Air Transport Liberalization A Critical Assessment - Finger2017

Air Transport Liberalization
Matthias Finger and Kenneth Button - 9781786431851

Downloaded from Elgar Online at 01/13/2018 08:55:07AM
via University of Durham
M4371-FINGER_9781786431851_t.indd 1 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 2 14/11/2017 11:14

Air Transport
Liberalization
A Critical Assessment
Edited by
Matthias Finger
Professor, Management of Network Industries, Ecole
Polytechnique Fédérale Lausanne, Switzerland
Kenneth Button
University Professor, Schar School of Policy and Government,
George Mason University, USA
Cheltenham, UK • Northampton, MA, USA

M4371-FINGER_9781786431851_t.indd 3 14/11/2017 11:14

© Matthias Finger and Kenneth Button 2017
All rights reserved. No part of this publication may be reproduced, stored

in a retrieval system or transmitted in any form or by any means, electronic,
mechanical or photocopying, recording, or otherwise without the prior
permission of the publisher.
Published by
Edward Elgar Publishing Limited
The Lypiatts
15 Lansdown Road
Cheltenham
Glos GL50 2JA
UK
Edward Elgar Publishing, Inc.

William Pratt House
9 Dewey Court
Northampton
Massachusetts 01060
USA
A catalogue record for this book

is available from the British Library
Library of Congress Control Number: 2017947105
This book is available electronically in the

Economics subject collection
DOI 10.4337/9781786431868
ISBN 978 1 78643 185 1 (cased)

ISBN 978 1 78643 186 8 (eBook)
Typeset by Servis Filmsetting Ltd, Stockport, Cheshire

M4371-FINGER_9781786431851_t.indd 4 14/11/2017 11:14

Contents
List of contributors vii
1 Introduction 1
Matthias Finger and Kenneth Button
PART I COUNTRY AND REGIONAL REPORTS
2 Airline liberalization in the US 7

James Peoples
3 Airline deregulation in Canada and the sustainability of
competition27
David Gillen and William G. Morrison
4 Australia – a reluctant liberalizer 51
Peter Forsyth
5 Air transport liberalization: the case of Ireland 69
Sean Barrett
6 The evolution of Indian civil aviation 92
Rajiv Nagpal and Haritha Saranga
7 Air transport development: a comparative analysis of China
and India 112
Yahua Zhang and Anming Zhang
8 European market: present and future 138
Volodymyr Bilotkach
9 Latin America and the Caribbean, thirty-plus years of
lukewarm liberalization of air transport markets 156
Henry L. Vega
10 Air transport in Africa 185
Gianmaria Martini and Davide Scotti

M4371-FINGER_9781786431851_t.indd 5 14/11/2017 11:14

vi Air transport liberalization
PART II TOPICAL ISSUES
11 Aviation safety in the age of liberalization 205

Clinton V. Oster Jr, John S. Strong and C. Kurt Zorn
12 Small community impacts of liberalization and the provision of
social air services 220
Aisling Reynolds-Feighan
13 Oligopolization of markets 244
Sveinn Vidar Gudmundsson
14 Domination of hub-and-spoke systems 266
Marc C. Gelhausen and Peter Berster
15 Market instability 284
Kenneth Button
PART III FUTURE CHALLENGES
16 Economic perspectives on aviation security 305

17 The need to evolve air traffic management: Europe as a
laboratory339
Matthias Finger, Marc Baumgartner and Engin Zeki
18 Canada and USA: a tale of two ANSPs 359
Rui Neiva
379
Index

M4371-FINGER_9781786431851_t.indd 6 14/11/2017 11:14

Contributors
Sean Barrett, Professor, Economics Department, Trinity College, Dublin,
Ireland
Marc Baumgartner, Air traffic controller, SESAR/EASA coordinator
IFATCA, Switzerland
Peter Berster, Dr, Institute of Air Transport and Airport Research,
German Aerospace Center (DLR), Germany
Volodymyr Bilotkach, Senior Lecturer in Economics, Newcastle University
Business School, United Kingdom
Kenneth Button, University Professor, Schar School of Policy and
Government, George Mason University, Arlington, USA
Matthias Finger, Professor, Management of Network Industries, Ecole
Polytechnique Fédérale Lausanne, Switzerland
Peter Forsyth, Adjunct Professor of Economics, Monash University and
Southern Cross University, Australia
Marc C. Gelhausen, Dr, Institute of Air Transport and Airport Research,
German Aerospace Center (DLR), Germany
David Gillen, YVR Professor of Transportation Policy, Sauder School of
Business, University of British Columbia, Vancouver, Canada
Sveinn Vidar Gudmundsson, Senior Professor Strategic Management,
Department of Strategy, Entrepreneurship and Innovation, Toulouse
Business School, France
Gianmaria Martini, Professor of Economics, University of Bergamo, Italy
William G. Morrison, Associate Professor, Lazaridis School of Business
and Economics, Wilfrid Laurier University, Waterloo, Canada
Rajiv Nagpal, Indian School of Business, Hyderabad, India
Rui Neiva, PhD, Policy Analyst, Eno Center for Transportation,
Washington, DC, USA
vii

M4371-FINGER_9781786431851_t.indd 7 14/11/2017 11:14

viii Air transport liberalization
Clinton V. Oster Jr, Professor Emeritus, School of Public and Environmental

Affairs, Indiana University, USA
James Peoples, Professor of Economics, Department of Economics,
University of Wisconsin-Milwaukee, USA
Aisling Reynolds-Feighan, Professor of Transport Economics, School of
Economics, University College Dublin, Ireland
Haritha Saranga, Dr, Production and Operations Management Area,
Indian Institute of Management Bangalore, India
Davide Scotti, Research Fellow, University of Bergamo, Italy
John S. Strong, CSX Professor of Finance, Raymond A. Mason School of
Business, College of William and Mary, USA
Henry L. Vega, Research Associate, Schar School of Policy and
Government, George Mason University, USA
Engin Zeki, PhD student, Ecole Polytechnique Fédérale Lausanne,
Switzerland and Eurocontrol, Brussels, Belgium
Anming Zhang, Professor, Sauder School of Business, University of
British Columbia, Vancouver, Canada and China Academy of Financial
Research, Shanghai Jiao Tong University, China
Yahua Zhang, Senior Lecturer, School of Commerce, University of
Southern Queensland, Toowoomba, Australia
C. Kurt Zorn, Professor, School of Public and Environmental Affairs,
Indiana University, USA

M4371-FINGER_9781786431851_t.indd 8 14/11/2017 11:14

1. Introduction
Matthias Finger and Kenneth Button
This book offers a critical and wide-ranging assessment of the air trans-
port liberalization process during the past 40 years.
This process is first located within the broader deregulation movement
of the late 1970s and 1980s, which, in addition to specific national events,
was partly the product of economic globalization at that time, resulting in
economic integration and competitive pressure on nation states and their
protected industries. All network industries were concerned, to varying
degrees. In addition, air transport was only one sector subject to reform;
other affected networks include the liberalization of national telecommuni-
cations and postal sectors, of energy, of surface transport and of the water
and sanitation sectors. The Anglo-Saxon countries typically deregulated the
operations of their network sectors early, with continental Europe typically
following suit, often driven by the European Commission, although Eastern
and Central Europe countries were driven by larger shifts in their political
structures. While such regulatory reform – or, to use the American jargon,
“deregulation” – triggered fierce ideological debates, it gradually became
clear that greater liberalization of both operations and infrastructure sectors
allowed the network industries to better adapt to the efficiency and com-
petitiveness requirements of an increasingly integrated and global economy.
Of course, there were often good reasons why the different national
network infrastructures and their operators were originally protected and
regulated in order to, depending on their intrinsic natures, minimize the
distortions that can occur in excessively competitive or monopolist struc-
tures. This was particularly the case when networks were vital elements in
the economy, either in terms of conventional economic development or as
conduits for allowing greater social and political integration. In aviation in
the 1920s and 1930s, such reasons initially pertained to ensuring mail net-
works, to tying ex-colonies to their mother countries, to passenger services
to less populated areas and to social services more generally, to national
interests and corresponding control over flag-carriers, to the support of
the military, and of course to safety, along with broadly defined national
interests, as later enshrined in the 1944 Chicago Convention.

M4371-FINGER_9781786431851_t.indd 1 14/11/2017 11:14

2 Air transport liberalization
While national interests were initially of primary importance in large

local markets, such as that of the United States, globalization subsequently
played a major role in transforming aviation. However, the changes were
not only brought about by national and international expediency. As we
see in several of the contributions, new economic ideas about regulatory
capture and about forms of competition, including theories of competi-
tion-for-the-market, contestability, and competitive tendering, were also
driving forces, as were empirical studies showing the distortions that
excessive regulation was inflicting on economies and social welfare. But
aviation was not transformed by ideas alone. Part of the evolution of avia-
tion simply had to do with the very practical matter of economic growth
and the increasing demand for air travel and for air freight transportation,
especially more recently in the emerging countries of Asia and to a much
lesser extent of Latin America and Africa.
On the supply side, there have been important technological develop-
ments, especially when it comes to aircraft, such as the adoption of wide-
body jets, longer-range airplanes, standardization of equipment, and the
possibility of two-crew cockpits. Significant changes have also taken place
at the airport level, where new financing mechanisms have allowed for the
development of existing capacity and the emergence of new capacity, and
as new business models have been introduced with greater emphasis on
complementary revenue sources, including car parking and retail conces-
sions. Some technological and managerial developments have also been
introduced into air traffic control that have increased its capacity and
facilitated the growth in traffic that has accompanied airline liberalization
and the growing air transport market. These have included incremental
improvements in radar-controlled traffic management, better weather
forecasting, and greater coordination of air traffic management systems
and financial arrangements. Airlines eventually found new business models
and practices, such as low-cost operations, new yield management tech-
niques, frequent flyer programs, and alliances. A large portion of global
air traffic is now carried by three strategic alliances, affording considerable
benefits of economies of scale and scope.
However, despite technological, economic, commercial, financial and
ideological trends favoring national, regional and global air transport
growth, liberalization remains – with the exceptions of Europe and Central
America in the specific case of air traffic control – largely a national
matter. Consequently, there have been significant differences regarding the
state and especially the outcomes of the liberalizations in the various coun-
tries and regions. This book begins by discussing regulatory liberalization
and its effects in the world’s main aviation markets – including the United
States (Chapter 2), Canada (Chapter 3), Australia (Chapter 4), Ireland

M4371-FINGER_9781786431851_t.indd 2 14/11/2017 11:14

Introduction 3
(Chapter 5), India (Chapter 6), and China (Chapter 7) – and regions, such
as Europe (Chapter 8), Latin America (Chapter 9), and Africa (Chapter
10).
After considering the main national and regional developments in the
aviation sector, the second section of the book discusses the main fears
or objections that have been leveled against deregulation of air transport;
namely, safety (Chapter 11), social services (Chapter 12), oligopoliza-
tion (Chapter 13), the ensuing persistence of the hub-and-spoke system
(Chapter 14), and market instability (Chapter 15). While most of these
fears have turned out to be unfounded, air transport deregulation, and
especially its success, has actually led to new, originally unanticipated
challenges. In the third section, we discuss two emerging challenges in the
aviation sector: mounting concerns over security (Chapter 16), and the
challenges in air traffic control in Europe (Chapter 17) and other parts
of the world (Chapter 18). Air traffic control is the remaining element of
global aviation that is still heavily regulated with continuing diversity in
national approaches. As well as national institutional rigidities, in terms of
technology, the global network is still based upon radar systems awaiting
a shift to more flexible and efficient satellite-based air traffic management.
There are inevitably other challenges that we have not addressed, includ-
ing reducing the adverse environmental imprint of aviation; developing
institutions that permit the maximum exploitation of economies of scale,
scope and density without the economic distortions that can accompany
monopoly power; and ensuring that less developed parts of the globe are
not excluded from the benefits of adequate air transportation. However,
such challenges will have to be left for further research and publications.
Here, we have simply gathered the most relevant and highest quality
inputs on all the various dimensions of air transport liberalization.

M4371-FINGER_9781786431851_t.indd 3 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 4 14/11/2017 11:14

PART I
Country and regional reports

M4371-FINGER_9781786431851_t.indd 5 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 6 14/11/2017 11:14

2. Airline liberalization in the US
James Peoples
INTRODUCTION
Air transportation has become an essential mode of international travel for

passengers and for economic trade, especially during the 21st century. In
the US, the number of passengers flying US–international routes increased
appreciably, from 166 million in 2011 to 203 million in 2015.1 Tons of
freight hauled on US–international routes increased from 9 648 920 to
9 822 771 tons for the same period. The economic activity associated with
increasing shipment of freight by air transport is better appreciated when
noting that most air freight consists of high-value, low-volume goods such
as electronic devices. Technological gains such as the development of more
fuel-efficient aircraft and improvements in logistics systems contribute to
affordable flights for passengers and efficient transport of low-volume,
high-value cargo.2 However, customers’ ability to take full advantage of air
transport cost-savings depends in part on easy access to a reasonably large
number of competing carriers. Without route competition, lower costs
could serve as a source of rent for carriers, rather than allowing for the
charging of low fares to customers. The potential gains to customers and
to the economy serve as a key reason for liberalization of international air
transport. Indeed, past research provides evidence suggesting significant
consumer welfare gains associated with international regulatory reform
(Cristea et al., 2014 and Winston and Yan, 2015). These studies also reveal
spillover effects on routes with non-signatory countries such that passen-
gers flying to these countries also experienced welfare gains. Given this
evidence of global welfare gain, there is the possibility that liberalization
policy may be converging toward the limits of its effectiveness in a way that
future agreements may not facilitate the type of gains experienced by early
signatories, due to the beneficial spillover effects enjoyed by passengers on
routes that include services of non-signatories.
This chapter examines fare differentials for US–international routes
serving signatory and non-signatory countries, using a more recent sample
population compared to past research. Initially, this chapter presents a

M4371-FINGER_9781786431851_t.indd 7 14/11/2017 11:14

historical review of liberalization of US–international aviation regula-

tion. This review includes analysis of the extent to which US liberalization
policy has promoted competition along routes. That section is followed by
a review of past research empirically examining the effect of liberalization
policies on fares and passenger traffic. Analysis of past research reveals
the need for fare estimation using more recent information that includes
countries which had not signed Open Skies Agreements for the observa-
tion period used in that research. The succeeding section describes the
data sources and empirical approach used to estimate fare equations for
US–international routes. That section is followed by the presentation
of the results derived from estimating fare differential for routes serving
US–international city pairs covered by Open Skies Agreements and US–
international routes serving city pairs that are not covered by Open Skies
Agreements. Concluding remarks are presented in the final section.
EARLY DEVELOPMENT OF INTERNATIONAL

AVIATION ACCESSIBILITY
Initial regulation of international air service was initiated with the 1919
Paris Convention. The overriding objective of the convention was to
establish regulations that were universally applicable to international avia-
tion transport. In addition to states’ rights to regulate entry and flights,
safety regulation was also a key component of the Paris Convention. The
general principles of the convention recognized that each contracting state
was entitled to maintain sovereign rights over its airspace, and therefore
had the right to regulate entry and flights into and through its airspace.
Regulation following the 1919 Convention imposed increasingly stringent
regulations on US air transport services. For instance, foreign investment
in US air carriers was significantly restricted following passage of the Air
Commerce Act of 1926. This Act required US citizens to own at least 51
per cent of any air carrier registered in the United States, and at least two-
thirds of the board of directors were required to be US citizens; US citi-
zenship was also required of the president of the company to operate in the
US (Furlan, 2006; Hardaway, 2010; Westra, 2007). Succeeding legislation
further increased the US citizenship threshold for air carrier ownership, as
the 1938 Civil Aeronautic Act (CAA) also established the Civil Aeronautics
Board (CAB) as a regulatory body with authority over the setting of rates
and the restriction of air carrier market entry. This regulatory oversight
applied to carriers serving national and international routes departing and
arriving at US airports. In sum, early legislation of international aviation
operations in the US created a highly regulated business environment with

M4371-FINGER_9781786431851_t.indd 8 14/11/2017 11:14

Airline liberalization in the US 9
the stated goal of promoting economic growth and development, as well as

assisting US flag carriers in competing with foreign competition.
A more liberal approach toward regulation of US–international avia-
tion began to take shape after World War II. Technological advancements
providing high-capacity and long-range air transport used for military
purposes offered the potential for profitable international commercial
air transport. However, the possibility of the routes closing following the
war facilitated the need to provide commercial air carriers the freedom
from regulation of international air transport. To address the need for
more open skies, the US convened the 1944 Chicago international avia-
tion conference to establish certain ‘principles and arrangements in order
that international civil aviation may be developed in a safe and orderly
manner and that international air transport services may be established on
the basis of equality of opportunity and operated soundly and economi-
cally’ (Preamble to the 1944 Chicago Convention on International Civil
Aviation). The conference re-established sovereignty over airspace above
a country’s territory (Article 1). In addition, the conference negotiated
agreements that addressed the issue of freedom of foreign carriers using
sovereign member states’ airspace. Two freedoms that address political
conflict that could arise from flying over foreign territory are contained in
the International Air Services Transit Agreement. Freedom 1 provides the
right to fly over a foreign country without landing and Freedom 2 extends
the right to refuel or engage in aircraft maintenances in foreign countries
without embarking or disembarking passengers or cargo. In addition to
Freedoms 1 and 2, three additional freedoms (Freedoms 3, 4 and 5) were
negotiated in the International Air Services Transit Agreement that make
it more economically viable to fly planes from one country to another by
allowing carriers to pick up, fly and discharge passengers and cargo on
international routes. For instance, the third freedom of the air allows the
domicile plane to fly from its own country to another; the fourth freedom
of the air allows the domicile plane to fly from another country to its
own country; and the fifth freedom of air allows the domicile plane to fly
between two foreign countries on a flight originating or ending in its own
country.
While the Chicago Convention provided a format for negotiating air
service agreements, challenges remained over economic issues for interna-
tional commercial flights (Hannappel, 1980). Guidelines on the capacity
and frequency of flights as well as the setting of rates were not settled and
would require later clarity from future negotiations. Indeed, the desire of
American carriers for greater frequency of flights served as the impetus for
the landmark 1946 US–UK aviation negotiations known as the Bermuda
I agreement (Hannappel, 1980), called Bermuda I because the site of the

M4371-FINGER_9781786431851_t.indd 9 14/11/2017 11:14

negotiations between the US and UK occurred in Bermuda. The agree-

ment served as the template for future bilateral aviation agreement for the
succeeding three decades. It included rules on route entry, flight capacity/
frequency, rate setting and airport fees. The prototypical Bermuda agree-
ment generally permitted countries to exchange traffic rights on a quid
pro quo basis, and entry was limited in a way that generally each country
designated one of its flag carriers on a city-pair route (Dempsey, 2008).
Provisions for capacity and flight frequency prohibited host countries from
unilaterally limiting the volume of traffic and the frequency of service
(Button, 2009).3 Provisions for rate setting allowed airlines to initially set
fares, subject to approval after 30 days of notice and upon double govern-
ment approval (Hannappel, 1980). Provisions for the setting of airport fees
prohibited host airports from charging fees to foreign carriers above fees
charged to domestic airlines.
MOVEMENT TOWARD LIBERALIZATION
After 30 years, the UK and US terminated the Bermuda I agreement, in

part due to UK dissatisfaction with the development of US traffic shares
exceeding that of UK carriers. The new agreement, called Bermuda II,
came into force in 1978 and imposed more restrictive provisions than its
predecessor. In contrast to the less liberal policy shift among the US–UK
aviation operations, US domestic policy enacted significantly more liberal
policies for domestic flights by US carriers. Motivated in part by evidence
showing that regulation of domestic air transport limited competition,
inflated prices and institutionalized inefficiency,4 the US Congress passed
the 1978 Airline Deregulation Act (ADA). This Act heightened the role
of the free market as a mechanism for setting fares and creating a busi-
ness environment allowing carriers greater freedom to compete on high-
demand routes. Complete deregulation of rate setting and market entry
occurred by 1983 (Shepherd and Brock, 2016). The effectiveness of this
policy change as a facilitator of efficiency enhancement is revealed by
research reporting an average increase in domestic US carriers’ load factor
of 52 per cent and a 15 per cent reduction in real cost per revenue ton mile
immediately following deregulation (Winston, 1998). The same research
also revealed consumer welfare gains associated with deregulation, as
average domestic US air fares reportedly fell 33 per cent and service fre-
quency improved significantly during the decade following passage of the
ADA.
While the ADA only applied to domestic air transport service, its
passage preceded a shift toward more liberal open-market agreements.

M4371-FINGER_9781786431851_t.indd 10 14/11/2017 11:14

Bilateral agreements ratified during the 13 years succeeding the enactment

of the ADA are characterized by liberalization of foreign carriers’ access to
US markets, enhanced opportunity for US carriers to increase the number
of alliance partners and increased flight capacity (Button, 2009). This
agreement contributed to enhanced competition from a foreign carrier, as
highlighted by Button (2009) and Pitfield (2009). They observed that nego-
tiated provisions provided the Dutch carrier KLM the flexibility to meet
market demand anywhere in the US, as well as exempting it from anti-trust
restrictions with its Northwest Airlines alliance. This agreement is viewed
as the template for Open Skies Agreements (OSAs).
Following this initial 1992 agreement, a growing number of bilateral
OSAs between the US and other countries contributed to continued lib-
eralization of US–international aviation. These US–international OSAs
include regional agreements with the EU and Pacific Rim. The legal
framework within which the US and its bilateral partners now operate
is captured in a number of major provisions of the comprehensive
agreement as outlined by Dempsey (2008).5 The prototypical open skies
agreement allows member countries broader entry into cooperative mar-
keting arrangements for code-sharing, franchising and leasing. Gains from
these types of arrangements, especially code-sharing, can arise from the
enhancement of a carrier’s presence on the new route served by the operat-
ing carrier, as well as from expansion of flight options for the marketing
carrier without the cost of additional investment, and gains can also arise
from the opportunity to dominate the code-sharing route (McMullen
and Yu, 2012). The standard OSA also allowed unrestricted route and
traffic rights such that carriers can change gauge, engage in the practice
of co-terminalization and carry fifth-freedom traffic. An example of these
provisions is found in the 2007 US–EU OSA. Following ratification of this
agreement, EU carriers are now recognized as ‘community air carriers’ by
the US, allowing them to schedule flights between any EU member state
and the US without touching the home country and without any price or
capacity restrictions (European Commission, 2008). For example, a French
Air France flight can go from Barcelona to the US without having to pass
through France. Compared to open-market agreements, OSAs are also
characterized by even greater ease of entry. For instance, a provision of the
1995 US–Canada aviation agreement provided Canadian airlines the right
to serve any city in the US. Excluding Montreal, Toronto and Vancouver,
US airlines also gained unlimited access to Canadian cities. Greater access
to the excluded cities was granted three years following the ratification of
the agreement. However, this Canadian agreement did not provide new
fifth freedom rights, and only provides limited co-terminalization of cargo
services.

M4371-FINGER_9781786431851_t.indd 11 14/11/2017 11:14

Easing of restrictions negotiated in OSAs presents carriers of member

states with the potential to enhance operating efficiency, provides greater
scheduling flexibility for passengers and contributes to lower fares by pro-
moting international competition. Potential efficiency gains arise in part
from provisions such as the right to carry fifth-freedom traffic that allow
carriers to fly less circuitous routes. Greater scheduling flexibility is a pas-
senger benefit derived from provisions allowing unlimited flight frequency
along routes. Promotion of international competition arises from provi-
sions liberalizing code sharing, franchising, leasing, pricing, and routes
between the US and EU (Button, 2009; Pitfield, 2009). However, sig-
nificant consumer welfare gains derived from the ratification of OSAs are
not certain, a priori. Alliances between the US and foreign carriers could
suppress competition if alliance members cooperate rather than compete
as independent rivals (Shepherd and Brock, 2016). Past studies show that
the three largest US–international alliances control close to 90 per cent
of commercial traffic between Europe and North America (Shepherd
and Brock, 2016). Oneworld, Sky Team and Star are the three major
US–international alliances such that American Airlines is the US partner
for Oneworld, Delta Air Lines is the partner for Sky Team and United
Airlines is the partner for Star. Combined, they serve 524 countries daily,
and depart on average 15 672 times daily (Shepherd and Brock, 2016). In
addition, these air carrier alliances’ market share accounts for 74 and 81
per cent of seat-miles flown at London Heathrow and Paris Charles de
Gaulle, respectively. Seat-miles flown by these three alliances account for
87, 69 and 90 per cent of the market at Frankfurt, Hong Kong and Tokyo
International, respectively.
PAST FINDINGS ON WELFARE GAINS ASSOCIATED

WITH FLIGHTS COVERED BY OSAs
Even though high market concentration associated with US–international

aviation alliances could promote non-competitive pricing behaviour, past
research suggests non-trivial gains in flight traffic and enhanced con-
sumer welfare. Findings by the US Department of Transportation (2000)
reported that average fares fell 20.1 per cent on transatlantic flights covered
by Open Skies Agreements for the 1996–99 observation sample. Fare find-
ings from this report also show a 10.3 per cent fare reduction for flights not
covered by Open Skies Agreements. This report explains that alliances are
able to offer improved services to these countries from their hubs in Open
Skies countries. Department of Transportation (DOT) findings on traffic
patterns show the total number of passengers serviced on US–European

M4371-FINGER_9781786431851_t.indd 12 14/11/2017 11:14

routes increased by 134 per cent for alliance members for the 1992–99
sample period, compared to a 50 per cent increase for other carriers.
While mean fare and traffic findings reported by the US DOT support
the hypothesis that Open Skies Agreements create a business environment
that promotes relatively low fares and higher flight capacity, these find-
ings may be misleading, since they do not control for other determinants
such as quality of service. Nonetheless, findings of enhanced consumer
welfare and flight capacity are reported when adjusting for differences
in fare determinants such as flight distance and number of competi-
tors in the market. Using US DOT Databank 1b (D1B1) data and T100
International Segment data for the 1993 to 2008 sample observation,
Cristea et al. (2014) reported that passengers on US–international flights
covered by OSAs pay lower fares and have a choice of more direct flights.
These findings are consistent with fare and flight frequency findings
by Winston and Yan (2015). Using international airline data collected
by the International Air Transport Association (IATA) for the 2005–09
observation sample, they reported almost a 15 per cent reduction in fares
associated with operating flights on routes covered by OSAs. Winston and
Yan used the findings to simulate welfare gains that could arise if the US
were able to negotiate OSAs with other countries experiencing a signifi-
cant amount of international traffic. Their simulation results suggest that
negotiating OSAs with remaining passengers would generate a gain of $4
billion for passengers.
A significant number of countries have signed bilateral OSAs with the
US following the 2009 observation period used in past research (Winston
and Yan, 2015). Notable among these signatories are countries with large
volumes of passenger traffic with the US, such as Japan, Colombia and
Brazil. Pricing pressure on routes without OSA coverage is potentially
high, assuming OSAs facilitated low fares charged on routes serving these
new signatory countries. Indeed, over 109 countries had signed OSAs with
the US by 2012, depicting a large global network of international destina-
tions with the ability to charge low fares and the capacity to serve a non-
trivial number of passengers. Maintaining high fares on routes that are not
covered by OSAs has the potential to suppress passenger demand in those
markets. In addition, as the US DOT observes, carriers can take advantage
of their hubs in Open Skies countries to provide lower-cost services to non-
signatory countries. Hence, estimation of fares along US–international
routes using more current data is warranted to examine the potential gains
from negotiating OSAs with remaining non-signatories.

M4371-FINGER_9781786431851_t.indd 13 14/11/2017 11:14

DATA AND EMPIRICAL APPROACH
Data
Data used to examine fares on Open Skies routes relative to fares on

other US–international routes are taken from the fourth quarter 2012
international itineraries of the US Department of Transportation (DOT)
DB1B Airline Origin and Destination Survey.6 The DB1B dataset is a 10
per cent sample of tickets for air travel within, to, or from the US and
includes information on airfare paid, origin and destination airports, the
US carrier serving the route under observation, and the round-trip status
for US carriers.7 These fares are matched with the characteristics of those
flights using the US DOT’s T100 Segment and International Market Data.
Matching is done by carrier, origin and destination. The T100 Segment
Dataset provides information on the number of passengers travelling on
each international and domestic flight, the type of aircraft they are travel-
ling in (including seat configuration) and the number of flights made for a
particular airline between a specific origin and destination location during
a given time period. The T100 International Market Data provides infor-
mation on the number of passengers travelling on a specific airline from
a particular origin to a particular destination during a given time period,
regardless of the number of stops in between the origin and destination.
Additional information on slot-controlled airports as reported by the
International Air Transport Association (IATA) is also included in the
analysis. A sample of 725, 205 and 851 one-segment international routes
for US carriers serving Open Skies in-force routes, provisional Open Skies
routes and routes without in-force or provisional agreements are respec-
tively obtained when using these data sets.
While these data sources offer the benefit of fare comparisons by Open
Skies routes, a shortcoming associated with the use of this information is
that it is limited to US carriers. The lack of information on non-US carriers
can contribute to underestimating actual route competition. Recognition
of this potential problem is considered when analysing fare results.
Descriptive statistics derived from using the merged data set are pre-
sented in Table 2.1. These statistics are mean values computed at the
route level to allow for merging fare information at the passenger level
with route-level information from the T100 data files. In all of these
estimations, following previous researchers (Bitzan and Chi, 2006), fares
that are outliers are eliminated and fares are averaged by route and
carrier. This is done to eliminate fare variation where no variation in
independent variables exists. For example, there may be 200 fares from
Boston to Toronto on Pinnacle Airlines in the fourth quarter of 2012;

M4371-FINGER_9781786431851_t.indd 14 14/11/2017 11:14

for this chapter’s estimation, each of these fares would have the same
distance, the same load factor, the same equipment size, the same flight
frequency, and so on. By averaging the fares, there is one observation
for each carrier/route combination. Findings are partitioned into three
categories: (1) O–D pairs that are covered by Open Skies Agreements;
(2) O–D pairs that are only provisionally approved by Open Skies
Agreements; and (3) O–D pairs that are not covered by a ratified Open
Skies Agreement or a provisionally accepted agreement. These mean
findings suggest flights on routes covered by Open Skies Agreements
actually charge passengers higher fares per mile. Fares along these routes
have a mean value of 39 cents per mile compared to 24 and 34.5 cents per
mile for routes covered by provisional agreements or without any agree-
ment, respectively. Analysis of other information on US-international
flights suggests that differences in route and carrier characteristics are
likely to contribute to an inflated mean fare for Open Skies routes. For
instance, compared to US carriers serving routes with provisional Open
Skies Agreements, US carriers serving international routes covered by
Open Skies Agreements fly shorter distances with lower load factors,
greater number of seats per aircraft and greater frequency of service.
Standard airline economic theory suggests shorter one-segment flight
distances can contribute to lower fares, since airline costs are a function
of the number of take-offs and landings, which vary less than propor-
tionally with distance flown.8 Average load factor, or the average number
of passengers per seat, is expected to have a negative influence on fares,
as many airline costs vary less than proportionally with the number of
passengers. For example, flight-crew costs, maintenance costs, fuel costs,
gate fees and terminal rental do not vary proportionally with passengers.
Similarly, as shown by Graham et al. (1983), there are cost savings asso-
ciated with operating larger aircraft. Thus, we expect that average plane
size will have a negative effect on fares, a priori. As reported in Bitzan et
al. (2014), it is important to note that each of the variables hypothesized
to influence the costs of a particular movement may also potentially
influence demand. For example, higher load factors may mean less
comfort due to crowding, decreasing demand. As another example, as
Borenstein (1989) suggested, larger aircraft are often more comfortable
and perceived to be safer than smaller aircraft. This may increase the
demand for travel on larger aircraft. However, previous researchers have
shown that the cost effects of these variables dominate any demand-side
effects that might occur (e.g. Bitzan and Chi, 2006 or Borenstein, 1989).
The final cost variable depicts frequency of service and is likely to influ-
ence fares through its impacts on demand. Higher flight frequency is
likely to mean more convenient schedules for passengers, increasing the

M4371-FINGER_9781786431851_t.indd 15 14/11/2017 11:14

Table 2.1 Descriptive statistics on US carrier characteristics and route

characteristics (2012, Q4)
Variable Mean of Characteristics for US–International

O–D Routes
With In-Force With Provisional Without

Open Skies Open Skies Open Skies
Agreement Agreement Agreement
Fare per mile ($) 0.396 0.240 0.345
Distance (miles) 2192 3793 1778
Load factor 0.765 0.804 0.763
Frequency of service 121 95 90
Seat per aircraft 148 214 147
Herfindahl 0.70 0.76 0.79
Index of Route Level Competition
Number of carriers 2.42 1.98 1.92
serving the route
Slot control 0.48 0.87 0.23
Potential carriers 50.08 54.88 49.09
US–Mexico routes (%) 0 0 46.53
Observations 725 205 851
Sources: US Department of Transportation DB1B Airline Origin and Destination Survey;

US Department of Transportation T100 segment data set.
demand for those flights. This suggests an expected positive influence of

flight frequency on fares.
Comparison of columns (1) and (3) in Table 2.1 shows less of a cost dis-
advantage associated with distance flown, load factor and size of aircraft
when comparing route and carrier characteristics for Open Skies routes and
for those not covered by an agreement. These similarities probably contrib-
ute to the smaller fare-per-mile differential for these two US–international
O–D route groups. Nonetheless, US carriers serving Open Skies routes do
provide appreciably more flights per route for the quarter compared to
carriers serving routes not covered by such agreements. An additional O–D
route characteristic likely contributing to lower fares on routes that are not
covered by Open Skies Agreements is the high percentage of US–Mexico
flights. Information presented in Table 2.1 shows that over 46 per cent of
these routes serve locations in Mexico. As revealed in Table 2.2, only 41.4
per cent of the US–Mexico O–D pairs are served solely by legacy carriers.
The remaining routes present passengers with the choice of flying low-cost
carriers or regional/charter carriers who advertise low fares.

M4371-FINGER_9781786431851_t.indd 16 14/11/2017 11:14

Table 2.2 US carrier options for passengers on US–Mexico Origin–

Destination (O–D) Pairs (2012, Q4)
Carrier Option Number of O–D pairs served by

One Carrier Multiple Carriers
Low-Cost Carrier (LCC)a 32
Legacy Carrierb 118
Other (Regional/Charter)c 71
Multiple LCC Carriers –
Multiple Legacy Carriers 1
Multiple Other Carriers 1
Combination of LCC and Legacy 44
Combination of LCC and Other –
Combination of Legacy and Other 18
Combination of All Three Carrier Types –
Notes:
a Low-cost US carriers serving US–Mexico routes are JetBlue Airways, Airtran Airways,
Spirit Airlines, Frontier Airlines, Sun Country Airlines and Virgin America Airlines.
b Legacy US carriers serving US–Mexico routes are United Airlines, American Airlines,
Alaska Airlines, US Airways and Delta Air Lines.
c Other (regional/charter) carriers serving US–Mexico routes are American Eagle, Mesa
Airlines, ExpressJet Airlines, SkyWest Airlines and Horizon Air.
Source: US Department of Transportation Data Bank 1B (DB1B) Origin and Destination

Survey files.
While some characteristics of carriers serving Open Skies O–D pairs are
associated with high fares, evidence in Table 2.1 also indicates character-
istics associated with low fares. These routes face relatively greater actual
competition, as the Herfindahl index is lower than that for routes without
an in-force Open Skies agreement. The promotion of market competition
is consistent with the objective of OSAs and, as stated earlier, mean values
actually understate the true level of competition, since the data source used
in this chapter does not include non-US carriers serving US–international
routes. The mean number of slot-controlled airports at the originating or
destination airport is smaller for Open Skies in-force routes. Because take-
off and landing slots are limited at the slot-controlled airports, the ability
of an existing or new carrier to meet the demand for more flights is limited.
In this way, the slot control can serve as an inhibitor to competition.
The variation in characteristics across the three groups of US–
international routes highlights the need to employ multivariable estima-
tion approaches to provide greater insight on the relative fares of flights

M4371-FINGER_9781786431851_t.indd 17 14/11/2017 11:14

serving Open Skies in-force routes, Open Skies provisional routes and
routes without such agreements.
Empirical Approach
Equation 2.1 is estimated to analyse fare differentials for flights serving

Open Skies in-force routes compared to flights serving Open Skies provi-
sional routes, flights serving Open Skies in-force routes compared to flights
serving routes without Open Skies Agreements, and to flights serving
Open Skies provisional routes compared to flights serving routes without
Open Skies Agreements.
Ln FPMij5β0 1 β1 Ln Distanceij 1 β2 Ln Loadij 1 β3 Ln Equipij

1 β4 Ln Freqij
1 β5Herfij 1 β6 PCarrij 1 β7 Slotij 1 β8 Mexicoij
1 Open Skiesij εij (2.1)
where FPMij is the mean airfare per passenger-mile for the carrier between
origin i and destination j. This variable is often referred to in other studies
as yield. The variable Distanceij is the mean one-way distance between
origin i and destination j, the variable Load is the mean load factor for the
carrier between origin i and destination j, which is computed by taking the
ratio of passenger-miles/seat-miles weighted by the number of departures
performed, and the variable Equipij is the mean aircraft size for the carrier
between origin i and destination j, which is computed by weighting the
number of seats per aircraft by the number of departures performed. These
three dependent variables are included to account for differences in the cost
that carriers incur on different routes. It is predicted that factors contribut-
ing to higher unit cost would be passed on to customers in the form of higher
fares. Hence, negative values are expected for the coefficients on these three
cost variables. The variable Mexicoij is also included for sample populations
comparing fares on Open Skies routes to fares on flights without Open Skies
Agreements to account for the cost differences, since a substantial share
of US–international flights that depart or arrive in airports in Mexico are
served by low-cost carriers and low-fare regional and charter carriers.
In addition to the inclusion of variables influencing operating costs or
identifying countries with a large prevalence of low-cost carriers serving
US–international routes, this specification also includes factors influenc-
ing the strength of air passenger demand on a particular route and factors
influencing the effectiveness of competition and potential competition.
For instance, the variable Freqij, which depicts the mean number of

M4371-FINGER_9781786431851_t.indd 18 14/11/2017 11:14

flights per quarter for the carrier between origin i and destination j, is an
indicator of passenger demand. The variables Herfij, PCarrij and Slotij
account for route-level competition and potential competition. Herfi is
the Herfindahl–Hirschman Index for flights between origin i and desti-
nation j, and is computed by taking the summation of carrier shares of
passengers squared for US carriers. This variable is normalized to range
from zero to one; thus, a value of one is indicative of a monopolistic route
and a value of zero is indicative of a perfectly competitive route. While
standard microeconomics suggests lower prices in competitive markets,
past estimation of US–international fares provides evidence supporting
the theory of imperfect contestability, such that the presence of potential
competitors at the originating or destination airport contributes to lower
fares (Bitzan et al., 2014). The specification of Equation 2.1 accounts for
the possible fare effect of potential competition by including the variable
PCarri,j. This variable depicts the number of carriers serving the origin
airport and/or the destination airport, but not serving the route. The
remaining variable Slotij that accounts for competition along O–D routes
is a dummy variable that indicates whether landing and departure slots
are airport-controlled at origin i or destination j. Since slot control is
viewed as an inhibitor of competition, it is expected that its coefficient
estimate will be negative.
The remaining fare determinant, specified as Open Skiesij in Equation
2.1, is included to examine whether Open Skies Agreements facilitate car-
riers charging customers low relative fares, all else equal. The Open Skies
variable is a dummy equalling one if the route includes a non-US destina-
tion or originating location that is covered by an in-force or provisional US
Open Skies agreement. The estimated coefficient on this dummy variable
depicts the log-fare differential for flights on Open Skies Agreement routes
compared to flights without Open Skies Agreements. This coefficient is
converted into a percentage differential by using the formula (expb − 1)
× 100; however, interpreting findings derived from making this conver-
sion might be problematic if heteroscedasticity is associated with the log-
transformed specification (Ai and Norton, 2000; Baser, 2007). Distorted
estimated fare differentials arise due to the introduction of enhanced bias
that is associated with the anti-log transformation when heteroscedastic-
ity is present. A Breusch–Pagan test for the log-fare estimate for each of
the three population samples rejects the hypothesis of homoscedasticity
as the chi-squared value varies from a low of 56.27 to a high of 181.66,
which is statistically significant at the 0.01 level. Given the presence of
heteroscedasticity, a gamma-based generalized linear method (GLM) is
used to estimate the log-wage equation to compute consistent estimates
(Manning and Mullahy, 2001).9 Estimation bias could still persist even

M4371-FINGER_9781786431851_t.indd 19 14/11/2017 11:14

after correcting for heteroscedasticity if market concentration is endog-

enous. Standard microeconomic theory suggests endogeneity is possible if
dominant carriers practise limit or predatory pricing. This chapter follows
past research and uses the lag of the Herfindahl index, load factor and
flight frequency as instruments to correct for possible endogeneity. A test
to examine the validity of these instruments supports their use for address-
ing potential market concentration endogeneity. For instance, the test for
weak instruments suggests that the minimum eigenvalue statistic ranges
from 1089.31 to 1756.95, which exceeds the critical values for relative bias
at the 1 per cent level. The Durbin and Wu–Hausman test of the relevancy
of market concentration endogeneity, however, generates F-scores ranging
from 1.203 to 0.021, which lacks statistical significance at the 1 per cent
level. This lack of statistical significance at standard levels supports not
rejecting the null hypothesis that market concentration is exogenous and
validates limiting bias corrections to log-transformation heteroscedasticity
by using the GLM procedure.
INTERNATIONAL FARE FINDINGS
Fare findings derived when estimating Equation 2.1 for three sample
populations is presented in columns 1–3 of Table 2.3. Column 1 contains
results when estimating Equation 2.1 for the sample population of routes
serving in-force Open Skies routes and those serving routes without Open
Skies Agreements. Column 2 uses the sample population of routes covered
by in-force or provisional Open Skies routes, and column 3 uses the sample
population of routes covered by provisional Open Skies Agreements and
routes without an Open Skies Agreement.
Findings on the control variables are consistent with standard trans-
portation economic theory. Lower fares are associated with the vari-
ables accounting for cost variation by routes. For instance, the estimated
coefficients on route distance, load factors and seats per aircraft are
negative and statistically significant for all but the sample population
for routes covered by provisional Open Skies and routes without Open
Skies Agreements (column 3). Even for that sample, only the estimated
coefficient on the load factor parameter lacks significance. The estimated
coefficient on the Mexico destination–originating route dummy is also
negative and is consistent with the mean findings, indicating a large share
of US routes arriving to or departing from Mexican airports are served
by low-fare airlines. The remaining parameter estimates on the variables
accounting for factors influencing the strength of air passenger demand
on a particular route, and factors influencing the effectiveness of com-

M4371-FINGER_9781786431851_t.indd 20 14/11/2017 11:14

Table 2.3 Estimation of airfares for one-segment international routes

(dependent variable: log of fare per mile (2012, Q4))a
Open Skies– No Open Skies– Provisional– No

Open Skies Provisional Open Skies
Sample Sample Sample
Intercept 2.31*** 2.273*** 1.78***
(15.17) (10.20) −(7.79)
In-force Open Skies dummy −0.057** −0.009 –
(−2.10) (−0.22) –
Provisional Open Skies dummy – – −0.047
– – (−1.18)
Distance (logs) −0.24*** −0.257*** −0.255***
(−10.50) (−6.98) (−9.22)
Load factor (logs) −0.028 −0.235** −0.039
(−0.43) (−2.27) (−0.50)
Seat per aircraft (logs) −0.448*** −0.411*** −0.353***
(−13.19) (−7.68) (−7.95)
Frequency of service (logs) 0.1057*** 0.094*** 0.124***
(11.10) (5.78) (11.31)
Herfindahl index 0.136** 0.022 0.299***
(3.02) (0.36) (5.33)
Number of potential carriers −0.0017*** −0.0012 −0.0022***
(−2.75) (−1.36) (−2.86)
Slot control dummy 0.117*** 0.097*** 0.055
(4.32) (2.65) (1.45)
US–Mexico route dummy −0.17023*** – −0.156***
(−5.53) – (−5.02)
Number of Observations 1576 930 1056
Log-Likelihood −873.51 −588.52 −539.022
Notes:
a T-statistic presented in parentheses.
*** Statistically significant at the 1 per cent level.

** Statistically significant at the 5 per cent level.
Sources: US Department of Transportation DB1B Airline Origin and Destination Survey,

and the US Department of Transportation T100 segment data set.
petition and potential competition reveal the predicted signs. Findings

depicting the fare effects of passenger demand suggest fares are higher
on routes providing frequent transport service for this quarter. Findings
depicting route-level competition reveal a positive and statistically sig-
nificant coefficient estimate on the Herfindahl index parameter for all

M4371-FINGER_9781786431851_t.indd 21 14/11/2017 11:14

three population samples. This fare finding for actual competition does
not support the hypothesis of perfect contestability, but does support
the rationale for negotiating Open Skies Agreements that enhance direct
competition. The estimated coefficient on the airport slot control is also
positive but lacks statistical significance for the sample population of
routes with provisional Open Skies Agreements and routes without Open
Skies Agreements. Findings depicting potential competition at the route
level suggest fares are lower on routes with a larger number of carriers
that could compete for service on the observed route. This fare finding on
potential competition is consistent with past research showing support for
the theory of imperfect contestability on US–international flights (Bitzan
et al., 2014).
The estimated coefficient of the parameter on the key variable of
interest is the Open Skies dummy. Findings on this estimated coefficient
reported in column 1 of Table 2.3 suggest statistically significantly lower
fares of 5.5 per cent on routes covered by in-force Open Skies Agreements
compared to routes without Open Skies Agreement coverage. This finding
is consistent with past research on liberalization of US–international
airline operations (Winston and Yan, 2015, and Cristea et al., 2014). In
contrast, findings reported in column 2 reveal that airfares on routes with
provisional Open Skies Agreements closely resemble fares on routes with
in-force Open Skies agreements. However, fares are measurably lower on
routes covered by provisional Open Skies Agreements compared to routes
without Open Skies Agreements. Indeed, findings in column 3 reveal a 4.5
per cent provisional Open Skies discount, which is close to the discount
observed in column 1 for in-force Open Skies Agreements. However, the
provisional Open Skies fare results lack statistical significance, most likely
due to the relatively small sample of routes covered by this type of Open
Skies Agreement.
CONCLUDING REMARKS
Since 1992, 120 countries have negotiated Open Skies Agreements with
the US. A review of the history of US aviation service agreements shows
that over 70 years have elapsed from initial regulation of international
service to the development of the current state of international aviation
service agreements. On the one hand, economic theory suggests that the
easing of fare, entry and capacity restrictions derived from negotiating
these agreements should enhance consumer welfare and present pas-
sengers with the opportunity to enjoy greater frequency of air transport
services. On the other hand, international alliances formed following

M4371-FINGER_9781786431851_t.indd 22 14/11/2017 11:14

ratification of Open Skies Agreements could promote the development

of non-competitive markets and the practice of non-competitive pricing.
Findings from this study show Open Skies Agreements are associated
with lower fares and greater frequency of service. While such findings
aren’t novel, the analysis uses more current data than that used in pre-
vious research, and allows us to examine whether further welfare gains
are available even in a business environment presenting global competi-
tive pressures from a larger share of countries providing the benefits of
relatively open skies compared to the sample of countries used in past
research. The results from this study are interpreted as suggesting that
potential welfare gains are still available from the negotiation of OSAs
with remaining countries.
Even though this study does find gains from negotiating Open Skies
Agreements, significant barriers to competition still remain. Non-US car-
riers are still prohibited from engaging in cabotage in the United States, as
US law does not permit non-US carriers the right to transport passengers
or cargo originating at one point in the United States to another US point
(Westra, 2007). In addition, non-US flag airlines still face a 25 per cent
legislated cap on voting equity and a statutory ban on ‘actual control’
of a US air carrier by foreign nationals (Alford and Champley, 2007).
These remaining restrictions on foreign ownership, cabotage and routes
continue to create efficiency challenges for non-US carriers. Prohibiting
foreign companies from acquiring US airline companies makes it difficult
for foreign airlines to experience efficiency gains through economies of
scale, scope and density when providing service to the US market. Hsu
and Chang (2005) explained that removing provisions on foreign owner-
ship makes it easier for foreign carriers to acquire domestic air companies,
allowing newly allied carriers to spread fixed costs over more passengers.
Foreign acquisition promotes economies of scope by allowing newly allied
carriers to reconfigure their networks to connect more flights to hub air-
ports. Acquisition promotes economies of density by allowing newly allied
carriers to combine traffic to improve load factors. Remaining restrictions
on international air transportation also create impediments to competi-
tion by limiting the number of carriers that have the ability to serve
various routes and by limiting the scope of carrier operations (Bitzan et
al., 2014). Thus, removing these remaining restrictions has the potential of
promoting even greater fare competition from foreign carriers.

M4371-FINGER_9781786431851_t.indd 23 14/11/2017 11:14

NOTES
1. Source: US Department of Transportation. Available at https://www.transportation.gov/

policy/aviation-policy/us-international-air-passenger-and-freight-statistics-report.
2. Newer aircraft, like the Airbus A380 and Boeing 787, consume on average less than
3 litres per 100 passenger kilometres or more than 78 passenger statute miles per US
gallon. This is a fuel use which compares favourably to that of compact cars, although
aircraft travel much further distances, much faster. Source: http://aviationbenefits.org/
environmental-efficiency/technology/.
3. Airline capacity is the maximum number of passengers, cargo or number of aircraft
which a route can accommodate for the charged fare.
4. See Keeler (1972) for an analysis of airline market performance prior to regulatory
reform in this industry.
5. For a more detailed presentation of the provision of the 2007 comprehensive US–EU
Open Skies Act, see Button (2009), Pitfield (2009), Shetty (2011), Alford and Champley
(2007), Robyn et al. (2005), Westra (2007), Prokop (2014) and Fu and Oum (2014), among
others.
6. This chapter’s sample is limited to one-segment movements, as we are not able to identify
origins and destinations for multiple-segment movements. All airfares are for round-
trip movements and include taxes and fees, which vary by destination and originating
airports.
7. Although the type of ticket (for example, unrestricted first class) is identified in the
DB1B, the US Bureau of Transportation Statistics notes that different carriers may use
different standards to classify tickets. Therefore, they recommend that this classification
not be used for analysis.
8. Examples of costs associated with distance flown between landing and departure include
fuelling costs, boarding costs, luggage loading costs, security costs, landing fees and
maintenance costs.
9. The GLM procedure assuming error terms with Poisson, Gaussian, inverse Gaussian and
binomial distributions was also used to estimate Equation 2.1. The results using these
distributions mirror the results using the gamma distribution.
REFERENCES
Ai, C. and E. Norton (2000), ‘Standard Errors for the Retransformation Problem
with Heteroscedasticity’, Journal of Health Economics, 19(5), 697–718.
Alford, E. and R. Champley (2007), ‘The Impact of the 2007 US–EU Open
Skies Air Transport Agreements’, International Trade Administration Occasional
Paper, no. 07–001.
Baser, O. (2007), ‘Modeling Transformed Health Care Cost with Unknown
Heteroskedasticity’, Applied Economics Research Bulletin, 01, 1–6.
Bitzan, J. and J. Chi (2006), ‘Higher Airfares to Small and Medium Sized
Communities: Costly Service or Market Power?’, Journal of Transport Economics
and Policy, September, 40, 473–501.
Bitzan, J., A. Kones and J. Peoples (2014), ‘Airfares and Competition on
International Routes’, in J. Peoples (ed.), The Economics of International Airline
Transport, Bingley: Emerald Press.
Borenstein, S. (1989), ‘Hubs and High Fares: Dominance and Market Power in
the U.S. Airline Industry’, RAND Journal of Economics, Autumn, 20, pp. 344–65.
Button, K. (2009), ‘The Impact of US–EU “Opens Skies” Agreement on Airline

M4371-FINGER_9781786431851_t.indd 24 14/11/2017 11:14

Market Structures and Airline Networks’, Journal of Air Transport Management,

15, 59–71.
Cristea, A., D. Hummels and B. Roberson (2014), ‘Estimating the Gains from
Liberalizing Services Trade: The Case of Passenger Aviation’, University of
Oregon working paper.
Dempsey, P. (2008), ‘Air Traffic Rights’, available at: http://www.mcgill.ca/iasl/files/
iasl/ASPL633-Air-Traffic-Rights.pdf.
European Commission (2008), ‘The EU–US “Open Skies” Air Transport
Agreement: Q&A’, MEMO/08/185 28/03/2008.
Fu, X. and T.H. Oum (2014), ‘Air Transport Liberalization and its Effects on
Airline Competition and Traffic Growth: An Overview’, in J. Peoples (ed.), The
Economics of International Airline Transport, Bingley: Emerald Press.
Furlan, C. (2006), ‘Foreign Ownership and Control Restrictions in United States
Airlines: Barrier to Mergers and Restructurings’, University of Miami School
of Law working paper.
Graham, D., D. Kaplan and D. Sibley (1983), ‘Efficiency and Competition in the
Airine Industry’, Bell Journal of Economics, 14, 118–38.
Hannappel, P.P.C. (1980), ‘Bilateral Air Transport Agreements: 1912–1980’,
Maryland Journal of International Law, 5, 241–67.
Hardaway, R. (2010), ‘Of Cabbages and Cabotage: The Case for Opening Up
the U.S. Airline Industry to International Competition’, Transportation Law
Journal, 34(1).
Hsu, C.J. and Y.C. Chang (2005), ‘The Influence of Airline Ownership Rules
on Aviation Policies and Carriers’ Strategies’, Proceeding of the Eastern Asia
Society for Transportation Studies, 5, 557–69.
Keeler, T. (1972), ‘Airline Regulation and Market Performance’, The Bell Journal of
Economic and Management Science, 3, 399–424.
Manning, W. and J. Mullahy (2001), ‘Estimating Log Models: To Transform or Not
to Transform’, Journal of Health Economics, 20(4), 461–94.
McMullen, B. and Y. Du (2012), ‘Determinants of Successful Code-Sharing: A
Case Study of Continental and America West Airlines Alliances: A Discrete
Longitudinal Analysis’, in J. Peoples (ed.), Pricing Behavior and Non-Price
Characteristics in the Airline Industry, Bingley: Emerald Press.
Pitfield, D.E. (2009), ‘The Assessment of the EU–US Open Skies Agreement: The
Counterfactual and Other Difficulties’, Journal of Air Transport Management,
15, 308–14.
Prokop, D. (2014), ‘Government Regulation of International Air Transportation’,
in J. Peoples (ed.), The Economics of International Airline Transport, Bingley:
Emerald Press.
Robyn, D., J. Reitzes and B. Moselle (2005), ‘Beyond Open Skies: The Economic
Impact of a US–EU Open Aviation Area’, in D.S. Hamilton and J.P. Quinlan
(eds), Deep Integration: How Transatlantic Markets are Leading Globalization,
Washington, DC: Center for Transatlantic Relations and Centre European
Relations, pp. 50–73.
Shepherd, W. and J. Brock (2016), ‘Airlines’, in J. Brock (ed.), The Structure of
American Industry, 13th edn, Long Grove, IL: Waveland Press.
Shetty, K. (2011), ‘The Role of Chicago Convention and the ICAO in the
Development of International Air Law: A Critical Approach’, available at http://
www.studymode.com/essays/The-Role-Of-The-Chicago-Convention-622586.
html.

M4371-FINGER_9781786431851_t.indd 25 14/11/2017 11:14

US Department of Transportation (2000), ‘Transatlantic Deregulation: The

Alliance Network Effect’, International Aviation Developments, second Report,
available at https://www.transportation.gov/office-policy/aviation-policy/transat
lantic-deregulation-alliance-network-effect.
Westra, C. (2007), ‘US–EU “Open Skies” Agreements: A Dream of Liberalization
Deferred’, Boston College International and Comparative Law Review, 32,
161–76.
Winston, C. (1998), ‘US Industry Adjustment to Economic Deregulation’, Journal
of Economic Perspectives, 12, 89–110.
Winston, C. and J. Yan (2015), ‘Open Skies: Estimating Travelers’ Benefits from
Free Trade in Airline Services’, American Economic Journal: Economic Policy,
7, 370–414.

M4371-FINGER_9781786431851_t.indd 26 14/11/2017 11:14

3. Airline deregulation in Canada and
the sustainability of competition
INTRODUCTION
Canada, in 2016, has a domestic, transborder and international aviation

sector that is dominated by one airline alliance (Star) and by one airline
(Air Canada); it has one healthy ‘cost-oriented carrier’ (WestJet) and a
few charter/schedule carriers (Air Transat, for example) that serve leisure
destinations.1 Recently, an effective competitor entered the domestic and
transborder markets: Porter Airlines, which offers limited competition to
both Air Canada and WestJet on a small subset of domestic and transbor-
der routes in the eastern part of Canada.2 Transborder routes are domi-
nated by Air Canada and its Star Alliance partners. Regional airlines are
predominately contracted to or owned by either Air Canada or WestJet.
There has been limited entry post-deregulation, and the small carriers who
did enter failed for a range of reasons.
The Canadian air travel market and the amount of competition is sig-
nificantly influenced by several factors, including Canada’s geography – it
is 3000 miles from west to east coasts – a population of 35.2 million
people, the majority of whom live within 200 miles of the United States
(US) border, and the concentration of traffic, with Canada’s busiest four
airports handling 70 per cent of total passenger traffic in Canada.3
Current air transport policy and airline industry market structure is a
legacy of the government’s history of active participation in every part of
the air travel sector: building and managing airports, owning and manag-
ing airlines, setting economic regulations in domestic markets for entry,
routes, fares and imposition of geographic boundaries of service, defining
and negotiating air service bilaterals, providing ATC (Air Traffic Control)
infrastructure and services as well as setting, enforcing and investigating air
safety regulations. Air transport policy is a legacy of Air Canada, having
been government-owned and operated and considered to be the airline
that represented Canada in international markets. The policy was driven
by a view that transportation was a tool of government, and the airline
27

M4371-FINGER_9781786431851_t.indd 27 14/11/2017 11:14

was treated like a public utility. As a public utility, the airline provided a
public service and the policy was to preserve service and employment in a
financially sustainable way; there was no consideration of the consumer’s
interests. This narrow view of the public interest defined as ‘to preserve
service by a national airline in a financially sustainable way’ has driven
most – if not all – subsequent Canadian air policy decisions; Transport
Canada has chosen Air Canada to be ‘the winner’, and structures and
manages air policy accordingly.
EVOLUTION TO DEREGULATION
The shift from regulation to deregulation came in the early 1980s, with
both internal and external pressures. Domestic deregulation in the US in
1978 offered a window on the impact of such changes. Increased com-
petition, increased entry, and lower fares were appealing to consumers
and politicians alike. The leakage of large numbers of Canadian air pas-
sengers to nearby US points at that time (as it is now in 2016) was also a
stimulus. The Crown-carrier Air Canada (AC) was regulated under a new
Act passed in 1977; this Act placed the carrier on an equal basis before
the regulator, the Canadian Transport Commission – at least in principle.
Under the new Act, AC’s shares were held by the Minister of Transport on
behalf of the government. The carrier was ordered to have a more com-
mercial orientation and to be profit motivated. At the time, however, the
incentives, rewards or penalties for AC were not changed, thus pursuing
profit was not like the challenge at other carriers. There was little or no
downside risk, as the government would not have let AC fail. This is most
likely true today as well, even with privatization. To be fair, AC was still
charged explicitly, as well as implicitly in the eyes of Canadians, with pro-
viding service that under strict commercial rules would be unlikely to be
offered. The internal pressures came from the treatment of carriers other
than AC. The regional carriers were restricted under the Regional Air
Carrier programme, and Canadian Pacific (CP) was limited in a number
of ways from competing with AC including limitations on the capacity it
could offer on transcontinental and some overseas routes. These restric-
tions had disappeared by 1980.
The first vestiges of competition began with the government allowing
discount fares by the two transcontinental carriers, AC and CP. All carriers
introduced a series of ad hoc lower fares in various forms. These included
discounts to students, seniors and families travelling together, to name a
few. The discounts were not across the board and were not actually the
outcome of competition, but they did serve to respond in some degree to

M4371-FINGER_9781786431851_t.indd 28 14/11/2017 11:14

Airline deregulation in Canada 29
the leakage of passengers to US origin airports close to the border and to

make it clear that Canadians responded to lower fares; thus demand was
elastic. The move to deregulation in Canada was evolutionary, rather than
revolutionary as in the US. As a result of an evolutionary approach, which
was similar to what occurred in Europe in the 1990s, the incumbent car-
riers were able to entrench their positions. In the European Union (EU),
with the much larger market and the creation of a common market, a
large number of aggressive low cost carriers (LCCs) emerged; this did not
happen in Canada due both to Canada’s small air travel market and to
limitations on foreign ownership.
GOVERNMENT POLICY AND REGULATION
Currently, transportation policy in Canada, for modes under federal

government jurisdiction (air, rail and marine), is governed by the Canada
Transportation Act (CTA).4 These transport modes will also be governed
by some additional Acts; in aviation, this would include the Aeronautics
Act and Canadian Aviation Regulations, for example. The CTA sets out
the objectives of transportation policy and the means of achieving those
objectives. Canadian transportation policy states that the best way to
meet the needs of the Canadian economy and economic well-being of
Canadians is to have a transportation system that is competitive, economic
and efficient. The competitiveness of Canadian firms in Canadian, North
American and world markets depends on the efficiency of the transporta-
tion system. The CTA maintains the needs of users, and ensures that the
well-being of Canadians in rural and urban Canada is met, when transpor-
tation is provided in the most efficient way possible, namely, at the lowest
total cost.
The CTA is also clear in stating that competition and market forces,
both within and among the various modes of transportation, are the prime
agents in providing viable and effective transportation services. In essence,
the CTA specifies that transportation services should be supplied at rates
that cover the cost of providing transportation services. Therefore, both
the users of transportation services and the providers of transportation
services will have their respective needs met by a ‘competitive, economic
and efficient national transportation system’ (Canada Transportation
Act, Section 5). Despite this statement of policy and the statement of
the preferred means of achieving the objectives of the policy, Section 5 is
completely ignored when it comes to designing and implementing aviation
policy.

M4371-FINGER_9781786431851_t.indd 29 14/11/2017 11:14

GOVERNMENT POLICY TOWARD AIR CANADA
The creation of Air Canada preceded the imposition of economic regula-

tion of commercial aviation, which began in 1938 with the Transport Act.
Such regulation, according to C.D. Howe (1938), ‘should have the effect
of lowering rather than raising rates, by tending to eliminate wasteful and
destructive competition among the various forms of transportation’. The
view that competition is wasteful and that government policy-makers and
regulators can use their power to create a more orderly regime has been
endemic in Canada. In respect to the airline industry, this remained the
dominant view until the mid-1980s, undergoing a hiatus until the late
1990s and being re-established under the guise of Bill C-26, which con-
ferred significant powers on the Competition Bureau.
The Crown-owned carrier, for at least the first 40 years of its life, was
probably more important than economic regulation as a government
instrument in shaping commercial aviation in Canada. It was government
policy that Air Canada had a monopoly on all domestic transcontinental
routes between 1937 and 1959.
It became evident that CP’s proposed amalgamation of ten regional
and local carriers in 1942, if joined, would have created a transcontinental
system paralleling AC (earlier known as Trans Canada Airlines, the name
was changed to Air Canada in 1964). Hence, the federal government moved
in 1944 to order all surface carriers to divest all of their interest in air carri-
ers within one year of the end of the war in Europe. In the US, for example,
the federal government prevented aircraft manufacturers from owning air
carriers. For instance, Boeing had to divest United Airlines and Ford had
to sell off Eastern Airlines. This move was clearly aimed at CP, although
it was rescinded in 1946. In 1959 the government allowed CP to provide
a transcontinental service; this was limited to one flight per day each way
between Vancouver and Montreal in order to protect Air Canada’s finan-
cial position. Between 1959 and 1965, CP obtained an average of 12.7 per
cent of the transcontinental market. Capacity restraints were maintained
until 1979. However, capacity limits were gradually relaxed pursuant to
Ministerial statements in 1967 (two flights per day and 25 per cent of
capacity by 1970), 1974 and 1977 (capacity increased to 35 per cent of the
growth in 1978 and to 45 per cent in 1979). In 1979, all capacity constraints
on CP were removed.
Between 1937 and 1945 the government simply underwrote AC’s losses
– losses that occurred even after the effects of setting the mail contracts
at artificially high rates. Between 1946 and 1950 Air Canada’s losses
amounted to $11.5 million. Again, the government financed the losses,
but only after forcing the airline to draw down its $1.5 million in retained

M4371-FINGER_9781786431851_t.indd 30 14/11/2017 11:14

earnings. Deficits were much less frequent after 1950, but between 1937
and 1962 Air Canada received ten subsidies amounting to $23.9 million.5
It has been claimed that since 1964 the airline has paid dividends
amounting to $55 million. However, it should be noted that the govern-
ment forgave portions of the Crown airline’s debt. In 1977, for example,
the federal government converted a major portion of Air Canada’s debt
to equity, thus reducing its burden of interest payments. Air Canada states
that since 1978 it has repaid $120 million in government loans and that it
has paid interest on these loans totalling $148 million.
When Air Canada was privatized in 1989, there were two important
aspects to the privatization that affected the ability of others to compete
fairly. The first was limitations on ownership and the second was the way
that the privatization funds were distributed (discussed in detail below).
Both actions resulted in a strong equity position for AC and showed that
the competition between carriers was biased, as the private sector carriers
were faced with private sector market conditions. On the other hand, AC
has continued to receive favourable treatment by government as the gov-
ernment adjusted the requirements for AC to meet pension obligations,
and twice has threatened back-to-work legislation when groups within AC
threatened to go on strike.
NEW CANADIAN AIR POLICY: MAY 1984
The path to the new air transportation policy began in September 1983,
when the then Minister of Transport, Lloyd Axworthy, sent a letter to the
acting president of the Canadian Transport Commission (CTC) stating
that recent events relating to ‘discount fares highlighted the need for a
comprehensive discussion of the entire domestic air travel pricing policy’.
He activated a section of the National Transportation Act and asked
the Minister of Justice to appoint an advocate for the public interest. In
December 1983, Mr Axworthy formed an interdepartmental taskforce
on airline regulation. It was composed of a ‘steering group’ consisting of
deputy ministers and a working group of middle-level public servants.
In early 1984, the CTC began hearings on the regulation of domestic
airfares. Very quickly, a wide range of interveners pushed the task force
hearings toward the broader issue of deregulation. A group of airline
unions strongly protested against the idea of deregulation in Canada. Air
Canada, however, changed its position in midstream and ceased to be a
vociferous opponent of deregulation.
In May 1984 the government announced its ‘New Canadian Air Policy’.
It recommended some relaxation of regulation, ‘but not deregulation’. The

M4371-FINGER_9781786431851_t.indd 31 14/11/2017 11:14

highlights of the Minister’s proposals created the framework for eventual

deregulation, albeit in an evolutionary way that ensured incumbents were
entrenched. Canada was to be divided into two regions for the purposes
of airline regulation: north and south. All of the old regime rules were
to continue to apply to northern Canada. In southern Canada, airlines
were to have unlimited freedom to reduce prices. Price increases were to
be allowed up to the rate of inflation of an input price index excluding
labour. Additional price increases would have to be approved by the CTC.
There was to be unlimited entry into round-trip charter markets in south-
ern Canada. All mandatory booking and travel restrictions on discount
fares were removed. The Regional Air Carrier Policy was abandoned. Any
licensed carrier could apply to the CTC for any existing or new route.
Air Canada was to be restricted from engaging in anti-competitive
pricing and scheduling practices. The Crown carrier could meet the com-
petition of other carriers. Air Canada was not to receive funds from the
Government unless it met ‘accepted financial tests’. There was also an
increase in transborder competition. In mid-1984 a new agreement with
the United States was announced, under which American carriers were
granted automatic entry from any point in the US – other than major
gateways – to Mirabel airport. The US was later to name an airport to
enjoy the same privileges as Mirabel (San Jose was chosen). On transbor-
der routes of 400 miles between smaller centres in central Canada and 600
miles elsewhere, local and regional carriers would be granted automatic
entry and were encouraged to establish innovative pricing structures. This
was the prelude to the eventual Open Skies Agreement signed in 1995.
In 1984 the spirit and intent of liberalization was illustrated by the
significant reductions in the restrictions on discount fares. However, the
‘regulation’ advocates still managed to drag their feet. As carriers wanted
to develop market-specific fare strategies, the regulator insisted that any
discount to fares be applied system-wide. This resistance to reducing regu-
lation changed in 1984, with the election of a Conservative government
that accelerated the move to greater commercial freedom in the airline
industry, a move initiated by the previous Liberal government.
These changes represented a ‘shift from eroded protectionism to differ-
entiated liberation’; that is, a liberated south with a wide zone of pricing
freedom and an unchanged, protected, regulated north. But within the
south there was also differentiation: local carriers were to be free of entry
and fare restrictions in operating light aircraft between smaller cities lying
in a band approximately 500 miles across the US border.
The importance of the ‘New Canadian Air Policy’ was multifold. First,
it clearly signalled the intention of the federal government to move toward
deregulation, a move that the US had made seven years earlier. Second,

M4371-FINGER_9781786431851_t.indd 32 14/11/2017 11:14

under liberalization the carriers had more time and a somewhat more
sheltered environment (for example, entry to the industry was virtually
impossible, while entry to specific routes was easier but not free) in which
to get ready for deregulation. Third, the policy was followed, in only 14
months, by the federal Conservative government’s Freedom to Move policy
paper, which proposed outright deregulation throughout Canada. This
obviously reinforced the move toward deregulation and put more pressure
on all existing carriers to get their costs down, engage in consolidations to
achieve the benefits of a large network and reconfigure their route struc-
tures so as to survive under open competition.
It should be noted, however, that the May 1984 policy statement made
no reference to the future of Air Canada and whether it should be pri-
vatized. However, the Crown carrier was instructed to operate on sound
business principles in contemplation of profit, and to refrain from overly
aggressive competitive pricing and scheduling practices.
DEREGULATION IN 1988
On 4 November 1986, the government gave first reading to a significantly

revised National Transportation Act, incorporating provisions to deregu-
late the airlines in southern Canada while only slightly liberalizing regula-
tion as it applied to routes in the north, or to and from the north. The
Bill became law in 1988. The key elements of the legislation for the airline
industry included a new ‘fit, willing and able’ test requiring only proof
of insurance coverage, certified aircraft, licensed pilots and 75 per cent
Canadian ownership and control. There was to be no control over entry to
domestic routes by existing carriers operating in southern Canada.
In southern Canada, there was no longer any regulatory control over
route entry, fare levels or structure, frequency of service, product distri-
bution or contracts with large clients. In the case of exit, the carrier was
required to provide 120 days’ notice, but only if the carrier was abandoning
the route entirely and not simply removing a flight. Issues of complaint
were handled by a new regulatory agency designed to oversee the new
liberalized policy, the National Transportation Agency (later renamed
the Canadian Transportation Agency). Unlike the Canadian Transport
Commission, the Canadian Transportation Agency was charged with the
mandate of promoting both intra- and intermodal competition. Access
to transborder and international routes continued to be subject to bilat-
eral agreements and domestic political considerations regarding which
carrier(s) would represent Canada.
The implications of the new deregulated environment would have

M4371-FINGER_9781786431851_t.indd 33 14/11/2017 11:14

consequences for the structure of the industry in the longer term. Many
anticipated that the results observed in the United States after deregulation
in 1977 would be similar for Canada. Initially, some changes in industry
structure were seen, and there were some changes in the level of fares,
as well as the introduction of a greater variety of fares. In his study for
the Royal Commission on National Passenger Transportation, Morrison
(1993) said that deregulation in Canada had no effect on the average level
of fares, but had simply decreased long-haul fares and increased short-
haul fares. This is interesting, as the government – through the CTC – had
held the industry with such tight constraints by confining certain airlines
to certain geographic markets, that the potential efficiency gains from
deregulation were less than those in evidence in the US.
An important gap in the 1987 National Transportation Act was the
lack of recognition of the continued existence of a government-owned air
carrier. There was deregulation, but at least initially there was no privatiza-
tion. The importance of this gap is that while the policy intended that there
be competition among air carriers, the initial conditions of such competi-
tion had not been established. In particular, the effectiveness of a competi-
tive deregulation policy – when several small private-sector air carriers had
to compete with a still apparently privileged, well-branded government-
owned air carrier with over 50 per cent of the total market – was not dealt
with; nor was the control that Air Canada had at Canada’s major airport
in Toronto. In the late 1980s, for example, Air Canada enjoyed a larger
market presence than all other carriers combined. It served over 60 points,
whereas the other national carrier served 35 points and the smaller regional
carriers served far fewer. The most important aspect of Air Canada’s pres-
ence and dominance in the market was that it attracted a disproportionate
share of the high-yield business traffic.
PRIVATIZATION OF AIR CANADA
The next major policy change occurred in 1988, when Air Canada was pri-
vatized. Until this time, it had operated under the direction provided by the
Air Canada Act of 1977. This Act was ostensibly to place the airline on an
equal footing with other carriers, particularly in front of the regulator, the
CTC. Prior to 1977 Air Canada had listed only $5 million in paid equity
from its shareholder, the government of Canada. The airline operated with
significant amounts of government-guaranteed debt that was financed by
its then-owner Canadian National Railway. A significant amount of this
debt had been forgiven over the years, but with the new Air Canada Act
in 1977, $325 million in debt was converted to equity. As a result of this

M4371-FINGER_9781786431851_t.indd 34 14/11/2017 11:14

transfer Air Canada’s costs were reduced, as a major reduction in interest

expenses occurred at a time when interest rates were hovering near 20 per
cent.
Under the new Act, Air Canada was instructed to be profit-oriented and
to operate on a more commercial basis. Unfortunately, the Canadian gov-
ernment was unclear in the Act as to the social role that the airline would
play in the future, and the role that it had played in the past.
When Air Canada was sold, there were two important aspects to the
privatization. First, under the Air Canada Public Participation Act of
1987, no individual or coordinated group of individuals could own more
than 10 per cent of Air Canada’s shares. Furthermore, non-Canadians
collectively cannot own more than 25 per cent of the voting shares under
the conditions of the National Transportation Act. This policy led to the
failure of the Onyx bid for Air Canada and Canadian Airlines in 1999.
The second important feature was that the privatization which took place
in two tranches led to a substantial strengthening of Air Canada’s balance
sheet. This was not necessarily inappropriate or illegal, but it did go against
the spirit of the deregulation of the industry that took place in 1988. One
could argue that since the government received only a fraction of the
investment it had been making since 1937, these funds should never have
been returned to the airline. This action resulted in a strong equity position
for Air Canada and created an anomaly in the sense that the competition
between carriers was biased, as the private sector carriers were faced with
private sector market conditions.
INTERNATIONAL AIR SERVICE AGREEMENTS: NO

DEREGULATION IN INTERNATIONAL AIR SERVICE
Historically, Transport Canada approached the issue of bilateral air service

negotiations on an ad hoc basis that featured incremental negotiations. In
order to facilitate a proactive approach to liberalizing Canada’s bilateral
air transportation agreements, Transport Canada, in 2006, announced
Blue Sky, Canada’s new international air policy. Broadly speaking, the
Blue Sky Policy indicated that Canada would seek to negotiate Open-
Skies-type agreements when deemed in Canada’s and Air Canada’s overall
best interest.
The Blue Sky policy stated the following five objectives:6
1. Provide a framework that encourages competition and the develop-

ment of new and expanded international air services to benefit travel-
lers, shippers, and the tourism and business sectors.

M4371-FINGER_9781786431851_t.indd 35 14/11/2017 11:14

2. Provide opportunities for Canadian airlines to grow and compete suc-

cessfully in a more liberalized global environment.
3. Enable airports to market themselves in a manner that is unhindered
by bilateral constraints, to the greatest extent possible.
4. Support and facilitate Canada’s international trade objectives.
5. Support a safe, secure, efficient, economically healthy and viable
Canadian air transportation industry.
Objectives 1 to 3 are, in our view, the primary objectives of Blue Sky

and are the aspects of Canada’s air transportation industry that can be
influenced most directly by the policy. Objectives 4 and 5 are secondary
objectives, of which Blue Sky has a supporting role. Transport Canada
maintains that there is no such hierarchy, and that all five objectives are of
equal importance. These five objectives are the intended consequences of
new and/or expanded bilateral agreements. In pursuing these objectives,
the negotiating process is guided by the following four principles:
●● recognize that air transportation is a direct contributor to a dynamic

economy and is a leading trade and tourism facilitator;
●● market forces should determine the price, quality, frequency and
range of air services options;
●● Canadian air carriers should have the opportunity to compete in
international markets on a reasonably level playing field; and
●● air liberalization initiatives will continue to be guided by safety and
security considerations.
It has been stated that the primary objective of Canada’s international

air policy is to seek to negotiate reciprocal Open-Skies-type agreements,
similar to the one negotiated with the US in November 2005, where it was
deemed to be in Canada’s overall interest. Open-Skies-type agreements
cover the following elements, known as air freedom rights, for scheduled
passenger and all-cargo services: open bilateral markets/access (third and
fourth freedom rights); no limit on the number of airlines permitted to
operate; no limits on the permitted frequency of service or aircraft type;
market-based tariff/pricing regime for bilateral and third-country services;
open and flexible regime for the operation of code-sharing services; unre-
stricted services to and from third countries (fifth and sixth freedom rights);
and rights for stand-alone all-cargo operations (seventh freedom right).
The Blue Sky policy document states that in determining Canada’s
negotiating priorities, several issues are taken into consideration in consul-
tation with both airlines and airports: Canadian airline and airport priori-
ties and interests; the likelihood and extent of new Canadian and foreign

M4371-FINGER_9781786431851_t.indd 36 14/11/2017 11:14

carrier services, and giving preference where early start-up of air services
is planned; the size and maturity of the air transportation markets and
the potential for future growth; foreign government requests; Canada’s
international trade objectives; safety and security issues; foreign relations;
and bilateral irritants and disputes. Note that the consumer’s interest is not
considered.
The introduction of the Blue Sky policy did not indicate the changes
from the previous policy, nor the reasons for the changes. There was no
statement as to why, or in what way, the prior policy was deemed inad-
equate. It was not clear how the Blue Sky policy was going to fix something
that was not working. Given the way in which the ‘new’ international air
policy was designed, it was clear it was not open skies by any means –
hence the term ‘Blue Sky’.
One seeming failure of the international air service policy is the low
frequency of service allowed for the majority of foreign carriers. In many
cases, foreign carriers do not fly more than three flights per week. With
such infrequent service, there is little chance of serving high-yield business
traffic. Such low allowable service frequencies make it nearly impossible to
develop and grow a market. When examining the countries that Canada
has agreements with, many are small and in many cases fares are restric-
tive, with single disapproval conditions, and when designating intermediate
points, they are nearly all Star Alliance hubs. The US, ICAO and IATA all
define a bilateral agreement as ‘liberal’ if frequencies for third, fourth and
fifth freedom traffic are unlimited. Canada does not use this definition
of ‘liberal’. Canada uses a Canadian maximization model rather than a
global maximization model. If the market demands only three weekly fre-
quencies, the market will yield that outcome; therefore, permitting single
daily service will do no harm. Defining limited frequencies ex ante never
gives the market an opportunity to work. Limiting access by foreign carri-
ers is clearly designed to protect AC.
CANADA–EU OPEN SKIES AGREEMENT
On 9 December 2008, Canada and the European Union successfully con-

cluded negotiations on a Comprehensive Air Transport Agreement. The
negotiations had commenced in November 2007 and the Agreement came
into effect in the first half of 2009.
This is a significant agreement pursued as part of Canada’s Blue Sky
international air policy, given the number of member states in the EU.
As stated, the purpose of the Blue Sky policy was to negotiate increas-
ingly liberalized international air transport agreements. At the time of

M4371-FINGER_9781786431851_t.indd 37 14/11/2017 11:14

the negotiations with the EU for a common EU agreement, Canada had

bilateral agreements with 19 of the 27 European Union member states. A
single agreement now governs Canada’s air transport relations with every
member of the European Union.
The Canada–EU Agreement is more far-reaching and open than any
similar pact that the EU or Canada has with other countries. Canada and
the EU agreed to remove restrictions on direct flights between Canada and
EU member states, and ultimately to remove restrictions on foreign owner-
ship of Canadian air carriers. The agreement provides for increased traffic
rights, allowing for unrestricted direct services (on an airline’s own aircraft
or that of another air carrier) between Canada and the EU, without any
limitations on the number of flights operated, the routes or the fares to
be offered. All-cargo airlines were allowed to operate to or from a third
country on flights involving Canada or the EU.
The agreement provided for a phased market opening linked to the
granting of greater investment freedoms by both sides:
●● Phase I took effect in the first half of 2009. Airlines now have unlim-
ited freedom to operate direct services between any point in the
European Union and any point in Canada. There are no restrictions
on the number of airlines flying between the EU and Canada, or on
the number of services operated by any airline. Cargo airlines have
the right to fly onward to a third country.
●● Phase II of the Canada–EU Open Skies Agreement anticipates
that Canada will amend its legislation to enable European investors
to own up to 49 per cent of a Canadian carrier’s voting equity, an
increase from the current 25 per cent. At that time, further traffic
rights will be granted, including the right for cargo carriers to
operate services to a third country, for Canada or the EU, to a third
country, without connection to their point of origin.
●● In Phase III, both sides would allow investors to establish and
control new airlines in each other’s markets. At that point, passenger
airlines would be able to fly onward to a third country.
●● In Phase IV, EU and Canadian carriers would be granted full rights
to operate between, within and beyond both markets, including
between points in the territory of the other party (known as cabo-
tage). These rights would be granted after both sides complete the
necessary steps to allow each member full ownership and control of
their carriers.
If Canada and the EU have indeed made binding commitments with

respect to Phases II to IV, it represents a fundamental change in Canada’s

M4371-FINGER_9781786431851_t.indd 38 14/11/2017 11:14

policy framework for aviation and will present significant opportunities

for carriers in both jurisdictions. If these commitments were realized, the
agreement would likely have led to consolidation and integration between
air carriers in the two jurisdictions. However, the recently approved metal
neutral7 joint venture A11 involving Lufthansa, United-Continental and
Air Canada on the transatlantic, together with similar metal neutral joint
ventures by the other two global alliances in the transatlantic, have effec-
tively blunted any possible gains in liberalization and competition that the
Canada–EU air service agreement might have generated. In our view, we
will never see Phases II to IV.
TRANSBORDER MARKET
Prior to 1995, air services between Canada and the US were governed by
a relatively restrictive bilateral air services agreement. This agreement set
out conditions on fares, capacity, cities served, carrier designation and
airport access. As a result, there was limited service between the two coun-
tries. However, after the 1995 Open Skies Agreement was signed, within
three days 17 new city-pair routes were approved by the US Department
of Transportation (DOT), of which five were completely new routes. By
the end of 1995, 16 US and Canadian carriers were offering service within
the Open Skies Agreement. By August 1997, a total of 79 new routes
were established between Canada and the US (20 of these routes were
subsequently discontinued). Canadian carriers established 29 of these new
routes, and American carriers established 50 (see Dubey and Gendron,
1999).
Canada and the US signed an Air Service Agreement (ASA) on 24
February 1995.8 This agreement allowed Canadian and US carriers to
operate air services between any point in either country using any aircraft
and at any service frequency, and effectively removed price controls on
transborder routes.9 However, there were phase-in provisions over a three-
year transition period that: (a) limited the presence of US-based carriers
on routes involving Montreal, Vancouver and Toronto (all existing route
rights at that time held by US carriers were not affected); (b) restricted
Canadian airlines from accessing slots in US airports; and (c) reduced the
ability of airlines to engage in code-sharing services across the Canada–US
border.
At the time that the ASA was signed, it was expected that the removal
of constraints on the ability of carriers to freely serve transborder routes
would result in greater competition on these routes. As of 2012, this was
not the case.

M4371-FINGER_9781786431851_t.indd 39 14/11/2017 11:14

Significant liberalization of the Canada–US aviation market through

the 1995 Open Skies Agreement did increase the overall number of trans-
border routes. From 1994 to 2005 the number of US destinations served
from Canadian airports increased from 38 cities in 1994 to 59 by 2005. The
number of Canadian cities with transborder service rose from 21, in 1994,
to 27 in 2005. The number of air carriers providing transborder air service
grew from 11 to 19 carriers, and the average transborder daily seat capac-
ity grew by 49 per cent (Mak, 2005). The majority of new flights between
Canada and the US linked more US cities with Canadian gateways (that is,
hubs of Air Canada and Canadian Airlines International).
However, the increase in the number of transborder routes did not
generally result in increased competition on individual transborder routes.
Market concentration as measured by the Herfindahl–Hirschman concen-
tration index (HHI) either stayed the same or declined in most markets
between 1995 and 1998. The HHI is a common measure of market con-
centration used in studies examining the impact of the number and size
distribution of firms on different performance measures such as prices,
profits, costs and so on (Pustay, 1999). Toronto was the one exception:
the HHI of flights involving Toronto actually increased due to the restric-
tions on US carriers to enter markets in Toronto during the three-year
phase-in period. From 1995 through to 1998, Air Canada and Canadian
Airlines International (CAI) increased total weekly transborder flights
from Toronto from 251 to 628 and from 5 to 128, respectively (Pustay,
1999).
There are some routes that are small and may support only one or
two carriers. However, observing a small number of carriers on a route
should not necessarily lead to the conclusion that the market is small.
Low numbers can also result from market dominance, which is what hap-
pened on a number of major transborder routes. When an airline like AC
dominates a market, there is insufficient residual traffic to support many
other carriers, which is a significant reason why we see very little competi-
tion between AC and WestJet on transborder routes.10 Air Canada entered
into a series of cooperative agreements with United Airlines, beginning in
1995. These agreements have allowed Air Canada and United to coordi-
nate on key aspects of competition on transborder routes and ‘generally
cover all of the parties in the joint venture’ on transborder routes. The
agreements between Air Canada and United include the Air Canada–
United Coordination Agreement (May 1995) and the Air Canada–United
Expanded Coordination Agreement (May 1996). As a consequence of
establishing market dominance so soon after the first liberal Canada–US
agreement, not much has changed since 1995.
Pustay (1999) illustrated the significant degree of concentration in trans-

M4371-FINGER_9781786431851_t.indd 40 14/11/2017 11:14

border markets that existed from 1995 through to 1997. The restrictions
placed on entry by US carriers during the phase-in period at Vancouver,
Toronto and Montreal airports allowed Canadian carriers to increase
their dominance on transborder routes. Using flights as a measure for
calculating the degree of concentration, Pustay (1999) showed that Air
Canada increased their dominance across all transborder routes during this
period.11 At Toronto, Air Canada’s primary hub, AC increased its market
share of flights to 55 per cent, while the next largest carrier (American
Airlines) had a mere 10 per cent of the market for transborder flights in or
out of Toronto. Pustay (1999) noted that after the phase-in period ended,
the US carriers were able to expand the number of flights; however, Air
Canada’s dominance was still significant. Montreal experienced a differ-
ent outcome, with Air Canada’s dominance declining as Delta, American,
Continental and Northwest all added flights into Montreal.
In 2005, Canada negotiated a revised Open Skies Agreement with the
US that was signed in 2007. This agreement liberalized so-called ‘fifth’
and ‘sixth’ freedoms.12 With these new provisions in the Open Skies
Agreement, a carrier can now use a single through flight number to
operate from a foreign country to its home country and then fly to another
foreign country. However, even this further liberalization has not resulted
in greater competition on most transborder routes, as only a small number
of carriers have actually initiated service that takes advantage of these
additional freedoms.
What is striking when examining the data for the top eight airports in
Canada is the relatively flat growth in transborder passenger traffic in
nearly all markets. From 1997 through to 2010, the annual growth rate has
ranged from a low of 0.2 per cent in Vancouver to 6.8 per cent in Halifax;
Toronto has grown by only 0.7 per cent annually. Such low growth rates
could reflect a number of factors, including a lack of aggressive pricing
and lack of competition. Despite liberalizing the transborder market in
1995, and further moves to an open skies policy signed in 2007, the growth
in this market has been relatively small. The reasons are high fares due to a
lack of competition and the market dominance of Air Canada and its Star
Alliance partners. Evidence of this appears in a study of demand leakage
in the Canada–US transborder market: Elwakil et al. (2013) find an esti-
mated leakage of 4.7 passengers in 2008 who flew from/to US domestic
airports rather than take a transborder flight. In addition, after controlling
for route-specific factors, they find that fares are approximately 28 per cent
higher in the transborder market. Elwakil et al. (2013) conclude that ‘The
lack of low-cost carrier competition likely contributes to the higher trans-
border fares as might the cozy relationship between US and Canadian
carriers resulting from alliance membership and antitrust immunity for

M4371-FINGER_9781786431851_t.indd 41 14/11/2017 11:14

price-setting and scheduling.’ (p. 55). Without sufficient competition in the

Canadian domestic market, such demand leakages are likely to continue.
THE CURRENT STATUS OF THE METAL NEUTRAL

JOINT AGREEMENT: AIR CANADA AND UNITED-
CONTINENTAL HOLDINGS (UCH)
As noted earlier, transborder routes are dominated by Star Alliance. This
dominance can be measured in a number of ways. First, an airline with
a greater share of flights at an airport will capture a greater share of
passengers; this is known as the ‘S-curve effect’. If an airline has more
capacity, it can offer higher levels of service to passengers, particularly
business passengers, because it is more likely to have a flight at a preferred
time. This effect is diminished in markets where there are high levels of
effective competition, because price becomes a more significant factor
for consumer choice. Star Alliance dominates flights at each of Canada’s
major airports. Air Canada/UCH are at least twice as large as the next
largest firm in terms of capacity and traffic for transborder flights from
Montreal, Toronto, Calgary and Vancouver, and depending on the
airport, are in most cases three times (or more) larger than the next-largest
firm.
The transborder joint venture and transatlantic joint venture (TJV)
provide AC with a broader and deeper integration of the passenger air
transportation services between AC and United. Such integration with
respect to TJVs would include (among other things) joint pricing, joint
route planning and scheduling, coordinated marketing, harmonization
of sales processes and revenue sharing between the airlines. The TJVs,
coupled with pre-existing alliance agreements between AC and UCH,
effectively merge AC and United-Continental into a single competitor on
transborder and transatlantic routes.
During 2014, Air Canada offered a non-stop passenger air transpor-
tation service between 93 transborder city pairs. Air Canada accounted
for approximately 34 per cent of the seat capacity on all transborder
flights, and it operated flights on 46 per cent of the transborder city
pairs where non-stop service was offered. The largest rival Canadian
carrier, WestJet, accounted for only 12 per cent of the seat capacity on
transborder flights, and it operated flights on 32 per cent of transborder
city pairs where non-stop service was offered during 2014. AC oper-
ates hubs at Toronto’s Pearson International Airport, Montreal’s Pierre
Elliott Trudeau International Airport, Calgary International Airport
and Vancouver International Airport. Approximately 97 per cent of Air

M4371-FINGER_9781786431851_t.indd 42 14/11/2017 11:14

Canada’s 2014 transborder capacity originated or terminated at one of

these hubs; this would include Origin–Destination (O–D) as well as con-
necting traffic.
While there may be pro-competitive effects from carriers forming an alli-
ance with respect to connecting passengers, as the alliance members offer
complementary inputs (that is, complementary flight segments that com-
prise a connecting itinerary), the prospect exists for alliances to create anti-
competitive effects on non-stop overlap routes, where they offer products
that are substitutes for one another. Reitzes and Moss (2008) demonstrated
that in the case of transatlantic routes O–D fares increased by nearly 11
per cent. In a recent paper, Bilotkach and Huschelrath (2013) showed that
metal neutral joint ventures have led to market foreclosure, such that non-
alliance carriers reduce their traffic by between 4.1 per cent and 11.5 per
cent. While it is not possible to undo 15 years of evolving market structure,
it is possible to reduce barriers to entry. Going forward, it is conceivable to
have market entry and alternative alliance entry. Two barriers are foreign
ownership restrictions and rights of establishment.
FOREIGN OWNERSHIP RESTRICTIONS
Rights of establishment work jointly with the removal of foreign owner-

ship restrictions. There is no compelling reason why the air transport
sector should have a restriction of 25 per cent foreign ownership. Some
may argue that the current bilateral system places some restrictions on
moving beyond 49 per cent foreign ownership. However, cross-national
mergers over the past several years within the EU and between the EU and
US would blunt any such argument, particularly since the US maintains a
25 per cent foreign ownership limit.
Currently, Chile is the only country in the world where there is no
limit on foreign ownership of airlines operating within or to and from
the country. The United States is restrictive, with the maximum level of
foreign ownership of voting stock at 25 per cent, yet the US may allow a
higher percentage of foreign ownership of non-voting stock on a case-by-
case basis. Both New Zealand and Australia have imposed a 49 per cent
ownership limit on international carriers, and there is no limit on foreign
ownership on purely domestic carriers. The EU has taken additional steps
that focus on liberalization within the European Economic Area, and sets
a 49 per cent limit on foreign ownership.
Restrictions have costs by limiting access to capital and management
expertise, limiting economies of density and scope, and can lead to
fragmented air service. Ownership restrictions pose a barrier to entry

M4371-FINGER_9781786431851_t.indd 43 14/11/2017 11:14

that may make entry impossible or expensive and that may invite state
aid, which would create market distortions. What we observe is coun-
tries moving to higher allowable foreign ownership limits when capital
is needed – such as Brazil and India, who moved to 49 per cent foreign
ownership limits.
In 2016, under pressure from a small group of aviation entrepreneurs,
Canada lobbied the government for a change in the foreign ownership
restrictions. After much consultation, the government moved to change
the restrictions from 25 per cent to an upper limit of 49 per cent foreign
ownership. The new rules ensure, however, that no single foreign investor
or group of foreign investors can have more than a 25 per cent ownership
stake.
What are the differences between a merger between carriers, a joint
venture within a strategic alliance, and allowing foreign capital to invest
in a national airline either up to some limit or with no restrictions? Would
the strategic direction, pricing, product offering, route choice and network
configuration differ under each of these three different scenarios? One
significant difference between mergers, joint ventures and allowing [more]
foreign capital is that the foreign capital can come from anywhere and
not be limited to the airline business, as mergers or joint ventures would
be. Also, and importantly, with mergers or joint ventures a competitor is
removed from the market, which – as economic theory and empirical evi-
dence show – generally reduces economic welfare.
Airlines have invested in other airlines for many reasons. Lufthansa’s
decision to acquire a 20 per cent shareholding in British Midland may well
have been motivated by a desire to gain more slots at Heathrow Airport,
and they may have been similarly motivated with their 18 per cent pur-
chase of JetBlue, which has a major hub at slot-constrained JFK in New
York. Developed countries with small populations, or countries that are
isolated, have tended to have a more liberalized approach to their airline
markets in order to expand the networks of their national airlines. It could
be one of the key reasons that Australia and New Zealand formed a single
aviation market to allow their citizens to invest in each other’s airlines. It
is also why Korea and the Netherlands have taken a liberal approach to
aviation access. Cross-border airlines are springing up more frequently;
in express delivery, UPS, FedEx and DHL, in Europe, AF/KLM, LH/SR/
OS/SN and BA/IB. Airline families have emerged: Grupo TACA, LAN
Airlines, Air Asia and Tiger as well as mergers such as TACA1Avianca
and LAN1TAM, which produced LATAM.

M4371-FINGER_9781786431851_t.indd 44 14/11/2017 11:14

SUMMARY
The legacy of Canadian air policy is to have effectively one international

(acknowledging the charter airline Air Transat has some minor presence)
and one major regional airline and several small regional carriers; to have
dominance by one alliance; and to have one Canadian national/interna-
tional carrier integrated into one of the three global alliances and into two
metal neutral joint ventures that control capacity, scheduling and fares in
key markets in the transatlantic and transborder markets.
The consequence of the Canadian air policy is to have marginal – if any
– domestic entry by carriers into multiple markets; Porter is recognized as
a constrained entrant since it operates from the Toronto Island Airport
and operates only turboprop aircraft, along with the airport having no
US preclearance. Some other outcomes of the air policy include foreign
ownership constraints, domestic airfares exceeding those in the US, and
higher international and transborder airfares. Some of the differences in
fares are a result of Federal and Provincial taxes and fees, but a significant
explanation for the fare differences is the lack of competition. The Blue
Sky approach to air service agreements has had no statistically significant
impact on total air passenger traffic, total tourism GDP, foreign direct
investment in Canada or the amount of trade in exports.
A new approach to air policy would define the public interest as more
than just the financial sustainability of Canadian airlines and employment
in the aviation sector. The Canada Transportation Act makes it clear that
transportation is to be efficient and effective to meet the needs of consum-
ers and shippers. It also makes it clear that effective competition within
and between modes is the mechanism through which to achieve efficiency.
A broad view of what constitutes the public interest in air policy would
take into account the impact on consumers, tourism, trade and connectiv-
ity to global markets. We cannot disassemble what history and policy have
created, but it is possible to remove barriers to entry and adopt a new Blue
Sky policy that recognizes that interests vary from region to region; what is
good for Toronto is not necessarily good for Calgary, Halifax or Vancouver.
A new approach to air policy would let markets work, have more effective
and potentially effective competition by putting in place rights of estab-
lishment, raising foreign ownership limits to 49 per cent and having air
service agreements that have a specific template and are not negotiated
on a case-by-case basis. The specific features would include pricing (at
most double disapproval) and capacity freedom, a minimum of allowed
single daily service, full fifth and sixth freedom rights for passengers and
the seventh for cargo airlines, and no airline designation – meaning any
Canadian airline can provide service. Specifying at a minimum of a single

M4371-FINGER_9781786431851_t.indd 45 14/11/2017 11:14

daily service means that an airline can provide this level of service if it sees
a business case in doing so. Airlines would not be constrained to fewer
weekly frequencies and would not be constrained to have symmetry with
Canadian and foreign carriers’ frequencies. The US, for example, pursues
a ‘single’ open skies policy with every country, with the ‘open skies’ being
as liberal as possible: unrestricted first to sixth freedoms with an optional
cargo seventh freedom, market-based pricing freedoms, extensive doing-
business protections, and strong safety and security articles are contained
in the agreements.
A new Canadian air policy would have an air service policy that lets
carriers of the two negotiating countries decide when and where to enter
a market, and how often. The carriers choose what is in their commercial
interests, rather than having government decide, subject to competition
laws.
The dynamics of the market come from innovation in products and
processes, and these outcomes generally come from outside the existing
market structure. How much competition is too much or not enough?
Most important markets in Canada are dominated by a small number of
firms, such as banking, telecommunications, large accounting firms and,
of course, air transport. As John Kay (2010) pointed out, ‘in all these
industries established businesses will tell you they welcome competition,
but they also explain enough is enough, that users already have a wide
array of products at competitive prices’. The firms argue that public
policy should be directed at investment and innovation and that it is global
competitiveness, not domestic competition, that matters. Unfortunately,
government seems to be all too accepting of these industry arguments. The
benefits of competition are not about today’s customers, but about future
customers. However, competition and competitiveness are different things:
tomorrow’s customers will have innovative products produced and deliv-
ered through innovative processes; without competition, the LCC business
model would not have evolved, and WestJet or Porter would never have
come into existence.
Fundamental to aviation policy and deregulation is to decide how
Canada defines the public interest for air policy, especially international
air policy. What weight is, or should be, placed on the interests of con-
sumer and aviation-dependent economic sectors vs. air carrier interest?
How explicit does this have to be? Do carrier interests include the interests
of the alliances that a Canadian carrier joins, or are they confined to the
Canadian carrier’s own flights only? Are we protecting Canadian carrier
capacity or its profits, including alliance-derived profits? Should Canadian
air policy protect one foreign carrier aligned with a Canadian carrier from
another foreign carrier?

M4371-FINGER_9781786431851_t.indd 46 14/11/2017 11:14

Has airline and airport deregulation been a success? By almost any

measure, one would have to say yes. Every assessment we have read in the
academic and industry literature makes it clear that fares, service, entry,
new business models, employment and network redesign and reach have
been net positive. Do airlines fail? Yes, but so do firms in other industries;
aviation is no different in that regard. It is also true that airlines, being at
the end of the supply chain, face tough competition. This is not unique:
all firms at the end of the supply chain face the same circumstances,
unless they are able to vertically integrate or contract to control entry or
form strategic alliances with upstream suppliers. Another way airlines can
attempt to achieve this same result is via government protection. This need
not be formal; simply assistance that constrains entry and influences factor
markets. Such assistance creates rents, and economic agents such as labour
and airports chase rents. Rents are translated into higher costs, and these
higher costs are in perpetuity. Air Canada will continue to be a high-cost
carrier unless the source of rents is removed. One approach is to reduce or
remove barriers to entry in order to allow actual or potential competition
through the rights of establishment. In addition, the recent move to allow
up to 49 per cent foreign equity ownership would have a positive effect on
enhancing competition. A significant change to the 2005 revised Blue Sky
policy is needed that places the property rights for access to Canada and
the Canadian market in the hands of Canadians.
NOTES
1. We use the term ‘cost-oriented carrier’ since WestJet is no longer a true low-cost carrier
(LCC); it has deviated significantly from its original LCC model based on Southwest
Airlines of the US.
2. The Canada–US market is referenced differently from other international aviation
markets given the proximity to the market over the entire US–Canada border and the
fact the US population is 316 million, nearly ten times that of Canada. The econo-
mies are also tightly linked as a result of various trade policies, particularly the North
American Free Trade Agreement between Canada, Mexico and the US.
3. In 2015 total passenger traffic in Canada was 133 354 859, with the top four airports
(Toronto, Vancouver, Montreal and Calgary) handling 70 per cent of traffic, and the top
eight airports handling 84 per cent of total traffic. Toronto Airport handled 31 per cent
of total traffic while Vancouver, the next closest, handled half that amount of passenger
traffic.
4. For rail, this only applies if the railway crosses provincial boundaries. For marine it
includes all waterways, even those wholly within a province.
5. Early in its life, Air Canada received some cash subsidies from the federal govern-
ment. These took two forms: the government paid for the airline’s deficits in its first
few years of operation, and Air Canada was given contracts to carry the mail at high
rates. The mail contracts represented 94 per cent of the airline’s operating revenues
in 1938, but this share fell to 37 per cent in 1943 and 1944, and to only 15 per cent in
1951.

M4371-FINGER_9781786431851_t.indd 47 14/11/2017 11:14

6. The discussion in this section is based on the official Blue Sky policy document; see
http://www.tc.gc.ca/pol/en/ace/consultations/blueSky.pdf.
7. Metal neutral joint venture is a form of partnership between airlines. It is structured
so that partners in the venture are indifferent as to which one operates the ‘metal’ (air-
craft) when they jointly market services. Metal neutrality can be achieved through cost-,
revenue- and/or comprehensive benefit-sharing arrangements.
8. The United States considered the 1995 agreement with Canada to be an ‘open trans-
border agreement’ which, however, did not meet all the requirements of an ‘Open Skies
agreement’ as defined by the DOT.
9. Standing Committee on Transport ‘Air Liberalization and the Canadian Airports
System: Interim Report’, May 2005, available at: http://www.parl.gc.ca/content/hoc/
Committee/381/TRAN/Reports/RP1858323/TRAN_Rpt04/TRAN_Rpt04-e.pdf.
10. WestJet serves five US cities with its own aircraft (Chicago O’Hare (ORD), Los Angeles
(LAX), Las Vegas, New York (LGA) and Hawaii). It provides service to approximately
30 destinations in the US via its codeshare partners.
11. Measuring flights can underestimate the degree of concentration and hence poten-
tial market power. For example, Alaska Airlines increased flights into Vancouver
using turboprop aircraft, which seat 50–70 people, while other carriers may be
using 120–140 seat jet aircraft. Seats would be a better measure of input metric of
concentration.
12. The fifth freedom right is the right for the carrier to enplane traffic at one foreign point
and deplane it in another foreign point as part of continuous operation, also serving the
carrier’s homeland. For example, a Canadian carrier could fly from Vancouver to any
city in California, drop off and pick up passengers, and continue the trip to a point in
Mexico.
BIBLIOGRAPHY
Bilotkach, V. and K. Huschelrath (2011), ‘Antitrust Immunity for Airline Alliances’,

Journal of Competition Law & Economics, 7(2), 335–80.
Bilotkach, V. and K. Huschelrath (2013), ‘Airline Alliances, Antitrust
Immunity and Market Foreclosure’, Review of Economics & Statistics, 95(4),
1368–85.
Dubey, S. and F. Gendron (1999), The U.S.–Canada Open Skies Agreement: Three
Years Later, Statistics Canada Cat: 8 7–003-XIB, 18(3).
Elwakil, O.S., R.J. Windle and M.E. Dresner (2013), ‘Transborder demand leakage
and the US–Canadian air passenger market’, Transportation Research Part E:
Logistics and Transportation Review, 57, 45–57.
Gillen, D. (1988), ‘Entry Barriers and Contestable Canadian Airline Markets’,
International Journal of Transport Economics, 15(1), February, 29–41.
Gillen, D. (2006), ‘Airline Business Models and Networks: Regulation, Competition
and Evolution in Aviation Markets’, Journal of Network Economics, 5(4),
December.
Gillen, D. and T. Hazledine (2011), ‘The New Pricing in North American Air
Travel Markets: Implications for Competition and Antitrust’, in J. Peoples (ed.),
Pricing Behaviour and Non-Price Characteristics in the Airline Industry: Advances
in Airline Economics, Volume 3, Bingley: Emerald Publishers.
Gillen, D., M. Hansen and R. Ramos (1990), ‘Assessing the Alternatives to Liberalize
the Canada–U.S. Air Transport Bilateral’, Journal of the Transportation Research
Forum, 31(2), 326–36.

M4371-FINGER_9781786431851_t.indd 48 14/11/2017 11:14

Gillen, D., R. Harris and T. Oum (2002), ‘Measuring the Economic Effects of
Bilateral Liberalization on Air Transport’, Transportation Research, 28E(3–4),
May.
Gillen, D., T. Oum and M. Tretheway (1985), Canadian Airline Deregulation
and Privatization: Assessing Effects and Prospects, Vancouver, Centre for
Transportation Studies, University of British Columbia.
Gillen, D., T. Oum and M. Tretheway (1986), Airline Cost and Performance:
Implications for Public and Industry Policy, Vancouver, Centre for Transportation
Studies, University of British Columbia.
Gillen, D., T. Oum and M. Tretheway (1986), ‘Entry-Barriers and Anti-Competitive
Behaviour in a Deregulated Canadian Airline Market’, Transportation Research
Forum Proceedings, pp. 483–93.
Gillen, D., T. Oum and M. Tretheway (1986), Identifying and Measuring the Impact
of Government Ownership and Regulation on Airline Performance, Economic
Council of Canada, 1986.
Gillen, D., T. Oum and M. Tretheway (1988), ‘Duopoly in Canada’s Airline
Industry: Consequences and Policy Issues’, Canadian Public Policy, 14(1),
March, 15–31.
Gillen, D., T. Oum and M. Tretheway (1989), ‘Privatization of Air Canada; Why
it is Necessary in a Deregulated Environment’, Canadian Public Policy, XV(3),
September, 285–99.
Gillen, D., T. Oum and M. Tretheway (1990), ‘The Cost Structure of the Canadian
Airline Industry’, Journal of Transport Economics and Policy, 24(1), January,
9–34.
Gonenc, R. and G. Nicoletti (2001), ‘Regulation, Market Structure and Performance
in Air Passenger Transportation’, OECD Economic Studies, No. 32, 183.
Howe, C.D. (1938), Canadian Parliamentary debate, 21 March, available at http://
www.lipad.ca/full/1938/03/21/17/.
Iacobucci, E., M. Trebilcock and R. Winter (2007), ‘The Political Economy of
Deregulation in Canada’, in Martin Levin (ed.), Creating Competitive Markets:
The Politics of Regulatory Reform, Brookings Institute.
Kay, J. (2010), ‘Radical Innovation Rarely Comes from Within’, Financial Times,
24 November.
Levine, M. (1987), ‘Airline Competition in Deregulated Markets: Theory, Firm
Strategy and Public Policy’, Yale Journal on Regulation, 4(6), 393–466.
Mak, D. (2005), ‘Transborder Air Services since 1994’, in Canadian Aviation
Intelligence Report, InterVISTAS, August, available at http://www.intervistas.
com/downloads/CAIR/issues/2005/08_aug2005.pdf.
Mentzer, M.S. (2000), ‘The Impact of Discount Airlines on Domestic Fares in
Canada’, Transportation Journal, 39(4), Summer.
Morrison, S.A. (1993), ‘Deregulation and Competition in the Canadian Airline
Industry’. Paper prepared for the Royal Commission on National Passenger
Transportation.
Morrison, S.A. and C. Winston (2000), ‘The Remaining Role for Government
Policy in the Deregulated Airline Industry’, in S. Pelzman and C. Winston (eds),
Deregulation of Network Industries: What’s Next?, Washington: AEI-Brookings
Joint Center for Regulatory Studies.
Oum, T., A. Zhang and Y. Zhang (1993), ‘Inter-firm Rivalry and Firm Specific Price
Elasticities in Deregulated Airline Markets’, Journal of Transport Economics &
Policy, 27(2), 171–92.

M4371-FINGER_9781786431851_t.indd 49 14/11/2017 11:14

Pustay, M.W. (1999), ‘Competition and Concentration in Canadian–U.S.

Transborder Aviation Markets’, Transportation Journal, 38(4), 5–17.
Reitzes, J. and D. Moss (2008), ‘Airline Alliances and Systems Competition’,
Houston Law Review, 45(2), 293–332.
Ross, T. and W. Stanbury (1999), ‘Avoiding the Maple Syrup Solution: Comments
on the Restructuring of Canada’s Airline Industry’, Public Policy Sources, No.
32, November.

M4371-FINGER_9781786431851_t.indd 50 14/11/2017 11:14

4. Australia – a reluctant liberalizer
Peter Forsyth
INTRODUCTION
Australia has long been a reluctant liberalizer in air transport. This has
been true for both domestic and international markets. While being cau-
tious with air transport liberalization, this has not been the case with other
industries, such as energy, telecommunications and the finance sector.
Australia has been deregulating at about the same time as other industrial
countries other than the US, and it has privatized major government firms
earlier than most.
With domestic markets, Australia was a keen observer of the perfor-
mance of US deregulation. In the late 1980s it decided to deregulate, and
deregulated in 1990. While the market is now totally deregulated, this has
come about by accident rather than design. Although entrants were free to
enter from 1990, there was a significant barrier: the incumbent airlines had
control of the terminals and gates, and entrants were not successful. The
collapse of one of the incumbent airlines – Ansett – in 2001 and the pri-
vatization of the airports over 1997 until 2002 opened up terminal access,
and entrants were able to prosper.
In the international market Australia had a traditionally regulated
industry. However, by the late 1970s, there was strong pressure from the
newly established Asian airlines to enter the market; with significantly
lower costs, they were able to offer much lower fares to Asia and Europe
than the incumbent airlines such as Qantas. The government fought hard
to keep the sixth freedom airlines out, but eventually, responding to public
pressure for lower airfares, it relented and allowed some measure of liber-
alization. Since then it has gradually liberalized, though it still prefers to
impose (fairly lax) capacity controls. A measure of liberalization is that it
has been prepared to accept a low market share for the Australian airlines.
The Australian experience with the competition from Asian and now Gulf
airlines has been similar to that now faced by European and US carriers
facing competition from Gulf airlines.
This chapter begins with a brief background. It examines, first, domestic
51

M4371-FINGER_9781786431851_t.indd 51 14/11/2017 11:14

liberalization and secondly, international liberalization. Five case studies

on international liberalization are then considered, after which conclusions
are drawn.
BACKGROUND
Australia is a large, highly urbanized country which makes considerable

use of air transport. Much of the traffic is between large cities (such as
Sydney, Melbourne and Perth) or tourism destinations (such as the Gold
Coast, Cairns and Darwin). For most long distance journeys (that is, over
500 km), the main options are only air and car: there are very few long-
distance trains (and there are no high-speed trains), and buses only have a
small share of the market. Almost all of Australia’s air transport markets
to and from other countries are a long distance away.
The air transport industry consists of international services, trunk
domestic services, and domestic regional and commuter services. The
federal government has responsibility for international and trunk services
as well as for some regional services, and state governments have respon-
sibility for regional services. Domestic airlines use aircraft such as Airbus
A330s, Airbus A320s and Boeing 737s, and most of the regional services
use turboprop aircraft such as Dash 8s and ATR 72s, along with smaller
aircraft for the low-density routes. Domestic trunk routes are served by
full-service carriers (FSCs) such as Qantas and Virgin Australia, and by
low-cost carriers such as Jetstar and Tiger Australia. Qantas and Virgin
Australia serve regional routes, as do dedicated regional airlines such as
Regional Express (REX) and smaller airlines. A relatively unique aspect of
the Australian market has been the development of charter operations to
serve the fly-in-fly-out (FIFO) flights to serve the mining industry.
Most of the air transport industry in Australia is now privately owned.
In 1994 the federal government sold Qantas, which is slightly more than
50 per cent Australian-owned (the Qantas Sale Act prescribes a minimum
of 50 per cent Australian ownership). By contrast, Virgin Australia comes
under different ownership regulations: the domestic operation is permit-
ted to be 100 per cent foreign-owned, but its international operation is
required to be at least 50 per cent Australian ownership. Currently the
Virgin Group, Etihad, Singapore Airlines and now two Chinese airlines
have major stakes in the airline, though there is a small percentage which
is listed on the stock exchange. Qantas has long been concerned at the dif-
ferent ownership rules governing it and its major Australian competitor.
The federal government privatized all of the major airports between
1997 and 2002 (Forsyth, 2002). Most of these are now unregulated, the

M4371-FINGER_9781786431851_t.indd 52 14/11/2017 11:14

Australia – a reluctant liberalizer 53
exceptions being Sydney, Melbourne, Brisbane and Perth, which are now
subject to light-handed regulation. Some airports are listed on the stock
exchange (Sydney, for example) and others are owned by investors such as
Australian and overseas pension funds. Smaller airports are mostly owned
by local authorities.
As with other countries, international and domestic aviation are regu-
lated under different arrangements. International air transport is handled
by bilateral agreements, with the exception of routes to New Zealand,
which are handled under the Single Aviation Market. Domestic trunk ser-
vices are governed under the federal government, and regional services by
the states (some of which still impose regulation).
DOMESTIC LIBERALIZATION AND THE

IMPORTANCE OF TERMINAL ACCESS
For most of the period between the end of WWII and 1990, Australia
imposed very tight regulation of domestic trunk airlines. The policy was
termed the Two Airline Policy. This policy was implemented rigorously
from about 1960 until the early 1980s, when it was loosened slightly
(Brogden, 1968; Forsyth, 1991). After a review in 1986 (May, 1986), in
1987 the government announced that it would deregulate the airlines in
1990, which it did. There was also tight regulation of regional services,
though the regulatory arrangements differed from state to state.
The underlying theme of the Two Airline Policy was a concern that com-
petition between a government-owned airline, Trans Australian Airlines
(TAA), and a private airline, Ansett, was inherently unstable – the govern-
ment airline would use subsidies to increase its market share, and drive
the private airline out of business. In the earlier, though not the later years
there were some grounds for this fear (Brogden, 1968). The only way of
eliminating this was to make the two airlines as similar as possible (and
thus preserve ‘competition’).
The core of the policy was rigorous control of capacity: the two airlines
were allowed to have exactly the same capacity in terms of seats. This was
achieved by the two airlines having exactly the same aircraft and additions
to capacity in the form of new aircraft arriving in Australia on the same
day. Capacity was carefully monitored, and additions to capacity were
allowed when target load factors were achieved. The outcome was that
actual load factors were constantly at 65 per cent, unless recessions meant
that actual load factors fell below this temporarily. Airfares were regulated
by the government, though in the 1980s an independent tribunal was set
up to regulate them (May, 1986). There was little diversity in airfares:

M4371-FINGER_9781786431851_t.indd 53 14/11/2017 11:14

there were first and economy fares, plus some discounts for students and
pensioners. Fares were set on a distance basis (which meant that long-haul
flights cross-subsidized short-haul flights) (Forsyth, 1991).
The nature of the regulation in Australia was rather different from that
imposed in the US before deregulation. There was no capacity control in
the US, but there was control over airfares, along with control over routes
(see Bailey et al., 1985). In Australia there was no control over routes, and
the airlines could operate the network they chose. As it turns out, they
chose to operate very similar networks. One feature of the Australian
system was that the airlines tended to operate ‘parallel schedules’ (Gannon,
1979; Hocking, 1972), which meant that there were two flights per day
from Sydney to a destination, one from each airline, and they tended to
depart within five minutes of one another (economists noted the relevance
of the Hotelling model of spatial competition).
The Two Airline Policy Under Scrutiny
The Two Airline Policy was implemented for many years without much
discussion. However, in the late 1970s it started being discussed more
critically, reflecting the worldwide re-evaluation of regulation, which had
begun in the US (Forsyth, 1991). Australia was slightly behind the US,
but by the late 1970s banking, telecommunications and transport regula-
tion were all being subjected to critical scrutiny. It was not surprising that
the Two Airline Policy became a topical issue. There were a small number
of economic evaluations of it; some politicians advocated its removal
(notably those from Western Australia, the most remote state and the one
most affected by high airfares), and airlines other than the two chosen
trunk carriers began seeking access to the market.
The complaints about the policy were several. There were not many
complaints about the airlines’ networks or about the quality of services,
which by international standards were high. However, there were consider-
able complaints about airfares, concerning both their level and structure
(Holcroft, 1981). Australian airfares were significantly above those of the
trunk airlines in the US. Long-distance flights, such as Sydney–Perth, were
particularly high relative to those in the US, or even between Australia
and New Zealand. There was little diversity in airfares, although the air-
lines did make some discount fares available from the late 1970s, notably
standby fares. Fares were regulated, and profits were not excessive, though
they were very stable. This suggested that the productive efficiency of
airlines was low compared to that of the US carriers. A number of pro-
ductivity studies confirmed that this was so (Forsyth and Hocking, 1980;
Mackay, 1979). While most states continued to regulate regional airlines,

M4371-FINGER_9781786431851_t.indd 54 14/11/2017 11:14

South Australia deregulated them; the results from this experiment were
encouraging (Starkie and Starrs, 1984).
While the Two Airline Policy became a policy issue, the two airlines were
very comfortable with it. The airlines and the government recognized that
there was pressure for change, but they did not wish to have major change,
let alone full deregulation. The government instituted a number of inquir-
ies (Holcroft, 1981; May, 1986), and there were several modifications to
the policy, both at the regulatory level (a new body to regulate airfares,
the Independent Air Fares Commission; see May, 1986) and at the airline
level (such as selected discount fares). By the middle of the 1980s the US
had deregulated its domestic airlines, and the results were regarded as
good, though there were some problem areas (for example, the failure of
some carriers). The US experience was very relevant for Australia, since it
became clear that there were alternatives to the Two Airline Policy.
Changes within the private carrier Ansett were significant. For many
years, ownership and control of Ansett was dominated by its founder, Sir
Reginald Ansett (who set up the airline in the 1930s as a way around crip-
pling road freight regulations). The airline was tightly held and difficult to
take over. However, by the 1980s a combination of the transport company
TNT and News Limited (Ansett owned some TV stations) managed to
take it over. The new owners were much less keen on regulation, and saw
opportunities in a deregulated environment (Forsyth, 1991). The upshot
was that the government decided to open up the domestic market quite
extensively in 1987, and implemented its new policy in 1990.
Deregulation and its Impacts
Perhaps the best way of summarizing the effects of airline deregulation is

that domestic airline policy is not an issue any more. There is no call for re-
regulation. There have been, from time to time, issues in domestic aviation,
such as the collapse of the major private carrier Ansett. Tighter regulation,
such as that embodied in the Two Airline Policy, might have been able to
avoid this, though at some cost.
Australia now has two major domestic airline groups: Qantas and
Virgin Australia. It is sometimes said that there has been little change in
the market, which has gone from two airlines to two airlines. However, this
statement misses out on the dynamic process of competition. At various
stages during the past 25 years, the number of competitors has varied
between four airlines and about one and a half (a major airline plus a small
new airline, Virgin, at the end of 2001).
As with other countries, the introduction of low-cost carriers (LCCs)
has been a significant part of the competitive process. Currently, there are

M4371-FINGER_9781786431851_t.indd 55 14/11/2017 11:14

two LCCs: Jetstar, which is owned by Qantas, and Tiger Australia, which
is owned by Virgin Australia. The first LCC entry was that of Compass,
which commenced shortly after the market was opened. This airline faced
a number of problems; some of these were managerial failures (such as the
choice of very large, difficult to fill, aircraft) and other problems included
difficulties in obtaining gates (see below) (Nyathi et al., 1993). It only
lasted a year. It was resurrected, but Compass Mark II was short-lived.
After this, it appeared that Qantas and Ansett would continue to dominate
the market, and there were no further entries until 2000.
There were two entries in 2000: Impulse and Virgin Blue (for discussion
of Australian LCCs, see Forsyth, 2003; Gross and Luck, 2013; Budd and
Ison, 2014). The latter was founded by the Virgin Group, which also has
airline investments such as Virgin Atlantic. Impulse was soon taken over
by Qantas, but Virgin grew rapidly. Qantas established an LCC subsidiary,
Jetstar, which operates both domestic and international services, and has
been one of the most successful parts of the Qantas group. It was very
well placed when Ansett collapsed in September 2001. It has continued
to grow, to the extent that it has about one third of the domestic market.
Another airline, Tiger Australia, entered the market; recently, this airline
has been taken over by (the renamed) Virgin Australia, which has also
transformed itself to become an FSC (while there have been no entries by
FSCs, it is interesting that Virgin has seen a market for an FSC in addition
to Qantas). Thus, at the moment, Australia has two FSCs, each having
an LCC subsidiary. Since 2000 there have been some smaller operators,
such as Alliance Airlines, which have concentrated on (mining) charter
operations.
There were several problems facing the earlier entrants. In the early
days post-deregulation, access to funding was not much of a problem.
However, after the failure of Compass Mark II, potential new entrants
had great difficulty in persuading capital markets; it was argued that the
domestic market had reached its equilibrium of two airlines. However, the
two entrants in 2000 did have some advantages. Impulse was based in an
established regional carrier, and Virgin Blue was a venture of the interna-
tional Virgin group.
However, the really difficult problem which was faced by the early
entrants was access to terminals and gates (Nyathi et al., 1993). When the
government deregulated the airlines it protected the incumbent airlines by
signing long-term leases of their terminals with them. This had the (inten-
tional) effect of making it difficult for new airlines to operate. For most
of the major destinations, such as Sydney and Melbourne, there were two
domestic terminals, leased by the two airlines, Australian (formerly Trans
Australia Airlines) and Ansett, and there was also a common-user inter-

M4371-FINGER_9781786431851_t.indd 56 14/11/2017 11:14

national terminal that was operated by the government-owned airport.

In order to gain terminal access and gates, the entrants had to negotiate
with the two incumbents – who had a strong incentive to reject them. In
the end, Compass was able to secure some gates from Australian (such as
in Melbourne) and to set up some temporary terminals in other cities. For
its brief life, terminal access and gates were a constant problem. For this
period, there was effective deregulation of the incumbent airlines, but not
effective entry. This meant that deregulation was incomplete.
Airport privatization and the demise of Ansett changed this scenario.
While the incumbents were still reluctant to provide access to terminals,
the owners of the airport were often quite keen to have additional custom-
ers and constructed temporary low-cost terminals, and in some cases used
the international terminals for domestic flights. Then, Ansett’s terminals
became available when it collapsed in 2001. In many cases the airports
purchased the leases for the terminals, and they became operated on a
common-user basis. In the years since, Qantas (the government merged
Australian with Qantas in 1992, and privatized the merged airline in 1994;
see Benns, 2009) has sold the leases for many of its terminals to the air-
ports, though it continues to operate them.
The greater freedom enjoyed by the airlines, and the introduction of the
LCC entrants did not make much of a difference to the networks of the
airlines (unlike in the US). This was not surprising, given that in Australia
the trunk airlines could choose their own networks. Deregulation did have
an impact on fares and their levels. Partly prompted by the LCC entrants,
but also partly of their own accord, the incumbents developed new air-
fares. As of now, the pattern of airfare offerings is similar to those which
are offered in the US and Europe. The two major airlines moved away
from a cosy, and to some extent collusive oligopoly, to some degree of
competition. Total factor productivity (TFP) grew during the 1990–2000
period, even though for much of this time there were only two airlines
(lack of data precludes accurate TFP measurement after 2000) (Forsyth,
2001).
The two airline groups are very well aware of each other’s existence, and
this conditions their behaviour. Nevertheless, there has been a price war
in recent years. This came about because Virgin decided to expand, but
Qantas drew a ‘line in the sand’ in terms of what market share it would
allow its competitor to have. Both airlines added capacity, and prices fell.
During this period, both airlines recorded losses (though in Qantas’s case
there were other factors contributing to this, such as international factors)
(O’Sullivan, 2015). Eventually, Qantas decided to allow its market share to
fall below 65 per cent, and fares rose again. Qantas has returned to profit-
ability, and losses at Virgin have fallen.

M4371-FINGER_9781786431851_t.indd 57 14/11/2017 11:14

INTERNATIONAL LIBERALIZATION: GRADUAL

AND PRAGMATIC
Australia has been a slow liberalizer on international routes. The liberaliza-

tion process has been taking place since the late 1970s / early 1980s, and it
still has some way to go. There has not been a major policy shift as there
was with the US when it adopted open skies; rather, air service agreements
with individual countries have been gradually liberalized from time to time.
As for the 1970s, international air service regulation was much the same
as for many other countries. A bilateral system of regulation was in place,
and Australia and a partner country each operated one airline. Most of
these countries opted for a government-owned flag carrier (as Australia
did). The main exception was the US, which operated with the bilateral
system, but its carriers were privately owned. The only airline Australia
permitted to operate on international routes was Qantas, which had been
operating on international routes since the 1930s. By the 1980s there were
many international airlines serving Australia, from Europe, Asia, North
America, New Zealand and the Pacific Nations, and a few from Africa and
South America.
The industry has changed somewhat since then. In 1992 the govern-
ment’s domestic airline, by then called Australian Airlines, was merged into
Qantas, and the merged airline was privatized in 1994. While Qantas has
been the dominant carrier, it has been joined by other Australian airlines.
In the 1990s Ansett began to operate international services, mainly to Asia,
including Japan. With the demise of Ansett, there was no major Australian
airline serving international routes. Virgin began to operate services from
some Pacific Nations, though these were operated as subsidiaries of the
Australian company. However, by 2009 Virgin began to operate its own
flights (using Boeing 777s) to the US, and other destinations followed,
including Singapore and Abu Dhabi. Qantas remains by far the largest
carrier on international routes. In the past ten years, the Qantas-owned
LCC Jetstar has entered the market. Its services are very much coordinated
with those of its parent; for example, it serves tourism destinations such as
Bali, as well as routes which are marginal for Qantas.
The pattern of foreign airlines has changed significantly. British Airways
is the only European airline now flying to Australia (and Australian airlines
now have only two flights per day into Europe, though this will increase
to three when a non-stop Perth–London service is operated from 2017).
Gradually, the major European airlines, such as KLM, Lufthansa and Air
France, and the smaller airlines such as Swissair and finally Austrian have
ceased service. The Australia–Europe service is almost totally dominated
by the sixth freedom carriers. The Gulf carriers have become a major force

M4371-FINGER_9781786431851_t.indd 58 14/11/2017 11:14

in Australia–Europe and Australia–Asia air transport. Asian airlines have

a major role, and the China market is now very large and growing rapidly.
Traffic to and from North America is growing steadily, and there are four US
carriers and one Canadian carrier. Traffic from New Zealand continues to
grow, and there still remain some carriers from Africa and South America.
The Process of Liberalization
Up until around 1990, Australia imposed quite detailed regulation on inter-

national airlines flying to Australia (ICAP review, 1978; Findlay, 1985).
Normally there was only one carrier per country, but this was eased around
the 1990s. In most cases, fares were regulated – Australia and the partner
country would set fares used on IATA recommendations. Over time, this
fare regulation was dropped, country by country (this did not apply to
US routes). There have been occasional reviews of international aviation
policy. An early and important one was the 1978 ICAP (International Civil
Aviation Policy) review, which investigated the role of sixth freedom car-
riers. Further reviews include the 1988 Review (Department of Transport
and Communications, 1988) which took some steps towards liberalization,
and the generally liberal Productivity Commission review (Productivity
Commission, 1998).
An interesting aspect of Australian policy evaluation is that it took very
much an economic or ‘cost–benefit’ approach to determine whether liber-
alization, on specific routes, was in Australia’s interest (in the sense that it
was welfare improving). The approach was to determine whether the ben-
efits, in terms of consumers’ surplus gains, outweighed the costs in terms
of producers’ surplus losses. This approach was used even in the early
ICAP report of 1978, then continued in the 1988 report and in the Bureau
of Transport and Communications Economics study of the Australia–
New Zealand Single Aviation Market (1991) and reached its zenith in the
Productivity Commission report of 1998 (for details of the approach, see
Gregan and Johnson, 1999; see also Street et al., 1994). This approach is
very consistent with a pragmatic approach to liberalization, and it is pre-
pared to accept that in some cases liberalization is not always in Australia’s
interest (in contrast with the US approach, which seeks open skies with all
countries). The cost–benefit approach has been used in other countries, for
example, in assessing the costs and benefits of US airline domestic deregu-
lation (Morrison and Winston, 1986): the early ICAP 1978 study predates
this widely quoted study. There seems to have been little (published) use of
the cost–benefit by the Australian government’s approach to assessment
since the Productivity Commission report, though there is some interest in
using computable general equilibrium (CGE) modelling (IAC, 1989).

M4371-FINGER_9781786431851_t.indd 59 14/11/2017 11:14

There are, however, two types of regulation which Australia still tends
to enforce: one of these concerns gateways, and the other is about capac-
ity. If a country seeks to establish air services with Australia, the default
position is that it may do so with no capacity controls except for flights to
Sydney, Melbourne, Brisbane and Perth. This is to encourage flights to
smaller centres, such as Adelaide, Cairns, the Gold Coast and Canberra.
If the country seeks permission for its airlines to access the four major
centres, Australia’s normal position would be that it would set capacity
controls (the ultimate outcome will depend on the bargaining power of the
country). Capacity would be set such that it is sufficient to handle expected
demand. As demand grows, additional capacity can be negotiated. There
are few flights which go to cities other than the big four.
On the face of it, this seems to be a fairly liberal arrangement. However,
there are some problems. For a start, the system relies on the government
regulators being able to forecast demand ahead; sometimes demand gets
ahead of capacity (for example, from time to time in the rapidly growing
China market). Secondly, and less obviously, this system can still result in
a degree of restriction which pushes fares above what they would be in a
competitive market – it is one thing to adjust capacity such that it is suf-
ficient to meet demand, but at what price? A good question would be ‘why
is it necessary to regulate capacity at all?’ Other airline markets in Australia
and elsewhere do not need this regulation and yet they perform well.
However, this regulation does have the effect of affording the Australian
carriers some degree of protection.
Given that there is limited capacity for the Australian airlines which
would like to service a specific route, an allocation problem arises. For
example, both Qantas and Virgin may wish to serve a route (either with
their own equipment or on a code-share basis), which could be potentially
profitable. To solve this problem, the government has set up an arm’s-length
body, the International Air Services Commission (IASC) (see Findlay and
Round, 1997), to allocate routes. Since this may pose some competition
policy issues, this body works closely with the main competition regulator,
the ACCC (Australian Competition and Consumer Commission).
The result of this incomplete deregulation is that while some routes
are very competitive, others are not, and airfares are moderately high.
The Europe route is very competitive; there is a wide variety in quality of
service and fares, and overall fares are very competitive. On some of the
less dense routes, such as those from Africa and South America, fares are
somewhat higher. Airfares depend on the tightness of regulation and on
competitive conditions (Forsyth and Seetaram, 2014). The US route is
not greatly regulated, though fares have been high sometimes (see below).
Currently they are low, reflecting the fact that there are several airlines

M4371-FINGER_9781786431851_t.indd 60 14/11/2017 11:14

competing. However, ten years ago, with Qantas, United and Air New
Zealand accounting for most of the traffic, they were moderately high.
Like most airlines, the Australian international airlines have used alli-
ances to further their objectives. Qantas has been in alliances since the
1930s, when it had an alliance with Imperial Airways (forerunner of British
Airways) to operate flying boat services to the UK. It is a member of the
oneworld alliance, and as is common, it has alliances with individual air-
lines. For many years it had a joint venture (involving pricing and sharing
of revenues) with British Airways (this joint venture was approved by the
UK and Australian competition authorities). However, in 2013 it termi-
nated this arrangement in favour of an alliance with Emirates (which had a
very convenient hub for Australia–Europe traffic) (O’Sullivan, 2015); this
too was approved by the competition authorities. However, this was not
the case when it sought to develop an alliance with Air New Zealand, as it
was not able to obtain approval from competition authorities, which feared
lack of competition in the Trans Tasman market (O’Sullivan, 2015).
Virgin Australia has a small international operation and relies very much
on its alliances with larger carriers; it has an extensive Australian network,
but only operates a small number of flights to a range of international des-
tinations. As a result it is able to offer a wide variety of destinations. Thus,
it offers flights to Singapore with one of its owners, Singapore Airlines,
and New Zealand, with one of its former owners, Air New Zealand.
However, potentially its most important alliance is with Etihad, which is
creating an alliance of its own.
Given that issues and responses depend very much on routes and times,
a good way of analysing the issues is by brief case studies. Here five are
identified:
●● Europe and the sixth freedom issue;

●● the slow liberalization of the Japan route;
●● the Transpacific route and Singapore Airlines;
●● the Australia–New Zealand single aviation market; and
●● adaptation to Asia.
Europe and the Sixth Freedom Issue
Until the 1970s, sixth freedom was not an issue for Australia or, for that
matter, most other countries. Qantas had flights to Europe (especially the
UK), and a range of European airlines had flights to Australia. Asian
airlines were in their infancy – they had few flights and were being given
substantial technical and financial help by the developed country carriers.
They were not in the position of competing on sixth-freedom markets.

M4371-FINGER_9781786431851_t.indd 61 14/11/2017 11:14

This changed quickly over the 1970s. The airlines of several Asian
countries started services to Europe and to Australia. Initially they were
not seen as being very competitive, although this changed over time. The
capacity which they were able to devote to Australian routes was limited
by regulation. Some did offer flights to Europe, selling the flight with two
separate tickets; for example, one might have one ticket from Sydney to
Singapore, and another from Singapore to London. In addition, for a time,
ship/jet charters became popular: one would take a ship from Australia to
Singapore, and there board a flight to London. This was worthwhile, given
the high (regulated) fares of the Australia–Europe flights. The British LCC
Laker Airways repeatedly sought to enter the Australia–UK market, prom-
ising significantly lower fares, but it was never permitted to enter. Both the
government and the flag carrier were very concerned about ‘malpractices’
such as fare discounting, and the carrier was losing market share. The issue
of airfares became a live political issue.
The government set up an inquiry, the ICAP (International Civil
Aviation Policy) (Findlay, 1985; ICAP Report, 1978) committee, which
reported in 1978, and following this, the government changed its policy in
1979. The policy was one of more tight regulation. Some new fares were
introduced, such as advance booking fares, though stopovers were not
permitted on these. The intention was to concentrate traffic on hubs such
as Sydney to try to increase load factors. Capacity controls were tightened
with the intention that the Asian carriers would only have enough capacity
to serve their own market; that is, there was a conscious attempt to freeze
out sixth freedom traffic.
The Asian countries, led by Singapore, were very much opposed to this
policy, since they lost airline revenue and tourism revenue from stopovers.
The ASEAN (Association of South East Asian) countries were able to
negotiate as a bloc, and they succeeded in gaining some modifications
to the policy (Findlay, 1985). For a time, perhaps two or three years,
Qantas did well out of the policy. However, it was not really possible to
limit capacity of the Asian airlines to simply that warranted by third and
fourth freedom traffic. Essentially, the policy broke down, and discount-
ing became common again by the mid-1980s. The Australian experience is
relevant to the current debate in the US, and to a lesser extent in Europe,
about the impact of the Gulf carriers on US and European airlines. There
have been several suggestions that the Gulf carriers should be limited to
carrying only third and fourth freedom traffic, and that they should not
be allowed enough capacity to carry sixth freedom traffic (Partnership
for Fair & Open Skies, 2015). If the Australian experience is any guide, it
would be very difficult to police this (as it was not feasible even in a world
where regulation was the norm).

M4371-FINGER_9781786431851_t.indd 62 14/11/2017 11:14

Initially, it was the Asian carriers that developed sixth freedom traffic
from Australia. More recently, it has been the Gulf carriers which have
become the pacemakers, and Emirates now is one of the two main carri-
ers on the Australia–Europe route (the other being Singapore Airlines).
Australia still controls capacity, but it has chosen to grant adequate
capacity for them to expand rapidly, even though these carriers are almost
entirely carrying sixth freedom traffic. Qantas was particularly concerned
about the growth of these carriers on Australian routes. However, the situ-
ation has changed, with the two Australian carriers forming close alliances
with Gulf carriers: Qantas with Emirates, and Virgin with Etihad.
The Slow Liberalization of the Japan Route
Of the Asian routes, historically the most important has been the Japan
route, though its importance has been falling recently with a decline in
Japanese tourism to Australia. This route has also been the most tightly
regulated one. Airfares were very high from the time that Japanese tourism
started growing in the 1970s. There has been some liberalization, but it was
not until around ten years ago that the fares on the route fell to something
like those on other routes. The Japanese government preferred tight regula-
tion on this and all other international routes (Yamauchi, 1997). There was
some sixth freedom traffic, and for a time, Northwest Airlines from the US
operated on a fifth freedom basis, though without success.
A significant feature of this route is that the tourism industry first
became a major stakeholder in the international aviation market. The
tourism aspects of the Europe route were not discussed much: most of the
discussion was in terms of the benefits to the home consumer versus the
airline profit effects (though the ICAP report estimated both home and
foreign consumer surpluses in its evaluations). The Japan route was pri-
marily a tourism route, and the tourism industry developed a voice. Since
then, governments have been seeking a balance between airline industry
interests, consumer interests and now tourism industry interests in most
air transport discussions.
The route is an interesting one, since the two end-countries were acting
against their own best interests. By far most of the traffic was Japanese,
and the airline capacity was shared equally by airlines of the two countries.
This meant that the high fares were paid by Japanese nationals, and the
airline revenues were shared between Japanese and Australian airlines.
Since for much of the time the Japanese airlines had very high costs, the
profits from the route accruing to them were not large. This meant that
Japan was paying a very high price to protect its airlines.
By contrast, Australia kept pressing the Japanese to liberalize. This

M4371-FINGER_9781786431851_t.indd 63 14/11/2017 11:14

would have the effect of reducing profits of the Australian airlines.

Australia would gain very little by way of consumer benefits, though it
would gain tourism revenues. The benefits from increased tourism would
have to be high to outweigh the substantial loss of profits (for many years,
the Japan route was Qantas’s most profitable route).
The North Transpacific Route and Singapore Airlines
One of the important routes is that to North America. Currently it is quite

competitive, though at times in the past there has been little competition.
In the past, several US carriers have tried their fortunes on the route, and
left. The profitability of the route has depended more on competitive
dynamics than on regulation. For many years, Australia has been reluctant
to go the way of full open skies; however, pressure from the US has ensured
that the route has been fairly open.
Currently, both Qantas and Virgin operate on the route, and US carri-
ers are present alongside Air Canada and Air New Zealand. However, one
airline that has been very keen to enter has been Singapore Airlines – this
was particularly the case around 2005, at a time when the competition was
weak (with only one US carrier apart from Hawaiian, which did not offer
direct services to mainland US) and profits were high. Singapore Airlines
sought to operate as effectively a seventh freedom route (passengers from
Singapore would not normally fly to Los Angeles via Sydney). The US
approved this plan, since it was consistent with its treaty obligations with
Singapore. The single Australian carrier, Qantas, was very much opposed
to this, as it would have resulted in loss of market share and profits.
Meanwhile, the tourism industry supported Singapore Airlines. Given
that airfares on the route were high and there was public pressure for
more competition, the Australian government reviewed the proposal, but
rejected it partly on the grounds that there were better ways of increasing
competition. It encouraged Virgin to commence the route, which it did in
2009. Since then, Singapore Airlines has continued to express its interest in
flying the route, though it has not pressed strongly to be allowed to do so.
The Australia–New Zealand Single Aviation Market
As might be expected, Australia and New Zealand have close relations,

both economic and cultural. In 1996 the two countries concluded an agree-
ment for a Single Aviation Market (SAM) (InterVISTAS, 2015; see also
Bureau of Transport and Communications Economics, 1991). This agree-
ment made it possible for airlines of both countries to operate domestic
services in the other country and it allowed unlimited frequencies between

M4371-FINGER_9781786431851_t.indd 64 14/11/2017 11:14

the two countries for their airlines and liberalized and harmonized owner-
ship arrangements. In addition, the two countries introduced an open skies
agreement which commenced in 2002. This made it possible, for example,
for an Australian airline to fly to New Zealand and then pick up passengers
and freight to the US.
The SAM has resulted in the effective integration of the two aviation
markets. So far, Air New Zealand has not entered the Australian domes-
tic market, though it has considered doing so. On the other hand, the
Australian airlines have been active in the New Zealand domestic market:
both Qantas and Virgin Australia operate services through subsidiaries.
Qantas flights from Australia to New Zealand are operated by a New
Zealand subsidiary, using both New Zealand-registered aircraft and New
Zealand crews (costs in New Zealand are lower than in Australia). In the
past, Air New Zealand held a major shareholding in Ansett, but the Ansett
collapse has made it wary of Australia. For a period it held a share of
Virgin Australia, though it disposed of this in 2016. Qantas held a share-
holding in Air New Zealand for some time: it was planning an alliance and
joint venture (one which would cross alliance boundaries, since Qantas
is in oneworld and Air New Zealand is in Star Alliance). However, the
joint venture was rejected by competition authorities, and Qantas sold its
holding soon after.
Adaptation to Asia
Australia is close to Asia, and the aviation links to Asian countries are
important. One critical problem is that Australia’s costs are much higher
than the costs of most Asian countries, excluding Singapore and Japan
(Oum et al., 2000). This has made it difficult for Australia’s airlines to
compete. As a result, the Australian airlines, especially Qantas, have been
at the forefront of internationalizing their labour forces. Thus, Qantas
makes considerable use of foreign crews on international flights; it also
relies heavily on maintenance done offshore. It has established subsidiar-
ies, such as Jetstar Asia, based in Singapore so as to access Asian markets.
Most of these ventures have been successful. In spite of this, the Australian
airlines have not been able to gain a large share of the vast and rapidly
growing China market.
Australia lies in a region in which its airlines are not particularly cost-
competitive, at least in some key routes, such as those to Europe and Asia
(Oum et al., 2000). Most airlines in the Asian region (with the exception of
Singapore and Japan) and the Gulf countries face lower labour costs, and
if they are productive, can achieve higher cost competitiveness than the
Australian airlines. The airlines are cost competitive on North American

M4371-FINGER_9781786431851_t.indd 65 14/11/2017 11:14

routes, but even with the New Zealand route, there has been pressure over
the years – since the 1970s – from consumer interests, and more recently
the tourism industry, for lower fares. There has been a succession of cases
where foreign airlines have been offering lower fares if they are permitted
to enter the routes. Governments have been prepared to allow these airlines
access, though rarely on an open skies basis. This has had implications for
the Australian airlines, and their market share has gradually fallen.
CONCLUSIONS
Australia has been a reluctant liberalizer of both domestic and interna-

tional air transport. It took some time after US deregulation before it
deregulated its domestic airlines, and when it did so it made new entry
quite difficult. It was not until the privatization of the airports, and the col-
lapse of Ansett, that new entrants were able to establish themselves, since
the incumbent airlines had control over the terminals. The government was
prepared to accede to consumers’ wishes for more competition and lower
air fares, but only to the extent that incumbent airlines were not harmed
too much.
Liberalization on international routes has been a gradual process since
the 1970s. Governments have been pragmatic, and at times, have analysed
the benefits and costs of liberalization very carefully. They have been trying
to balance the different interests of the industry, consumers/shippers, and
in more recent times, the inbound tourism industry, rather than adopt-
ing a systemically pro-liberalization stance, as the US has done with its
emphasis on open skies. At various stages, governments have sought to
limit liberalization, as they did in the 1970s, and more recently, in denying
Singapore Airlines the right to fly to the US. At other times, they have been
willing to allow liberalization, as they have done with the Japan route and
the Australia–New Zealand single aviation market. The Australian airlines
operate in a low-cost region: Asia. Nonetheless, Australian international
routes have become, in the main, quite liberal. One measure of the will-
ingness to accept liberalization is the fact that policies have been adopted
which have had a significant decreasing effect on the market share of the
Australian airlines in the past decades.
REFERENCES
Bailey, B., D. Graham and D. Kaplan (1985), Deregulating the Airlines, Cambridge,
MA: MIT Press.

M4371-FINGER_9781786431851_t.indd 66 14/11/2017 11:14

Benns, M. (2009), The Men Who Killed Qantas, North Sydney: Heinemann
Australia.
Brogden, S. (1968), Australia’s Two Airline Policy, Carlton: Melbourne University
Press.
Budd, L. and S. Ison (eds) (2014), Low Cost Carriers: Emergence, Expansion and
Evolution, Farnham: Ashgate.
Bureau of Transport and Communications Economics (1991), Costs and Benefits
of a Single Australasian Aviation Market, Canberra: Australian Government
Publishing Service.
Department of Transport and Communications (Australia) (1988), Negotiating
International Aviation Rights, Canberra: Department of Transport and
Communications.
Findlay, C. (1985), The Flying Kangaroo: an Endangered Species? An Economic
Perspective of Australian International Civil Aviation Policy, Sydney: Allen and
Unwin.
Findlay, C. and D. Round (1997), ‘Multiple designation policy in Australia’, in
C. Findlay, Chia Lin Sien and Karmjit Singh (eds), Asia Pacific Air Transport
Challenges and Policy Reforms, Singapore: Institute for Southeast Asian Studies,
pp. 84–105.
Forsyth, P. (1991), ‘The regulation and deregulation of Australia’s domestic airline
industry’, in K. Button (ed.), Airline Deregulation: International Experiences
London, London: David Fulton Publishers, pp. 48–84.
Forsyth, P. (2001), ‘Total factor productivity in Australian domestic aviation’,
Transport Policy, 8(3), 201–207.
Forsyth, P. (2002), ‘Privatisation and regulation of Australian and New Zealand
airports’, Journal of Air Transport Management, 8, 19–28.
Forsyth, P. (2003), ‘Low cost carriers in Australia: experiences and impacts’,
Journal of Air Transport Management, 9, 277–84.
Forsyth, P. and R. Hocking (1980), ‘Economic efficiency and the regulation of
air transport’, Monograph M62, Committee for the Economic Development of
Australia.
Forsyth, P. and N. Seetaram (2014), ‘The future of Australian international avia-
tion: liberalisation, competition and the Dutch disease’, in D. Duval (ed.), Air
Transport in the Asia Pacific, Farnham: Ashgate, pp. 143–56.
Gannon, C. (1979), ‘Parallel scheduling’, Appendix 12 Department of Transport,
Domestic Air Transport Policy Review, Appendices, Canberra: Australian
Government Policy Review.
Gregan, T. and M. Johnson (1999), Impacts of Competition Enhancing Air Services
Agreements: A Network Modelling Approach, Productivity Commission, Staff
Research Paper, Canberra: AusInfo.
Gross, S. and M. Luck (eds) (2013), The Low Cost Carrier Worldwide, Farnham:
Ashgate.
Hocking, R. (1972), ‘Some economic aspects of Australia’s Two Airline Policy’,
Monograph M35, Committee for the Economic Development of Australia.
Holcroft, W. (Chairman) (1981), Domestic Air Fares: Report of the Independent
Public Inquiry, Canberra: Australian Government Publishing Service.
ICAP (1978), ‘Review of Australia’s International Civil Aviation Policy (ICAP):
Report of Review Committee’, Canberra: Australian Government Publishing
Service.

M4371-FINGER_9781786431851_t.indd 67 14/11/2017 11:14

Industries Assistance Commission (Australia) (IAC) (1989), Travel and Tourism,

Report No 423, Canberra: Australian Government Publishing Service.
InterVISTAS (2015), ‘The economic impacts of air service liberalization: updating
the Landmark 2006 Study to Reflect the New Realities of Commercial Passenger
Aviation’, Washington and Vancouver: InterVISTAS.
Mackay, K. (1979), ‘A comparison of the relative efficiency of Australian domestic
airlines and foreign airlines’, Appendix A6.1, Department of Transport, Domestic
Air Transport Policy Review, Appendices, Canberra: Australian Government
Publishing Service.
May, T. (Chairman) (1986), Independent Review of Economic Regulation of
Domestic Aviation, December, Canberra: Australian Government Publishing
Service.
Morrison, S. and C. Winston (1986), The Economic Effects of Airline Deregulation,
Washington: Brookings Institution.
Nyathi, M., P. Hooper and D. Hensher (1993), ‘Compass Airlines: 1 December
1990 to 20 December 1991: what went wrong?’ Parts I and II, Transport Reviews,
13, 119–49 and 185–206.
O’Sullivan, M. (2015), Mayday: How Warring Egos Forced Qantas off Course,
Melbourne: Viking Books.
Oum, T.H., C. Yu and M. Li (2000), ‘Modeling performance: measuring and com-
paring unit cost competitiveness in airlines’, in D. Hensher and K. Button (eds),
Handbook of Transport Modelling, Oxford: Pergamon, pp. 609–24.
Partnership for Fair & Open Skies (2015), ‘Restoring open skies: the need to
address subsidized competition from state-owned airlines in Qatar and the
UAE’, available at http://skift.com/wp-content/uploads/2015/03/White.Paper-2.
pdf.
Productivity Commission (1998), International Air Services, Inquiry Report No. 2,
Canberra: AusInfo.
Starkie, D. and M. Starrs (1984), ‘Contestability and sustainability in regional
airline markets’, Economic Record, 60(3), pp. 274–83.
Street, J., J. Smith and S. Savage (1994), ‘An analysis of the trade-offs in interna-
tional aviation rights’, Australasian Transport Research Forum, 19, Transport
Research Centre, University of Melbourne, pp. 183–206.
Yamauchi, H. (1997), ‘Air transport policy in Japan: limited competition under
regulation’, in C. Findlay, C. Lin Sien and K. Singh (eds), Asia Pacific Air
Transport Challenges and Policy Reforms, Singapore: Institute of Southeast
Asian Studies, pp. 106–22.

M4371-FINGER_9781786431851_t.indd 68 14/11/2017 11:14

5. Air transport liberalization: the case
of Ireland
Sean Barrett
INTRODUCTION
Ireland played a leading role in the liberalization of European aviation.

It is the home of the EU’s largest airline in terms of passenger numbers
(Ryanair); its bilateral deregulation of the Ireland–UK market in 1985 gave
Ireland an early experience of liberalization in advance of the completion
of EU liberalization; and its legacy airline, Aer Lingus, successfully phased
in liberalization. As an outer offshore island, aviation is important to
Ireland; the Irish Competition Commissioner, Peter Sutherland, promoted
liberalization in European aviation, while countries with land frontiers
gave this issue a lower priority. Island people need strong airline services.
Ireland enjoyed a first-mover advantage in European aviation in 1985
when Aer Lingus, the State airline, was exposed to competition for the first
time from a rival Irish airline, Ryanair, on the Dublin–London route. The
liberalization of the aviation market in the rest of the European Union
was not completed until 1 April 1997. As a broad indicator of the effects,
Table 5.1 contrasts the low output of the sole Irish airline in 1985 (Aer
Lingus) with Ryanair and other carriers that subsequently entered the
market.
THE EMERGENCE OF RYANAIR
The argument for liberalization in 1985 across Europe was that the closed
market access system of non-competing national airlines would produce
high fares, low productivity and a smaller overall market output and that
new market entry was required. The Irish case study vindicates the advo-
cates of liberalization. Ryanair is the largest new entrant to the liberalized
European market. The 2015 passenger numbers for Ryanair (106 million)
exceed those of traditional national airlines such as the International
Consolidated Airlines Group (IAG), based on British Airways and Iberia
69

M4371-FINGER_9781786431851_t.indd 69 14/11/2017 11:14

Table 5.1 The Irish airline sector prior to liberalization in 1985 and in

2016
Year Airline Passengers (million) Planes Staff

1985 Aer Lingus 2.3 23 6500
2016 Ryanair 117.0 350 11 500
Aer Lingus 10.0 48 3615
Cityjet 2.1 27 800
Stobart 1.4 17 370
Total 2016 130.5 442 16 285
Source: Airline websites.
at 88 million, and Air France/KLM and Lufthansa (both at 79 million).

The second-largest new entrant airline, easyJet, had 69 million passengers.
The two leading new entrants, Ryanair and easyJet, had 216 million pas-
sengers in 2015. Thus, liberalization of market access in 1985 on Ireland/
UK routes and in 1997 on the remainder of EU routes facilitated new
market entry that was widely availed of by consumers to make choices that
had not been available before liberalization.
The impact of liberalization on Aer Lingus, the previously protected
national airline, was a 4.3-fold increase in passenger numbers and a
remarkable increase in productivity, from 354 passengers per staff member
per year in 1985 to 2740 in 2016. Thus, liberalization promoted the growth
of Aer Lingus significantly beyond its output when policy was to protect
it from competition.
Combined, Ryanair and Aer Lingus’s passenger numbers constituted
over 97 per cent of passengers on Irish airlines in 2016. Therefore, the
impact of liberalization on these airlines dominates debate on liberaliza-
tion in Ireland. For Ryanair, liberalization brought the opportunity to
enter markets from which the airline had previously been excluded.
Ryanair’s success required liberalization between Ireland and the UK and,
subsequently, with the rest of the EU. In 2016, Ireland only had two inter-
nal air routes, which means that Ryanair could not have succeeded without
international liberalization.
The impact of new entrants on aviation market efficiency in the US
and in Europe has been examined in several case studies (e.g. European
Economic Community, 1981; International Civil Aviation Organization,
1990; Jordan, 1970). These studies found that lower air fares pertained
within the state of California than on regulated interstate routes in the US
and that a similar situation existed regarding charter airlines in Europe

M4371-FINGER_9781786431851_t.indd 70 14/11/2017 11:14

The case of Ireland 71
Table 5.2 British Airways cascade analysis
Cascade Steps Route A Route B Route C

Scheduled cost per passenger 100 100 100
1. Deduct commission 92 92 91
2. Tourist class 86 84 87
3. Seating density 77 80 83
4. Local factor 56 59 60
5. Peak–trough ratio 60 62 63
6. Utilization 57 60 61
7. Standards 51 51 53
8. Not applicable 36 37 39
Derived charter 36 37 39
Actual charter 34–37 32 35
Source: European Economic Community (1981).
when compared to legacy scheduled airlines operating in the same geo-

graphical areas. In 1980, charter airline fares in the European Union were
between 32 and 37 per cent of scheduled airline fares.
The cost differences between aviation in Europe and North America
were examined by the UK Civil Aviation Authority (1983) and the
Association of European Airlines (1984). The former showed that, in 1981,
five cost factors accounted for almost four-fifths of excess costs in Europe.
These were sales costs (22.9 per cent), route and landing charges (18.3 per
cent), station and ground (13.8 per cent), fuel (13.4 per cent) and crew
(11.2 per cent). The AEA estimated that ticketing, sales and promotion
accounted for 26.8 per cent of excess costs in Europe; landing and en route
charges 24.2 per cent; station and ground 12.3 per cent; cabin services 9.7
per cent; and maintenance and overhaul 9 per cent.
Table 5.2 shows the results of a European Commission cascade analysis
of British Airways’ costs that can be compared to charter airlines. The
latter did not typically sell through travel agents and thus saved 8 per cent
of the ticket fare; by operating tourist class only, the cost of the ticket is
reduced by a further 8 per cent. In terms of load factors, taking Route B,
charters filled 85 per cent of their seats, compared to 65 per cent for sched-
uled airlines.
The potential fare reductions contrasted with a 72.6 per cent increase
in fares on the London–Dublin route between 1980 and 1985, compared
to 43.7 per cent over a sample of 11 short-haul routes including Dublin–
Paris, Amsterdam, Frankfurt, Brussels and six UK cities. Although traffic
between Dublin and London fell by 2.8 per cent over five years, the overall

M4371-FINGER_9781786431851_t.indd 71 14/11/2017 11:14

Table 5.3 Passenger growth and charter share on main European routes

from UK (1980–85)
Growth 1980–85 (per cent) Charter Share 1985 (per cent)

Portugal 120 78
Greece 56 81
Switzerland 39 28
Spain 38 82
France 21 12
Netherlands 17 2
Germany 16 18
Ireland −3 1
Italy −5 51
Source: UK Civil Aviation Authority, Annual Reports.
growth for 11 routes examined by the UK Civil Aviation Authority (1987)

was 29.9 per cent. Table 5.3 shows that charter airlines featured in high-
growth markets from the UK between 1980 and 1985, confirming their
importance as drivers in the industry.
The pre-liberalization performances were uncompetitive compared to
airlines in the other major world markets, North America and Asia-
Pacific, and compared to charter airlines operating in Europe. Europe’s
national airlines secured protection from possible competitors by captur-
ing their regulatory authorities à la Stigler (1971). This made such airlines
complacent about productivity, their products, and their prices.1 The
success of the airlines in dominating consumer interests in the short run
diverted entrepreneurial effort in the industry away from serving passen-
gers and increasing productivity in the airlines, towards rent-seeking activi-
ties in the hope of protecting legacy airlines from new market entrants and
the unlikely prospect of competition from other incumbent airlines.2
Ireland became independent in 1922. Aer Lingus, established in 1936,
was identified as a symbol of the new state. The airline came to symbolize
entrepreneurship, promote access to an outer offshore island, and develop
Ireland as a tourist destination as well as providing direct employment in
the airline itself. State enterprises would become development corpora-
tions in contrast to the conservative private sector in the new state. Aer
Lingus subsequently enjoyed repeated success in its rent-seeking and
regulatory capture of its parent department of state. The protection of
a low-productivity, high-cost national airline delayed competitive access
until after liberalization.

M4371-FINGER_9781786431851_t.indd 72 14/11/2017 11:14

The International Air Transport Association (IATA), the trade associa-

tion for 89 active and 20 associate airlines, defended the ‘three inter-related
cornerstones’ for ‘orderly air transport development’. Market entry had to
be controlled ‘to prevent speculative creaming of traffic by those operating
outside orderly agreements’. Capacity had to be harmonized by incumbent
airlines ‘to prevent the waste of resources but without rigid restrictions’.
Finally, tariffs were ‘to be developed by multilateral airline negotiation
subject to government approval’ ‘to avoid predatory pricing on competitive
routes so closely inter-related’ (International Air Transport Association,
1977). In the late 1970s the IATA added that ‘for twenty years this system
has served the world well’. In retrospect, protectionism in aviation was
already being questioned because of the high-cost airlines it shielded from
competition.
Several obstacles had to be overcome before liberalization could take
place. Doganis (2001) summarized several of these when discussing the
‘distressed state airline syndrome’ in European aviation with the following
elements: substantial losses, over-politicization, strong unions, overstaff-
ing, no clear development strategy, bureaucratic management, and poor
service quality.
Regulatory capture in Ireland was undermined by a parliamentary
revolt led by Desmond O’Malley. The Air Transport Bill, 1984, was intro-
duced in the Irish parliament because the Supreme Court had lifted an
injunction restraining Transamerica from selling unapproved airfares that
were lower than those approved by the minister. ‘Discounting and other
malpractices could take place on a scale that would undermine approved
tariff structures and could have serious implications for airlines generally
and for Aer Lingus in particular’ (Barrett, 2009).
The Minister for Transport had been in Japan when the Bill was intro-
duced by his junior minister. He sought to dilute opposition to it by
awarding Ryanair licences to UK routes with the agreement of the UK
government because they needed bilateral agreement to increase competi-
tion on EU air routes.
Ryanair commenced operations in 1985 and in May 1986 entered the
Dublin–London (Luton) route. The unrestricted Dublin–London fare
fell from £208 to £95 return on the first day of competition. In the first
full year of liberalized operations on the route, there was a 64.9 per
cent growth in passengers compared to 1985, the last full year of pre-
liberalization policies. This contrasted with a growth of only 2.8 per cent
over the five years before liberalization. The growth rate was highest in
August 1987, at 91.7 per cent, indicating the attractiveness of the low fares
in the holiday season.
By 1993, Ryanair had four routes serving Dublin and Knock in Ireland

M4371-FINGER_9781786431851_t.indd 73 14/11/2017 11:14

and Liverpool, Stansted and Luton in the UK and exceeded a million

passengers; by 1998, there were 5 million passengers on 26 Ryanair routes
including eight from Stansted to mainland Europe and two from Dublin
to mainland Europe. Seven million passengers were carried in 2000, 51
million in 2007, over 100 million in 2015 and, as we have seen in Table 5.1,
117 million in 2016. The network in 2016 has 84 base airports and serves
200 airports on 1800 routes with an average projected fare of €40.
Airline efficiency was part of the outcome of the emergence of the low-
cost carrier. By 2016, Ryanair had 33 staff per aircraft, a reduction from
58 in 1997 when the European Union market was liberalized. The average
number of staff per aircraft in the previously protected national airlines
of the Association of European Airlines was 116, double that of Ryanair.
Measured another way, there are 10 174 passengers per staff member on
Ryanair in 2016. This compares with 4377 in 1997 when the European
Union market was liberalized and when the average for 14 AEA members
was 752 passengers per staff member. In sum, removal of the regulatory
capture through which incumbent airlines could veto new market entrants
has allowed market entry by those previously outside the market, large fare
reductions and productivity increases, large increases in airline output and
employment and a large increase in the number of points served.
The success of the Irish carriers was also the result of complacency
among the protected airlines in Europe. For example, the cascade studies
cited above show that the scheduled airlines could have produced a
‘derived charter’ product at a price comparable to the actual charter fare
– that is, between 32 and 37 per cent of the scheduled legacy airline fare –
but chose not to do so. Barrett (2009) cited a Chamber of Commerce of
Ireland survey that found over 90 per cent of the chief executives of the top
thousand Irish companies regarded business class as bad value for money.
Others following the Ryanair model also benefited from this and, in 2015,
the four largest low-cost airlines in Europe – Ryanair, easyJet, Norwegian
Air and Air Berlin – had 250 million passengers.
Ryanair introduced several product changes. For example, it deleted
many customer service items such as sweets, newspapers, free food and
beverages, seat allocation and business class, allowing more seats per air-
craft and increasing load factors. It made use of secondary airports, some
that had no previous scheduled services, and had no interlining or connec-
tion journey tickets. Passengers and baggage had to be checked in at each
airport on a multi-stage journey. It ceased providing an airport business
lounge service and tickets were no longer sold through travel agents or
airline ticket sales offices. There was no frequent flyer programme and
there were strict penalties for ‘no-show’ passengers.
While significant fare savings have been the main consumer gain from

M4371-FINGER_9781786431851_t.indd 74 14/11/2017 11:14

deregulation of European aviation, there have also been service improve-

ments, such as avoiding congested hub airports, improved punctuality
at uncongested secondary airports and reduced walking distances and
baggage waiting times and fewer bags lost because of the simple point-to-
point product. There is also no longer a risk of being denied boarding on
overbooked flights, compared to traditional airlines whose overbooking
policy imposed the cost of no-show passengers on overbooked passengers
rather than on the no-shows.3
Some of these developments were as much the result of needing to
circumvent existing restrictions as from the business model per se. For
example, hub slot-constrained airport capacity is allocated by incumbent
airlines in order of seniority, meaning that new entrant airlines were
compelled to seek new airports to enter the market. This has had positive
results, such as reduced time spent at airports per journey, fewer lost bags
and improved punctuality because planes at uncongested airports encoun-
ter fewer delays.
Low-cost airlines like Ryanair quickly achieved considerable cost savings
by using the internet rather than travel agents. A study by the Association
of European Airlines found that ticketing sales and promotions costs were
the largest single component of European airline operating costs in 1983,
costs that were 2.3 times those of US airlines. The savings generated by
the low-cost carriers were achieved by substituting internet sales for sales
through travel agents and expensive high-rent airline sales office locations.
Non-price competition required promotional budgets for products sold
at identical fares by legacy airlines compared to price competition since
liberalization.
THE IMPACT OF RYANAIR
Ryanair, with 117 million passengers in 2016, is the largest airline in

Europe. It has more passengers than any of its previous national airline
rivals. Remarkably, as some of the old national airlines have grouped
together – with Lufthansa, IAG and Air France absorbing 11 previously
independent airlines to face the Ryanair challenge – Ryanair has emerged
larger than each of the three groupings. It is also 56 per cent larger than
easyJet and 400 per cent larger than Norwegian and Air Berlin in terms of
passengers carried.
Several factors have been instrumental in Ryanair expanding faster
than other European low-cost airlines. First, it had first-mover advantage
when it started to compete with British Airways and Aer Lingus in 1986.
Its founder, Tony Ryan, had previous experience in Aer Lingus in Dublin,

M4371-FINGER_9781786431851_t.indd 75 14/11/2017 11:14

London and New York and in aircraft leasing at Guinness Peat Aviation,
and the business model was based on that of Southwest Airlines in the
US; fares were low, leading to near immediate economies of density with
a rapid increase in traffic volume. Competitively, this also gave an advan-
tage on Ireland–UK routes because Ireland was in recession at the time of
the airline’s launch. Ryan recognized that another cartel airline was not
needed and that Ryanair had to differentiate itself from low-productivity
and high-cost legacy operations. Ryanair’s productivity could not have
been attained by adopting the traditional European national airline model.
Thus, Ryanair’s chief executive, Michael O’Leary, differentiated the airline
both from legacy airlines and from other start-ups.
Having achieved unprecedented productivity from its own staff com-
pared to incumbent airlines, Ryanair then targeted its suppliers. Discounts
were sought and achieved from aircraft manufacturer Boeing and from all
of the airports that Ryanair served. Airport competition was promoted to
create new routes and exert leverage on high-cost airports. For example,
the use of air bridges was declined and the design of terminals changed
to the low-cost model. Travel agents were removed from the selling of
Ryanair products and were replaced by online sales. Seats per aircraft were
maximized and load factors rose steadily to over 90 per cent. Free inflight
service was replaced by sales. Passengers readily took to the new product
because of the low fares on offer.
Ryanair achieved a 25-minute turnaround time at airports on the basis
that more round trips per day per plane would improve the economics
of the airline. Ryanair also located its fleet at 84 bases throughout the
network, an average of four per base. This contrasts with the old national
airline practice of locating aircraft at major hubs. All Ryanair planes
return to their home base each evening, in contrast to the expense incurred
by the legacy airlines of overnighting crews in hotels. Living away from
congested hubs reduced living costs for Ryanair crews and enhanced their
lifestyles. The traditional location of national airline fleets at busy hubs,
where they jointly operated the major air routes, left many second-city
markets open to new entrants such as Ryanair.
THE WIDER GEOGRAPHICAL CONTEXT
Critics of US airline liberalization have claimed disadvantages from the

policy such as financial and employment instability, reduced quality of
service, fewer flights and higher fares to smaller places and industry bank-
ruptcy (Goetz and Vowles, 2009). This seems to be the case to a lesser
extent in Europe.

M4371-FINGER_9781786431851_t.indd 76 14/11/2017 11:14

Irish airline financial instability affected Ryanair in its initial years, the
smaller airlines in Table 5.1 from time to time and Aer Lingus at intervals.
Weldon (2002) noted 28 years of loss-making between the establishment of
Aer Lingus in 1936/1937 and 2001, when it lost €140 million. Losses were
€317 million between 2006 and 2009 and €95.8 million in 2014. Ensuring
financial and employment stability at Aer Lingus was the motivation for
the government of Ireland to sell the airline to IAG in August 2015 and
delist it from the Irish Stock Exchange on 17 September. The alternative
bidder for Aer Lingus was Ryanair, which already held a 29.8 per cent
stake, but this option was opposed by the European Union, the UK com-
petition authorities and the Government of Ireland, which held a 25 per
cent stake.
Ryanair owes its existence to liberalization and has thrived in that
context. Its passenger numbers for 2017 to 2018 are predicted to be 125
million, and 135 million in 2019. Ryanair has annual profits of €250
million, a net margin of 15 per cent and a market capitalization of €15.5
billion. It employs 11 500 people and its dominant role in the Irish airlines
sector has secured overall profitability and employment growth in the
sector more than stability in its finances and numbers employed. The sug-
gested excessive instability does not seem to be present.
The second negative assessment cited by Goetz and Vowels is reduced
quality of service. The European airline that most prioritizes service
quality is CityJet, which offers leather seats, meals and the use of London
City airport with its proximity to the financial services district of London.
In 2016, CityJet has approximately the same passenger numbers as Aer
Lingus did before liberalization. Aer Lingus and British Midland dropped
their business class to compete with Ryanair. When food and bever-
age service on Aer Lingus was unbundled from high price tickets, it was
only on the Dublin–London and Dublin–Brussels routes that passengers
showed a willingness to pay for these services at the point of use. Many of
the services that were unbundled by price competition following liberaliza-
tion were inherited from the era of non-price competition by airlines. The
trade-off between price and non-price competition may change over time.
For example, in the European competition between Ryanair and easyJet,
Ryanair has sought to match its competitor by service improvements such
as flying to some hub airports, reserved seating and priority boarding,
while maintaining its claim to have the lowest prices. In 2017, Ryanair
secured a small presence at Frankfurt due to concerns over the dominance
of Lufthansa, but hubs such as Heathrow and Charles de Gaulle retain
grandfather rights in slot allocation.
It has also been argued that low-cost airlines will ultimately offer fewer
flights at higher fares to smaller places. In Ireland, this feared negative

M4371-FINGER_9781786431851_t.indd 77 14/11/2017 11:14

result of liberalization has not occurred: fares have fallen and frequencies
increased on all routes. Both Cork and Shannon, the pre-liberalization sec-
ondary airports, gained from increased frequency and reduced fares. Two
additional regional airports, Knock and Kerry, shared this experience. The
Ryanair network is hardly small, serving 1800 routes to 200 airports.4 Put
simply, there were many airports in Europe that the legacy airlines chose
not to serve because they based their operations overwhelmingly at hubs.
Low-cost carriers have filled this void.
The new entrant airlines were restricted by the slot system and the grand-
father rights of legacy airlines at hub airports. Therefore, the new entrants
sought out secondary airports. In 2010, Ryanair served 153 airports in
Europe. Only two of these, Madrid and London Gatwick, were in the top 10
airports (by passenger numbers), while 129 (84 per cent) of the Ryanair air-
ports were outside the top 100. Thus, liberalization in Europe brought sig-
nificant gains to secondary airport locations. The higher population density
in Europe compared to the US meant that secondary airport services had
more potential passenger numbers and the new routes proved viable. The
results of liberalization at Europe’s smaller airports were the opposite of
those feared in the US. The success of low-cost airlines at secondary air-
ports in Europe has prompted some hub airports to seek service from these
airlines at airports such as Brussels, Amsterdam and Copenhagen because
growth at such airports is reduced by reliance on legacy airlines alone.
Goetz and Vowles (2009) viewed industry bankruptcy as an undesirable
result of liberalization in the US. In stressing the role of freedom of entry
in the theory of contestable markets, William Baumol (1982) stated that
‘perhaps a bit newer is the emphasis on freedom of exit which is as crucial
a requirement of contestability as is freedom of entry. Thus, we must
reject as perverse the propensity of regulators to resist the closing down of
unprofitable lines of activity’.
Bankruptcy removes the control of assets from a management that has
failed the market test. Many long-established airlines, both in Europe and
the US, have either left the market or have been absorbed by other airlines,
as was the case with Aer Lingus. Having grown up in a tradition of pro-
tectionism, they were unable to survive in a liberalized market. Mergers
and takeovers were the alternative to bankruptcy for many airlines. The
industry is now more concentrated and this should be considered by com-
petition authorities dealing with further takeovers, both in the US and in
Europe. This aspect of liberalization is considered below.
Goetz and Vowles (2009, p. 259) also refer to ‘the renewed debate about
the efficiency of deregulation and liberalization policies, particularly at a
time when the global financial crisis has cast a harsh spotlight on the (un)
desirability of these polices’. The Irish example does not show any conta-

M4371-FINGER_9781786431851_t.indd 78 14/11/2017 11:14

gion linking the severe financial crisis and successful aviation liberaliza-
tion. Regarding airline liberalization in Ireland, the only regulatory change
was the removal of the restrictions on new entrants. All other standards as
applied to Aer Lingus were applied to the new entrants. Contagion may
well exist but as a contextual matter that did not apply to Ireland.
The Goetz and Vowles view of airline deregulation in the US is signifi-
cantly more pessimistic than the Irish experience since 1986. Kahn’s (1988,
p. 321) assessment of US deregulation is closer to the results in Ireland,
namely,
The last ten years have fully vindicated our expectations that deregulation
would bring lower fares, a structure of fares on average in closer conformity
with the structure of costs, an increased range of price-quality options and great
improvements in efficiency – made possible by the abandonment of regulatory
restrictions and compelled by the greatly increased intensity of competition – all
this along with a 35 per cent or so decline in accident rates.
LIBERALIZATION AND AVIATION SAFETY ON

IRISH AIRLINES
The Irish Aviation Authority (IAA) has responsibility for safety, jointly
with the United Kingdom and Canada, of a large section of the North
Atlantic. The IAA is also responsible for the safety regulation of some 500
civil commercial airliners in the Irish fleet, of which 350 belong to Ryanair
and 50 to Aer Lingus. IAA also regulates safety at each destination flown
by Irish airlines. It is itself regulated by the International Civil Aviation
Organization (ICAO).
According to the chief executive of the IAA, ‘ICAO ranks Ireland
consistently among the world’s top four regulators ahead of the US which
came sixth in 2015. The FAA categorizes Ireland as a category 1 state.
They take our certificates unreservedly’ (Irish Times, 8 July 2016). The
ICAO 2015 Universal Safety Oversight Audit ranked Ireland second in
Europe for effective implementation of almost 11 000 ICAO standards and
recommendations. Ireland’s compliance rate was 94 per cent, compared to
93.63 for the UK, 91.35 per cent for USA, 89.8 per cent for Germany, 94.4
for France and 95.3 for Canada. Ryanair has never had a passenger fatal-
ity, while the last Aer Lingus passenger fatality was in 1968.
The IAA website states that the IAA ‘regulates legacy and low-cost
airlines to the same safety and operating standards which conform to all
European Union and European Aviation Safety Agency regulations and
requirements’. Some early opponents of liberalization in Ireland claimed
that new-entrant airlines would use inferior aircraft and would be poorly

M4371-FINGER_9781786431851_t.indd 79 14/11/2017 11:14

regulated in safety matters. In fact, the Ryanair fleet is comprised entirely

of Boeing 737–800 with an average age of five years and the company’s
safety regulation compliance is among the highest of all ICAO member
states.
LIBERALIZATION ON THE NORTH ATLANTIC

ROUTES
The main emphasis in Irish aviation policy before the EU/USA Open
Skies agreement took effect in 2008 was to protect Shannon from competi-
tion from Dublin while reducing the market share of US airlines on the
Ireland–USA routes.5 The Ireland/US Air Services Agreement of 1945
only permitted flights to and from Shannon and gave Irish airlines rights
at New York, Boston and Chicago. Only in 2008 did Ireland finally allow
US airlines unrestricted access to Dublin under the EU/USA Open Skies
agreement.
The path to Open Skies over 63 years showed a variety of obstructive
measures by Ireland to keep American airlines out of Dublin for as long
as possible. The Aer Lingus New York to Dublin flight landed at Shannon
compulsorily in both directions. The plane was designated an Aer Lingus
flight between Dublin and Shannon and an Aer Linte flight between
Shannon and New York. US airlines could land at Shannon but their
Dublin passengers could only make their onward journey by Aer Lingus.
In 1971, the US Civil Aeronautics Board proposed banning Aer Lingus
from New York unless TWA could serve Dublin. As a result, TWA was
allowed to serve Dublin but only with a compulsory stop at Shannon. By
1993 the requirement was that 50 per cent of flights should either originate
or stop over at Shannon. From 2006, it was required to serve Shannon on
one in three flights that served Dublin.
The Irish restrictions on access by US airlines to Dublin were intended
to end in April 2008 but the EU/US Open Skies agreement came into effect
in March 2008. Between 2006 and 2009, Shannon’s transatlantic traffic fell
from 781 000 passengers to 442 000, and by 2012 it had further declined to
288 000. Passengers on all routes at Shannon halved from 2.6 million to
1.3 million between 2009 and 2012, the trough year at the airport. In 2013,
Shannon became an independent stand-alone state airport and between
2013 and 2015 its traffic grew by 31 per cent, although this was still only 57
per cent of its 2008 peak.
The Shannon stopover became unpopular with Aer Lingus, an intended
beneficiary, when it faced a liberalized European market. It was inefficient
to fly transatlantic aircraft over 112 miles from Dublin to Shannon and to

M4371-FINGER_9781786431851_t.indd 80 14/11/2017 11:14

have aircraft and crew wait there while passengers disembarked to engage
in duty-free shopping for an hour. On eastbound journeys, the stopover
required the planes to land at Shannon in the early morning, thus disrupt-
ing the sleep patterns of those on the overnight flight before flying on to
Dublin. Given the absence of direct flights to Dublin, many passengers
chose voluntarily to stop over at London instead of Shannon. In 2006,
when Shannon’s North Atlantic passengers peaked at 781 000, they were
37 per cent of the Dublin and Shannon combined market on the routes. By
2012 the Shannon share had declined to 15 per cent.
In 2015 Dublin was the fifth-largest airport in Europe in terms of
traffic to North America. The requirement to serve both Dublin and
Shannon had improved the competitiveness of Dublin routes serving
North America, winning back traffic that had been diverted over London.
Dublin has also engaged in competition with London by becoming a hub
that feeds its North Atlantic services from its large UK and European
mainland network of routes. In 2016, one million passengers from Aer
Lingus’s European routes transferred to services to North America from
Dublin. US customs and border protection services conduct customs and
immigration pre-clearance procedures at Dublin and Shannon so that pas-
sengers are treated as domestic passengers on arrival at their US destina-
tion airports.
While the Shannon stopover was a burden imposed by Ireland on
itself, and could have been removed at any time, its demise in 2008 as
part of Open Skies was a further benefit of liberalization. The end of the
Shannon stopover brought a Blue Skies air agreement between Ireland
and Canada, permitting services between the countries and allowing the
countries to use each other as platforms to serve other destinations. In the
summer of 2016 there were 20 daily flights from Dublin to the US and
four daily to Canada (to Toronto, Montreal, Vancouver and St John’s).
Added to this, a fifth freedom service was opened in 2015 by Ethiopian
Airlines, serving Dublin in both directions on its Addis Ababa–Los
Angeles route.
Norwegian Air’s attempt to enter the Cork–Boston route in May 2016
was delayed by opposition in the US to the granting of a Foreign Air
Carrier Permit. The service is now scheduled to begin in 2017. Norwegian
Air is registered in Ireland and sought to serve the US under the Open
Skies agreement between the EU and USA. After lengthy objections from
airlines in the USA, and aviation trade unions in the US and those rep-
resenting pilots at Aer Lingus, Norwegian was permitted on 3 December
2016 to serve the Ireland–US market. It has 55 aircraft on the Irish register
and plans to feed its Cork–Boston service from Barcelona. Further plans
in Ireland include serving Shannon and New York.

M4371-FINGER_9781786431851_t.indd 81 14/11/2017 11:14

LIBERALIZATION AND AIRPORT COMPETITION
The control of airport slots at hub airports by airlines in order of senior-

ity under grandfather rights is an obvious barrier to new entrant airlines.
Therefore, the development of new airports was vital for competitive
aviation in Europe. The four largest low-cost airlines had 250 million pas-
sengers in 2015, meaning that even if the grandfather rights issue had been
addressed by regulatory authorities, additional airport capacity would
have been required. Low-cost airlines like Ryanair highlighted the amount
of under-utilized airport capacity in Europe, but opening under-utilized
airports attracted opposition from legacy airports and airlines. Had this
opposition succeeded, the under-utilized airports would have been con-
fined to activities such as general aviation and flying clubs.
In Ireland, liberalization was initially associated with the construction of
several new airports in a market controlled by the state airport company,
Aer Rianta, which operated in Dublin, Cork and Shannon. Dublin was
not slot-constrained and accommodated the growth of Ryanair. All
airports participated in the growth of Irish aviation after liberalization
in 1986 until the banking and property crisis of 2008, when passenger
volumes declined until 2015.
The recession also coincided with the opening of an extensive motorway
network between 2010 and 2012; this combined impact of recession had
an adverse impact on airport traffic (Table 5.4). Dublin recovered from the
recession faster than the other regions and improved road access to Dublin
diverted traffic from the regional airports. Thus, two regional airports
have closed and one has lost scheduled commercial service. Efficiency
issues have arisen at all three state airports, with Cork and Shannon seeing
particularly large declines in passenger numbers. As Dublin increases its
dominance, the emphasis on its efficient regulation grows more vital.
Passenger numbers increased 9.5-fold between 1985 and 2008. Passenger
numbers in 2015 were 4.8 per cent below the 2008 numbers and market
share at Dublin in 2015 increased to 83.8 per cent from 75 per cent in 2008.
Passengers at Dublin in 2015 were 6 per cent higher than in 2008, while
Table 5.4 Passenger shares of Ireland’s airports
Passengers Airport passenger share (in percentage)

(millions)
Dublin Cork Shannon Knock Kerry Others
1985 3.3 69.9 7.8 22.3 0 0 0
2008 31.3 75.0 10.4 9.4 2.0 1.4 1.8
2015 29.8 83.8 6.9 5.6 2.3 1.0 0.4

M4371-FINGER_9781786431851_t.indd 82 14/11/2017 11:14

those at Cork and Shannon were 37 and 43 per cent lower, respectively,
with Cork declining by 1.2m passengers and Shannon by 1.3m passengers.
The period from 2008 to 2015 also saw a decline at the new airports, other
than Knock, which entered the market after liberalization. Galway, which
had 266 000 passengers in 2008, closed in 2011. Sligo had 42 000 passengers
in 2008 but also closed in 2011. Waterford was the base for Ryanair’s first
route, to London Gatwick in 1985, but when it chose to concentrate solely
on the Boeing 737 in 1991, Ryanair moved to other airports. Waterford
had 144 000 passengers in 2008 but lost passenger service in 2016 when
Belgian airline VLM left the Waterford–London route. The VLM chief
executive stated that ‘Waterford is close enough to Dublin and Cork, so the
catchment area will always be sort of limited’.6
Turning to matters of ownership, Shannon’s traffic fell steeply after
2008, falling from 3 million to a trough of 1.3 million in 2012 but rising to
1.7 million in 2015. It has had independent governance since 2013, with the
aspiration that this will enable it to compete with Dublin more effectively
than if remaining part of the Dublin Airport Authority.7 Independence for
Cork Airport will be re-examined in 2019. The Airport Authority narrowly
declined the option of independence in 2008 because of fears about the
airport’s debt. Its decline has been continuous since 2008, from 3.3 to 2.1
million passengers, but turned around in 2016.
Cork and Shannon are legacy state airports that predate airline liberaliza-
tion in 1986. In that era, all state airports were run by Aer Rianta. Airlines
did not compete on price and passed on higher airport charges in the form
of higher fares. In that era, there were no new entrant airlines choosing
between competing airports and there was no independent regulation of
airport charges. There were likely inefficiencies in such a market structure,
with dangers to the efficient allocation of both capital and labour resulting
in ‘Averch–Johnson effects’ (basically gold-plating) and low labour produc-
tivity.8 Capital projects were prestige buildings that embodied the region or
neo-Keynesian development measures, rather than commercial investments
where a wrong decision could harm the competitiveness of an airport.
Shannon invested £40 million in a terminal in 2000 when it had 2.4 million
passengers, a number that declined to 1.7 million in 2015. Cork invested
€120 million in a new terminal in 2006 and the Airport Authority took
responsibility for the debt in 2008, thus reducing the competitiveness of the
airport; however, as we have seen, passenger numbers declined until 2016.
Criticisms by the Commission for Aviation Regulation of capital spend-
ing at Irish airports, as proposed in 2001, included:
lack of consultation with user airlines; lack of transparency in regard to costs;

construction of facilities that are inefficient, and/or do not meet the needs of

M4371-FINGER_9781786431851_t.indd 83 14/11/2017 11:14

users in line with best international standards; inadequate or non-existent cost

benefit analysis or business cases to justify specific capital programmes and the
supply of inconsistent information to the regulator (Prasifka, 2002, pp. 51–2).
The regulator disallowed €50 million of the €106 million investment pro-
posed at Shannon and €79 million of the €136 million proposed at Cork.
Labour productivity at Cork and Shannon was examined by the
Infrastructure Management Group (2001). While Cork was 15 per cent
above the average of its UK peers, Shannon was only 37 per cent of that
average. Part of the justification for the Shannon stopover discussed
above was the creation of employment in the region. However, creating
employment by over-manning an airport undermines the competitive-
ness of a region. Cork’s entry to the North Atlantic market in 2017 with
Norwegian Air is a new market for the airport. Shannon will also be served
by Norwegian with the advantage of US immigration and pre-clearance.9
The distribution of Irish airport passengers towards Dublin was also
influenced by the development of motorways and the reduction of subsi-
dies to regional air services. As we have seen, the motorway network, with
connections from Dublin and Dublin Airport to Cork, Galway, Limerick
and Waterford, had an impact on air travel. Internal air routes to Dublin
peaked at 885 000 passengers in 2007 but declined to 61 000 in 2012 after
the completion of the surface network. The air services affected com-
prised a mix of commercial routes, such as Cork and Shannon to Dublin,
and subsidized services to smaller centres. Cork to Dublin declined from
half a million passengers in 2007 to under 5000 in 2012 and by 2015 had
no service. Similarly, Shannon to Dublin declined from 100 000 in 2006
to 1393 in 2012 and had no service in 2015. The subsidized routes from
Dublin to Galway, Sligo and Knock were withdrawn from the public
service obligation subsidy scheme in 2011, leaving the services to Kerry
and Donegal as the only remaining parts of the domestic route network.
These routes had 84 000 passengers in 2015, which accounted for only
0.3 per cent of passengers at Dublin. The PSO subsidies to Donegal and
Kerry are scheduled to end in 2018. Unless actions by airlines and airports
improve the economics of these routes, there will be no remaining internal
routes in Ireland.
Regional airports in Ireland gained from the initial stages of liberaliza-
tion when turboprop aircraft could undercut the high fares that the incum-
bent airlines were charging on services to the UK and mainland Europe,
and Aer Lingus responded by operating turboprop feeder services to its
hub at Dublin. The lower fares charged by Ryanair and a restructured Aer
Lingus subsequently undermined the economics of both direct turboprop
services to Irish regional airports and the Aer Lingus feeder services to

M4371-FINGER_9781786431851_t.indd 84 14/11/2017 11:14

Dublin. Liberalization of express coach services to Dublin Airport also

offered quick, low-cost access over the motorway network to the greater
network of destinations, frequency of flights, and choice of airlines at
Dublin.
A further aspect of airport competition within Ireland has been the
attraction of an estimated million passengers from Northern Ireland.
Dublin Airport is located on the motorway to Belfast, about 90 minutes
away. There are no border controls between the Republic and Northern
Ireland. Dublin serves five times more cities than Belfast’s two airports
and has 10 times more peak seat capacity. The air passenger duty tax levied
at Northern Ireland for journeys of up to 2000 miles is £13 for the lowest
class of travel and £26 for any other class, while for longer journeys it is
£75 for the lowest class and £150 for any other class. The United Airlines
Belfast–Newark service was exempted from the £75 and £150 taxes to
improve the finances of the route, but United withdrew in January of
2017, citing ‘the route’s poor financial performance’. The airline did not
pursue a subsidy proposal from the Northern Ireland government, fearing
that it would breach EU state aid rules. The Republic of Ireland abolished
air travel tax as a promotional measure for an island economy seeking to
attract tourism and trade. This move, combined with intense price com-
petition between Ryanair and Aer Lingus at Dublin and more frequent
flights to more destinations, led to passengers transferring from Northern
Ireland to Dublin.
Dublin Airport has also faced regulatory criticism of its labour produc-
tivity. In 1999, the consultants Infrastructure Management Group found
the airport’s operating costs were 30 per cent higher than the best of its
peers, namely Brussels, Copenhagen, Glasgow and Stansted. Dublin’s
workload units per staff member were under half the average of the four
comparator airports. Of the €757m capital projects sought by the airport,
the regulator approved only €160m. The regulator’s decisions were chal-
lenged by the Airport Authority between October 2001 and October
2014 when a Supreme Court appeal was withdrawn by the company.10
Disputes between airport and airline surface from time to time. In 2017,
for example, Ryanair will cut its capacity at Dublin by 2.4 per cent because
of the discontinuation of an incentive plan by the airport.
ON-GOING CHALLENGES OF LIBERALIZATION
Combining Kahn’s optimistic stance, the pessimistic view of Goetz and

Vowles, and the positive Irish experience of airline liberalization leads to
consideration of its future regulation.

M4371-FINGER_9781786431851_t.indd 85 14/11/2017 11:14

Button (2012) stated that ‘the time has come to begin experimenting with
more infrastructure reforms in the US’ and called for ‘more market-based
approaches to the supply and use of air transportation infrastructure’. At
airports, ‘fees do not reflect the prevailing congestion levels at airports and
the congestion caused by different types of aircraft at different times of
the day.’ Poole and Butler (1988) proposed that ‘airports should be free to
increase their capacity and to charge congestion pricing’. They also pro-
posed the privatization of air traffic control and the use of new technology
to increase capacity. The success of airline liberalization in generating extra
revenues for airports and air traffic control would fund infrastructure, in
contrast to the congestion that is currently being experienced.
The allocation of slots at hub airports based on airline seniority is an
obvious barrier to new entrant airlines. While, in economic terms, airlines
are customers of airports, airlines have acquired property rights to slots
over time and a market in slots has developed. An example at Heathrow
is the sale of a slot in February 2015 by SAS to Turkish Airlines for $22
million. The allocation of slots at hub airports by airlines rather than by
airport management is a barrier to new entrant airlines. The allocation of
slots at hub airports by grandfather rights rather than by price or admin-
istrative intervention has remained the practice. Competition authorities
have not intervened and airport management have abdicated the role. In
the takeover of Aer Lingus by IAG in 2015, no provision was made for slot
reassignment on the Heathrow routes from Dublin (1.7 million passengers
per year), Cork (400 000 passengers) and Shannon (200 000 passengers).
IAG formerly faced competition from Aer Lingus on the three routes and
from British Midland on the Dublin route but, since the takeovers, has had
a monopoly on Ireland–Heathrow routes.
Therefore, secondary airports in Europe were crucial to successful
airline liberalization. Europe’s large number of empty airports provided
the vital market access needed by new entrant airlines that were typically
excluded from hub airports. The new entrant airlines and the secondary
airport managements took the entrepreneurial risk in opening new air
routes from virtually unused airports. Passengers responded initially to
avail themselves of low airfares but they also grew to like low-cost airports
with less crowding, less complexity, less walking, cheaper car parking, and
nearness to local airports.
Charleroi airport, 29 miles from Brussels, is an example of the success
of a secondary airport in a liberalized market. Charleroi lost its service
to London by Sabena during the 1970s because of poor results. Ryanair
commenced a service to Dublin in 1997 and an early objector was the main
Brussels Zaventem airport. Charleroi increased its passenger numbers
from less than 20 000 in 1996 to 5.5 million in 2015. The European

M4371-FINGER_9781786431851_t.indd 86 14/11/2017 11:14

Commission upheld complaints against Charleroi and Ryanair in 2004

because the airport ‘took risks that a private investor acting in a market
economy would not have taken’. The advantages granted to Ryanair were
deemed state aid that distorted competition not just on one or more routes
but over the entire network of Ryanair. The assistance to Ryanair was
deemed ‘incompatible with the common market’, discriminatory, unlawful
under Belgian law and contrary to the principle of proportionality.
In 2008, the European Court of First Instance upheld Ryanair’s appeal
against the Commission’s decision of 2004 because the Commission had
not applied the private investor principle in its examination of the case.
The Court ordered the Commission to pay Ryanair’s costs and ordered the
Association of European Airlines, which entered the case in support of the
Commission in November 2004, to pay its own costs (Barbot, 2006).
In October 2014, Ryanair was cleared in the EU Commission’s inves-
tigation of alleged distortions of competition by assistance at Charleroi,
Hahn, Alghero and Vasteras airports. By September 2016, 24 airports had
been investigated under EU State Aid rules (Table 5.5).
The list indicates that the EU Commission, frequently acting on behalf
of incumbent airports and airlines, is an obstacle to airline liberaliza-
tion at secondary airports. The litigation costs and the time taken (in the
Charleroi case, for example) were borne by Ryanair, which is the most suc-
cessful new entrant. These costs might deter market entry by a small, start-
up airline. Given the Commission’s poor record in gaining access to hub
airports for new entrant airlines, its repeated opposition to new entrants
gaining market access at secondary airports is a double disadvantage that
new entrants face. The Market Economy Investor Principle, which the
Commission is required to apply in assessing complaints about airline
deals at secondary airports, does not consider factors such as network
externalities, ancillary revenues in airport retailing and car parks, the avail-
ability of incentives to other airlines, or how a privatized airport would
conduct negotiations with actual and potential competing airlines.
The final key factor in developing liberalized airline markets is the treat-
ment of airline takeovers and mergers. The failure of the competition
authorities in both the US and European Union to assess mergers and
takeovers from the wider public perspective is a threat to contestability.
Airlines in Europe that used to collude under bilateral agreements are
now jointly owned. This is likely to deter new entrants that have played a
major role in introducing competition to a sector where new entrants were
excluded and state airlines acted in concert in timetabling, pricing and
revenue sharing. The three major merged groups of legacy airlines, IAG,
Air France-KLM and Lufthansa, had previously absorbed potential com-
petitors within their home countries.

M4371-FINGER_9781786431851_t.indd 87 14/11/2017 11:14

Table 5.5 Ryanair Airport state aid cases as of September 2016
Airport Million passengers (2015) State Aid

Charleroi 5.5 No
Beauvais 3.7 TBD
Hahn 2.3 No
Niederrhein 1.9 No
Marseille 1.8 No
Cagliari 1.6 Yes
Girona 1.5 TBD
Schönefeld 1.4 No
Alghero 1.1 No
Bratislava 1.0 No
Carcassonne 0.4 TBD
Reus 0.3 TBD
Nîmes 0.2 Yes
Tampere 0.2 No
Aarhus 0.1 No
La Rochelle 0.1 TBD
Västerås 0.1 No
Lübeck 0 TBD
Klagenfurt 0 TBD
Târgu Mureș 0 TBD
Altenburg 0 Yes
Angoulême 0 Yes
Pau 0 Yes
Zweibrücken 0 Yes
Notes: No 5 finding of no aid by the European Commission; TBD 5 to be decided by the

European Commission; Yes 5 finding of aid by the European Commission. Six findings of
aid are under appeal or awaiting appeal.
Source: Ryanair, September 2016.
CONCLUSIONS
The value of airline liberalization to the Irish economy is great. As an outer

offshore island, Ireland penalized itself with its pre-1986 commitment to
keeping new entrant airlines out of its airports. The high fares charged by
the protected airlines amounted to a tax on access, paid to the protected
airlines, which had low staff productivity compared to the liberalized air-
lines operating currently. Low-cost aviation reduced the isolation of the
Irish economy and is important for tourism and foreign direct investment.

M4371-FINGER_9781786431851_t.indd 88 14/11/2017 11:14

Irish airlines moved quickly in the new, liberalized environment. Ryanair

rapidly availed itself of the market opportunities arising from liberaliza-
tion and carries about one-third more passengers than each of the three
sets of merged legacy airlines in Europe, and half as many again as the
second low-cost start-up, easyJet. It has the second-highest market capi-
talization on the Irish stock exchange and accounts for over 16 per cent
of the capitalization of the entire stock exchange. It is the most successful
airline in the era following liberalization in Europe. Opposition from the
European Union, the UK and Ireland to Ryanair’s takeover of Aer Lingus
and the requirement that it sell its Aer Lingus shares to IAG contrasts with
the ease with which Air France, Lufthansa and British Airways were able
to take over many potential competitors. The disallowed combination of
Ryanair and Aer Lingus might have been a competitive stimulus to IAG,
Air France-KLM and Lufthansa.
The consolidation of aviation with European Union approval recalls
Kahn’s (1988, p. 319) criticism of the same policy in the US:
The concentration process also reflected what many of the advocates of

deregulation would characterize as a lamentable failure of the administration
to enforce the policies of the antitrust laws, to disallow a single merger . . . or
attack a single case of predation . . . I take perverse satisfaction in having pre-
dicted the demise of price-cutting competitors like World and Capitol Airways
if we did nothing to limit the predictable response of the incumbent carriers to
their entry.
Given the tradition among Europe’s legacy airlines of not competing with
one another, low-cost airlines and new entrants are key to retaining the
gains from liberalization. New entrant airlines remain largely excluded
from major hub airports. The success of Ryanair in finding substitutes for
hub airports has occurred despite the opposition of the European Union,
notably at Charleroi.
NOTES
1. In Ireland, the capture of the Department of Transport by Aer Lingus was so successful
that the Department was described as the ‘downtown office of Aer Lingus’.
2. This followed the pattern described by Tullock (1967) and Krueger (1974), who warned
of the costs to society from successful rent-seeking and lobbying as a substitute for
competitive enterprise.
3. Before liberalization, airlines overbooked flights in anticipation of ‘no-shows’ and
denied boarding to those checking in late. The low-cost model imposes the cost of no-
shows on the passenger who fails to check in.
4. More broadly, Fewings (1998) looked at 431 airports in 13 European countries; exclud-
ing the special case of Iceland, with a ratio of 49 airports per million people, the average

M4371-FINGER_9781786431851_t.indd 89 14/11/2017 11:14

is 1.9 airports per million people. David Starkie (2008) listed 56 airports on the UK
mainland serving 58 million people.
5. For an assessment of the network effects of this deregulation, see Button (2009).
6. Irish Independent, 19 June 2016.
7. Cahill et al. (2017) provided more details on the relative efficiency of Ireland’s airports.
8. See Averch and Johnson (1962).
9. The devaluation of sterling following the Brexit referendum in mid-2016 has increased
the cost of visiting Ireland for UK residents. Ryanair has decided to reduce its capacity
at Shannon in 2017 by 100 000 seats.
10. The airport authority stated that ‘airline operators have a powerful incentive to oppose
large scale expansions of airport facilities, given that such expansion will frequently
allow the entry of further competitors’ and ‘so-called “low-cost” operators generally
attach little or no value to the provision of facilities at airports other than essential facili-
ties such as checking in, baggage handling and passenger embarkation’ (Barrett, 2009).
REFERENCES
Association of European Airlines (1984), Comparison of Air Transport in Europe

and the USA, Brussels.
Averch, H. and L.L. Johnson (1962), Behavior of the firm under regulatory con-
straint, American Economic Review, 52, 1052–69.
Barbot, C. (2006), Low-cost airlines, secondary airports and state aid: An eco-
nomic assessment of the Ryanair–Charleroi agreement, Journal of Air Transport
Management, 12, 197–203.
Barrett, S. (2009), Deregulation and the Airline Business in Europe, London and
New York: Routledge.
Baumol, W.J. (1982), Contestable markets: An uprising in the theory of industrial
structure, American Economic Review, 72, 1–15.
Button, K.J. (2009), The impact of EU–US ‘Open Skies’ agreements on airline
market structures and airline networks, Journal of Air Transport Management,
15, 59–71.
Button, K. (2012), Airline Deregulation: Myth or Reality, Mercatus Center Paper
108, Fairfax: George Mason University.
Cahill, C., D. Palcic and E. Reeves (2017), Commercialisation and airport perfor-
mance: The case of Ireland’s DAA, Journal of Air Transport Management, 59,
155–63.
Doganis, R. (2001), The Airline Business in the 21st Century, London: Routledge.
European Economic Community (1981), Scheduled Passenger Air Fares in the
EEC, Brussels: EEC.
Fewings, R. (1998), Provision of European airport infrastructure, Avmark Aviation
Economist, 15, 18–20.
Goetz, A. and T. Vowles (2009), The good, bad and the ugly: Thirty years of US
airline deregulation, Journal of Transport Geography, 17, 251–63.
Infrastructure Management Group (2001), Benchmarking Report for Commission
for Aviation Regulation, Bethesda: IMG.
International Air Transport Association (1977), Reason in the air: towards a
balance of international aviation needs, Geneva: IATA.
International Civil Aviation Organization (1990), Annual Survey of Air Fares,
Montreal: ICAO.

M4371-FINGER_9781786431851_t.indd 90 14/11/2017 11:14

Jordan, W. (1970), Airline Regulation in America, Baltimore: Johns Hopkins.

Kahn, A. (1988), Surprises of airline deregulation, American Economic Review, 78,
316–22.
Krueger, A. (1974), The political economy of the rent seeking society, American
Economic Review, 64, 291–303.
Poole, R. and V. Butler (1988), Airline Deregulation: The Unfinished Revolution,
Washington: Competitive Enterprise Institute.
Prasifka, W. (2002), Aer Rianta CPT and the Commissioner for Aviation
Regulation, Affidavit, High Court Judicial Review 2001/707, Dublin, dated 14
February.
Starkie, D. (2008), Aviation Markets, Aldershot: Ashgate/Institute of Economic
Affairs.
Stigler, G. (1971), The theory of economic regulation, Bell Journal of Economics
and Management Science, 2(1), 3–18.
Tullock, G. (1967), The welfare costs of tariffs, monopolies, and theft, Western
Economic Journal, 5, 224–32.
UK Civil Aviation Authority (1983), A Comparison between European and US Air
Fares, CAA Paper 83006, London: CAA.
UK Civil Aviation Authority (1987), Competition on the Main Domestic Trunk
Routes, CAA Paper 87005, London: CAA.
Weldon, N. (2002), Pioneers in Flight, Dublin: Liffey Press.

M4371-FINGER_9781786431851_t.indd 91 14/11/2017 11:14

6. The evolution of Indian civil aviation
Rajiv Nagpal and Haritha Saranga
INTRODUCTION
Civil aviation in India has come a long way since it took wings when the
Indian State Air Services, in collaboration with Imperial Airways, opened
the first domestic air route between Karachi and Delhi in December of
1912 (Association of Private Airport Operators, 2013). Currently, India
is the ninth largest and the fastest-growing aviation market in the world
and has the capacity to cater to more than 220 million passengers and 4.6
million tonnes of air cargo per annum. India is expected to become the
third-largest aviation market in the world, with 450 million passengers by
2020. Most of this growth has come about during the last 25 years, post-
liberalization of the Indian economy in the 1990s.
At the time of India’s independence in 1947, nine air transport compa-
nies carried air cargo and passengers within and beyond the boundaries of
the country, with Air India – owned by the Tata Group – being the largest
of these carriers. In 1948, the Government of India acquired 49 per cent
of Air India and permitted it to operate international services under the
name Air India International (ATIC, 1950). Soon after, with the enactment
of the Air Corporation Act 1953, the Government nationalized the airline
industry and the assets of nine existing companies were assigned to the two
new corporations: Air India International and Indian Airlines. These cor-
porations served on international and domestic routes, respectively, under
the control of the Director General of Civil Aviation India (DGCA), the
Indian governmental regulatory body.
However, after a period of national carrier monopoly, the economic
reforms in the 1990s saw a re-privatization of the airline sector with the
entry of prominent airlines such as Jet Airways and Air Sahara following a
relaxation of passenger, cargo and foreign direct investment (FDI) norms.
Significant growth in air traffic in India finally arrived with the advent of
low-cost carriers in the early to mid-2000s, as these airlines brought down
fares and expanded the airline network. These airlines have also given rise
to price wars, bringing profits under immense pressure. While private air-
92

M4371-FINGER_9781786431851_t.indd 92 14/11/2017 11:14

The evolution of Indian civil aviation 93
lines were allowed to operate, there were several restrictions with respect to
route decisions, pricing, funding and MRO facilities. In the mid-2000s, the
airport infrastructure was given a boost through partnerships with private
players and FDI. The Greenfield Airport Policy is expected to boost
further the airport infrastructure in areas that currently suffer from poor
connectivity. In addition, the Airports Authority of India has endeavoured
to upgrade air navigation systems in line with global standards to ensure
smooth operations over the airspace.
This chapter examines the major policy changes over the last 30 years
and their impact on airlines, airports and passengers. It also characterizes
the market structure post-deregulation and the fate of various players in
the course of these changes. Regulatory changes in airline operations,
airport and air navigation services are discussed in separate sections, fol-
lowed by a discussion of the environmental impact of the aviation sector.
Finally, the section on future outlook describes the most recent policy
changes that aim to improve regional connectivity and the ease of doing
business in the airline industry.
THE ONSET OF AIRLINE LIBERALIZATION IN

INDIA
Indian economic policy post-independence from the British in 1947 and

before the reforms of 1991 was influenced by an exploitative colonial expe-
rience. Import substitution, state intervention, micro-level monitoring,
regulation and protectionism characterized industry and trade. The Indian
business ecosystem came to be known as the ‘Licence Raj’ by virtue of
elaborate licences required at every stage of running a business. All major
industries were nationalized, and the government attempted to close the
Indian economy to the outside world. The aviation industry in India was
no exception: after nationalization in 1953, government-owned carriers
monopolized the aviation sector.
In the 1980s, efforts were made to spur industrial development and
improve the competitive efficiency of the Indian economy in the global
marketplace. The economy was undergoing a near-crisis situation reflected
by an unmanageably high rate of inflation and unhealthy current account
deficit (Joshi and Little, 1996). Economic liberalization and globalization
measures were rolled out by the government to overcome the crises and
foster high levels of GDP growth. As the Indian government reforms led
to rapid growth of incomes and employment, the air traffic was estimated
to grow at 9.7 per cent over the next decade (Hooper, 1997). Air travel
was expected to play a crucial role not only as a fast and efficient mode of

M4371-FINGER_9781786431851_t.indd 93 14/11/2017 11:14

transport, but also in connecting the diverse topological and demographic

corners of the country. In addition, under the Tourist Action Plan, the
government aimed to increase the number of international visitors from
1.6 million to 5 million per year. Therefore, the sector required greater
capacity to promote industrial development and tourism.
Indian Airlines, a mélange of eight parent airlines and the only domes-
tic carrier, lacked the capacity to meet the projected growth in demand
(Krishnan, 2008). Moreover, in the 1970s Indian Airlines Corporation
flew through turbulent times, as it was responsible for providing logis-
tics support during major engagements with neighbouring countries.
Operations were often interrupted and Indian travellers had to resort to
rail connections; it was also solely responsible for providing connectiv-
ity on unprofitable routes in underdeveloped regions like the north-east.
A host of import duties and high sales tax on fuel prices further limited
profitability, leading to low employee wages and eventually labour unrest
(National Transport Development Policy Committee, 2013). With seven
reported safety incidents (Aviation Safety Network, 1990) in the 1970s
attributed to pilot or crew errors, the airline exhibited a patchy safety
record in addition to a reputation for delayed flights and declining service
standards. Fleet purchase decisions, route selection and employee recruit-
ment suffered from political interference in the mid-1980s, and this often
cost the airline its elite image (Krishnan, 2008). While the national carrier
had built a vast domestic network, the prevailing situation led to calls for
the opening up of the airline sector.
In an attempt to boost tourism, the government introduced the Air Taxi
system in 1986, wherein private carriers were allowed to operate Air Taxis
but not scheduled services (Hooper, 1997). Air Taxis were small aircraft,
which operated on an on-demand basis, but with restrictions on seat
capacity, airports, timing and fare. While this was an initial step to usher
in private participation in the sector, it proved insufficient to generate real
interest.
LIBERALIZATION IN THE 1990s: OPEN SKIES

POLICY
The Indian government finally woke up to reality in April 1990, and

announced the ‘Open skies’ policy to usher in private players into the
aviation sector in earnest. As part of this policy, Air Taxi operators were
allowed to operate charter and non-charter flights from any airport, decide
flight schedules and set their own fares (Nathan Economic Consulting
India, 2012). East West Airlines – the first national private airline to

M4371-FINGER_9781786431851_t.indd 94 14/11/2017 11:14

operate in India since 1953 – took to the skies with a fleet of three Boeing
737–200s. Soon after its maiden flight from Bangalore to Mumbai, the
airline was asked by the Civil Aviation Minister to bring in more aircraft to
counter the impact of a disabling Indian Airlines pilot strike. This marked
a beginning of the re-privatization of the civil aviation industry in India.
Another significant move in the same direction was that Air Taxis were
permitted to obtain up to 40 per cent of equity from foreign sources, with
100 per cent equity participation for NRIs and Overseas Corporate Bodies
(Hooper, 1997). With this option and 20 per cent funding from each of
Kuwait Airways and Gulf Air, Jet Airways leased a fleet of four Boeing
737–300 aircraft and started operations in May 1993. The airline would
later go on to become a fully-fledged private airline, operating 300 daily
flights to 68 destinations worldwide. Damania Airways, ModiLuft and
Air Sahara were other prominent operators who took off in 1993 under
the Air Taxi system. As can be seen from Figure 6.1, the years following
the open skies policy marked a steady increase in the number of domestic
passengers handled by private carries.
The ‘open skies policy for air cargo’, adopted in 1990 for a period of
three years, was extended in 1992 on a permanent basis (Aeronautical
Information Circulars, 1992). This allowed both Indian and foreign carri-
ers to operate cargo services (scheduled and non-scheduled) to and from
Domestic passengers carried (in lakh)

90 80.68
80 73.79
70.82 69.99 69.46
67.46 68.94 65.38
70
60
50 56.92
40 48.93 49.08 49.14
41.35
30 36.1
20
20.92 Indian Airlines
10 Private Airlines
4.11
0
1992 1993 1994 1995 1996 1997 1998 1999
Source: Created using data from the website of the Directorate General of Civil Aviation.
Figure 6.1 Increase in air traffic of private airlines post-liberalization

M4371-FINGER_9781786431851_t.indd 95 14/11/2017 11:14

any airport in India, provided that custom/immigration services are avail-

able. Carriers could set their own cargo rates for major export commodi-
ties without regulatory control (International Civil Aviation Organisation,
2003). In 1994 the Indian Government enacted the Air Corporations Act
that led to the setting up of Indian Airlines Limited and Air India Limited,
registered under the Companies Act, 1956 (Parliament of India, 1994).
More importantly, this Act repealed the earlier Act of 1953, thus ena-
bling private carriers to operate scheduled services and publish timetables
subject to fulfilment of statutory requirements.
However, although the private sector was allowed to participate in civil
aviation, the carriers were not given a free hand. Restrictions were imposed
on the import of aircraft with 120 seats or more and access to airport ter-
minals. Airlines had to deploy their capacity equally on routes above and
below 700 kilometres (Hooper, 1997). Private operators also had to display
a sound financial position, a minimum paid-up capital ($8 million for using
aircraft with a take-off mass of 40 000 kg or more), a fleet of a minimum
of five aircraft and an adequate aircraft maintenance, repair and overhaul
facility certified by the DGCA. Given the capital-intensive nature of the
business, the requirement of a minimum paid-up capital and minimum
fleet of five aircraft for issuance of a new permit resulted in a high cost of
entry, restricting the entry of smaller players. This is unlike the regulation
in markets such as Australia, the European Union and the United States,
where potential market entrants require no paid-up capital or aircraft
fleet: only financial soundness, litigation and insurance policy informa-
tion is submitted to the authority (Nathan Economic Consulting India,
2012). Moreover, the Indian government mandated airlines to deploy a
portion of their seats on low-density routes with poor connectivity such
as the north-eastern states, Jammu and Kashmir and Andaman Islands
(Krishnan, 2008). This forced the airlines to operate a large network and
to maintain smaller aircrafts to cater to low-density routes, alongside the
wide-body Boeing 737s for regular routes. These regulations, combined
with high fuel costs (nearly two to three times the international average),
regulated fares and poor aviation infrastructure threatened the financial
viability of the airline companies. Consequently, private players like East
West airlines, Damania Airways, NEPC and ModiLuft ceased operations
between 1996 and 1997. Capital investment by foreign airlines in Indian
entities was prohibited, leading to divestment of the Gulf Air and Kuwait
Airways holding in Jet Airways.
Fortunately, in 1997 the government eased the barriers to entry by sim-
plifying the approval process, by scrutinizing applications to verify only
financial soundness, maintenance and safety standards (ASCI, 2009).
The restrictions on the type and seating capacity of the aircraft to be

M4371-FINGER_9781786431851_t.indd 96 14/11/2017 11:14

imported or acquired were also lifted. During this phase of deregula-

tion, the private airlines, operating as Air Taxis, added more capacity to
address the latent demand that could not be served by the public sector
Indian Airlines. Carriers such as Jet Airways and Sahara Airlines doubled
their fleet in the three years from 1996 to 1999 (Krishnan, 2008). Private
airline market share increased drastically, from 4.9 per cent in 1992 to 46.8
per cent in 1999. While the number of passengers grew at a compounded
annual growth rate of 5.38 per cent, that for private airlines grew at 56.4
per cent, since they began with a smaller base (Figure 6.1). While Air India
dominated the international air services, Jet Airways, Sahara and Indian
Airlines constituted more than 95 per cent of domestic air services. After
1992, Indian Airlines faced stiff competition from private carriers for the
first time in its existence, with its market share falling from 95.2 per cent in
1992 to 53.5 per cent in 1999, and it was compelled to streamline its opera-
tions, lower fares and revamp service offerings to include benefits such as
frequent flyer programmes.
The entry of the private carriers also led to a steady growth in the
amount of domestic cargo carried from 176 000 tonnes in 1992 to 259 000
tonnes in 1999. International air cargo increased from 300 000 tonnes to
420 000 tonnes during the same period.
ENTRY OF THE LOW-COST CARRIERS
The Indian economy witnessed a GDP growth rate of 6 per cent in the year
2000, which further propelled the demand for travel for business as well as
leisure. An important development in this context is the emergence of the
Indian middle class – a section of working professionals and businesspeo-
ple who fall between the upper and working classes. Sensing an opportu-
nity, Air Deccan entered the domestic market in 2003 in an attempt to take
air travel to the masses in India. This was based on the ‘low-cost’ model
of cutting out frills from airline operations, popularized by Southwest
Airlines in the US and Ryanair in Europe. The airline excluded services
such as free meals, limited the staff, offered point-to-point flights and sold
tickets only through the internet at throwaway fares. Competition increased
with the entry of low-cost carriers such as SpiceJet (2005), GoAir (2005)
and IndiGo (2006), with greater focus on cost and efficiency. Full-Service
Airlines (FSAs) like Kingfisher Airlines and Paramount started operations
in 2005. The advent of these private carriers brought down ticket fares
and helped not only to increase the frequency of travel, but also to convert
train and bus travellers to airline passengers (ASCI, 2009). The combined
fleet of scheduled carriers grew from 162 to 402 from 2004 to 2015, and

M4371-FINGER_9781786431851_t.indd 97 14/11/2017 11:14

Others, 1.2
Spicejet, 15.2 Air India, 17.2
Air Costa, 0.8

Jetlite, 3.6
Go Air, 9.3
Jet Airways,
15.4
Indigo, 37.2
Figure 6.2 Market share of airlines in India, 2014–15
the number of aircraft departures increased from 270 031 to 729 283 during

the same period. Figure 6.2 shows the relative domestic market share (cal-
culated using revenue passenger kilometres) of scheduled airline carriers
in 2015. This shows that most of the market is being operated by low-cost
carriers; the two major FSAs, Air India and Jet Airways, together consti-
tute less than 33 per cent of the market share.
India and the US inked an open skies agreement in 2005, lifting restric-
tions placed on destinations, aircrafts and pricing and providing for seam-
less code sharing between Indian and US carriers (Indian Express, 2014).
This helped to increase traffic by 85 per cent on the Indo-US route, from
447 000 in 2005 to 827 000 in 2007 in just three years (Economic Times,
2007). Between 2005 and 2007, India inked several air service agreements
with different emirates of the UAE and countries with business and
tourist interest such as Singapore, Italy, Thailand, Japan, New Zealand,
China and France (Directorate General of Civil Aviation India, 2008).
The number of international passengers travelling to and from India saw
a drastic increase, from approximately 14 million in 2004 to more than 43
million in 2014. India also signed a Horizontal Aviation Agreement with
the European Union at Marseilles in September 2008, bringing bilateral
agreements with 26 member states into conformity with EU law and thus
encouraging more airlines to operate services between the two countries
(European Commission, 2008). In the decade since 2003, aircraft regis-

M4371-FINGER_9781786431851_t.indd 98 14/11/2017 11:14

tered outside India were exempted from levy of all taxes and duties on
fuel and lubricants (Aeronautical Information Circulars, 2003), domestic
fares were no longer subject to the Inland Air Travel Tax and international
passengers no longer had to pay a Foreign Travel Tax (The Hindu, 2004).
Along with an easing of international tourist charter norms, these poli-
cies gave much-needed impetus to a growing aviation sector. In 2007, the
government announced the regional airlines policy, which allowed airlines
to operate between airports of designated regions. This policy was aimed
at increasing connectivity to Tier II and III cities – these airlines had lower
fleet and paid-up requirement (Bangalore Aviation, 2008). Such regionally
focused airlines were allowed to operate between one metro and all other
non-metro airports within that region.
IMPACT OF AIRLINE LIBERALIZATION ON

LARGER ECONOMY
The growth of the airline industry in the last decade has brought far-flung
destinations closer and generated great value for consumers by provid-
ing speed, reliability and reach. Train journeys spanning two days were
brought down to a few hours at affordable costs, creating high-speed access
to business and holiday destinations. According to Oxford Economics
(2011), in 2009 the monetary value of consumer benefit to Indian citizens
was estimated at $17.2 billion from air travel and $0.64 billion from air
cargo. This figure could be much higher when the catalytic impact on
tourism, employment and manufacturing is taken into account. A by-
product of the increased consumer travel was the rise of e-commerce in air
ticket purchase, allowing customers to choose from all available ticketing
options and select the most convenient flight at the lowest cost – often
competitive relative to air-conditioned railway fares.
Passengers benefited from improved frequency of service between major
Indian cities and connectivity to economically important destinations
around the world. In 2010, Indian carriers operated more than 59 flights
per day between Delhi and Bombay, a major trunk route for business trav-
ellers, and eight flights per day between Delhi and Dubai, an important
Middle East destination. IndiGo airlines, a low-cost carrier that com-
menced operations in 2006, grew its network to 40 destinations in India
and abroad, with 679 daily flights by 2015. Other low-cost airlines such
as SpiceJet and GoAir have established a number of direct links between
metros and hitherto unconnected cities. Cheaper and faster movement of
people and goods has prompted Indian companies to expand their cus-
tomer base, exploit economies of scale and eventually earn higher revenues

M4371-FINGER_9781786431851_t.indd 99 14/11/2017 11:14

and returns from capital employed (National Transport Development

Policy Committee, 2013).
Oxford Economics, in an India country report, explained how ‘improved
connectivity enhances overall productivity of the wider economy through
increased access to foreign markets and the free movement of invest-
ment capital and workers between countries’ (Oxford Economics, 2011).
Manufacturers rely on air transport to operate just-in-time production
operations and increase flexibility within their supply chains. The report
values the direct contribution of the aviation sector in 2009 at $6.82 billion,
with $3.04 billion from airlines, airports and ground services, $2.21 billion
contributed through the sector’s supply chain and the rest through the
spending of employees of businesses which are directly or indirectly linked
to aviation. The sector directly or indirectly employs 1.7 million people
in airlines, airport services, ticketing, maintenance, air navigation and
aerospace manufacturing. Aviation firms also add to public finances in the
form of taxes levied on the profits and wages, and contribute to foreign
exchange earnings in the form of international tourism receipts.
The catalytic impact of the aviation sector that relates to spill-over eco-
nomic activity facilitated by tourism, flow of foreign funds and manufac-
turing supply chains constitutes 82 per cent of the sector’s contribution to
GDP (Air Transport Action Group, 2014). Tourism was one of the driving
forces behind air transport liberalization in the 1990s. After the subsequent
policy changes, this sector has seen rapid growth. Foreign tourist arrivals
in India grew from about 2.5 million in 2000 to 7.7 million in 2014. About
86 per cent of these tourists arrived by air (Ministry of Tourism, 2014).
India has an abundance of cultural and historical attractions and is a large,
diverse market. Regional airline connectivity played an important role in
the growth of domestic tourists from 168 million in 1998 to 1280 million
in 2014.
AVIATION MARKET POST-DEREGULATION
After deregulation, the Indian civil aviation market is best described as an

oligopoly. The market share of low-cost carriers has increased substan-
tially, increasing price competition between airlines. These new airlines
used the freedom to decide routes and set prices and to spur demand.
The end customer has been a beneficiary of the market changes that have
brought promotional offers, discounted fares and the gradual expansion of
the airline network to various parts of the country. To woo air travellers,
airlines such as Air Deccan and SpiceJet declared fares as low as about
$2 before surcharges and taxes, when full-service airline fares were close

M4371-FINGER_9781786431851_t.indd 100 14/11/2017 11:14

7
6
5
4
3
2
1
0
19 –96
19 –97
19 –98
19 –99
20 –00
20 –01
20 –02
20 –03
20 –04
20 –05
20 –06
20 –07
20 –08
20 –09
20 –10
20 –11
20 –12
20 –13
20 –14
5
–1
09
95
96
97
98
99
00
01
02
03
04
05
06
07
08
10
11
12
13
14
19
Operating revenue per RPK Operating expense per RPK
Figure 6.3 Operating revenue and expense per revenue passenger kilometre

(RPK) for scheduled Indian airlines
to $200. Air Deccan was the first airline to introduce dynamic pricing
(Business Standard, 2015), changing prices based on demand and duration
from the departure date. As the ticket prices decreased due to increasing
competition from 2005, and expenses increased due to high fuel and main-
tenance costs, the average operating revenue earned per revenue passenger
kilometre (RPK) became less than the average operating expense incurred
per RPK at an aggregate level for all airlines combined, after the entry of
Low Cost Carriers (LCCs) from 2005 to 2014 (Figure 6.3). This suggests
that airlines spent more than they earned to fly each passenger over a dis-
tance of one kilometre. Intensive price competition, expensive taxation,
price-sensitive customers and the global recession in 2008 were some of the
factors that contributed to continued losses for most airline operators. As a
result, many carriers struggled to remain financially viable and were either
shut down or merged with other airlines.
By the year 2013, three airlines – Kingfisher, Air Deccan (Kingfisher
Red) and Paramount – had shut shop, and no major airline had entered
the market since the launch of IndiGo in 2006. The government gave the
sector a ‘shot in the arm’ by permitting foreign airlines to hold up to 49 per
cent equity in Indian companies operating scheduled and non-scheduled
air transport services (Aeronautical Information Circulars, 2013). The
policy also allowed Indian carriers to access pilot training, maintenance,
repair and overhaul facilities operated by foreign airlines. Abu Dhabi-
based Etihad Airways picked up a 24 per cent stake in the financially
strained Jet Airways, and paid $70 million to buy Jet’s airport slots at

M4371-FINGER_9781786431851_t.indd 101 14/11/2017 11:14

London Heathrow Airport (The Hindu, 2013). Air Costa, a regional airline
that focused on connectivity between Tier II and Tier III cities, com-
menced operations in October 2013. The airline obtained a regional opera-
tor’s permit according to the policy announced in 2007. The Tata Group,
an industrial conglomerate, launched two start-up airlines, Tata AirAsia
and Tata SIA, and marked a return to the aviation business. AirAsia India
is a low-cost carrier focused on providing connectivity to Tier II and Tier
III cities, and Vistara (Tata–Singapore Airlines joint venture) is a premium
full-service carrier to cater to the demands of high-end business travellers.
The performance of these carriers in a competitive market against the
backdrop of changing regulation is yet to be seen. However, the success
of low-cost airline IndiGo, which managed to post consistent profits for a
continuous stretch of eight years, demonstrates the potential for exploit-
ing operational efficiencies despite the changing regulatory environment
(Saranga and Nagpal, 2016).
AIRPORT LIBERALIZATION: AIRPORT

INFRASTRUCTURE POLICY, 1997
The rapid expansion of the air transport network in India since 1991 has
been heavily dependent on airport infrastructure. The capacity of airports
can be a bottleneck for steady growth of passenger traffic. Although the
construction of civil airports began, way back in 1924, with airports in
Calcutta and Allahabad, most airstrips in India were used for military
aviation. After Indian independence, many military airfields known as
‘Aerodromes’ came to be used for civil aviation under the control of the
Civil Aviation Department (Mehta, 2011). However, it was only in 1995,
with the formation of the Airports Authority of India (AAI) through an
Act of Parliament, that the management, upgrading and maintenance of
the civil aviation infrastructure in India were given a major thrust. This Act
also enabled the AAI to allow private players to operate a subsidiary busi-
ness for developing, financing and operating airports managed by the AAI
(Ministry of Civil Aviation, 2003a). Today, the AAI manages 125 airports,
provides ground handling and air traffic management services and ensures
safety of aircraft operations.
The Airport Infrastructure Policy (Ministry of Civil Aviation, 1997)
announced in November 1997 envisaged the development of international
and regional hubs with world-class facilities to provide capacity ahead
of the burgeoning volume of air travel. The policy also conceived of the
infusion of private capital for building new airports or for the moderniza-
tion of existing ones. This was in line with the global transition towards

M4371-FINGER_9781786431851_t.indd 102 14/11/2017 11:14

corporatization and private management of airports from state-owned

monopolies. Foreign equity (including from airport authorities) up to
74 per cent was permitted with automatic approval, and up to 100 per
cent with government approval. The government sought to bring flex-
ibility in the current ownership structure through various models, such
as the build–own–transfer, build–own–operate and joint venture. Fiscal
incentives in the form of income tax deductions were proposed for those
involved in airport infrastructure projects. Private participation was also
encouraged in ground-handling services to streamline processes and to
reduce drastically the dwell time of passengers and cargo, in line with
international norms. Based on the level of threat, private security agencies
were allowed to operate at certain airports to have a passenger-friendly,
efficient, technically trained and commercially conscious force. Keeping
in view the high investment costs and long periods of return, the policy
gave freedom to operators to raise revenue from non-aeronautical sources
such as shopping arcades, restaurants and lounges. In 1999, for the first
time in India, public–private participation was used to build Cochin
International Airport Ltd with the help of loans from non-resident
Indians, donations from industrial bodies and the state government
(Varkkey and Raghuram, 2001). This provided essential infrastructure
for air travellers constituting tourists and Keralite families from the Gulf
region.
With the emergence of private carriers towards the end of the twen-
tieth century, India’s metro airports had a long way to go to bridge the
capacity gap created by increased passenger and cargo growth. Long
passenger queues at check-in terminals and long waiting hours for
immigration procedures were a common sight at Indian airports. The
government realized the need to expand the airport infrastructure in
the country and approved two Greenfield airports at Bangalore and
Hyderabad under public–private partnership on a build–own–operate
and transfer basis in 2004 (International Civil Aviation Organisation,
2015). This was followed by approval for the restructuring and mod-
ernization of the Delhi and Mumbai airports, to be performed jointly by
the AAI and a consortium of international bidders in 2006. In the same
year, the government also decided on the development of 35 non-metro
airports using private involvement for development of infrastructure on
a revenue sharing or land-lease basis (The Hindu, 2006). In 2008, the
government decided to lay down policy guidelines for the setting up of
Greenfield airports and adopted a liberal approach towards the setting
up of such airports.
As per the Aircraft Act, 1934, ‘the Central Government had the sole
right to grant a license for setting up an airport’ (Government of India,

M4371-FINGER_9781786431851_t.indd 103 14/11/2017 11:14

2008). However, as per the new policy, the DGCA was given authority
to grant licences for the operation of Greenfield airports beyond 150 km
from an existing civil aviation airport, ‘no prior central government per-
mission is required’ (iGovernment, 2013), provided the airport is compliant
with the Aircraft Rules, 1937. Greenfield airports set up by the AAI were
to be structured and financed with private participation via open com-
petitive bidding. The policy outlined incentives such as concessional land,
airport connectivity, development rights and fiscal exemptions provided by
respective state governments. For Greenfield airport projects, 100 per cent
FDI was permitted under automatic approval. However, functions such
as traffic control, security, customs and immigration would be managed
by the central agencies. The government has since then given ‘in-principle’
approval to 14 Greenfield airports in the country under public–private
partnership, made available the necessary land and in most cases the forest
and environmental clearances required for construction. The Airport
Economic Regulatory Authority was established in May 2009 by an Act
of Parliament and was responsible for fixing and regulating tariffs for
aeronautical services and user fees levied by the airport operator. The
Authority also keeps a check on the quality standards of amenities pro-
vided by airport developers.
PUBLIC–PRIVATE PARTNERSHIP AND

GREENFIELD AIRPORTS
According to the Centre for Asia Pacific Aviation (2014), as of 2014 the
AAI earned $1.7 billion in revenue share from public–private partnership
(PPP) airports. The major PPP airports at Delhi, Hyderabad, Mumbai and
Bangalore accounted for 53 per cent of passenger traffic during the finan-
cial year 2013–14. As shown in Figure 6.4, the passenger traffic since 2004
has seen a steady growth, and even faster growth rates after 2013, with the
introduction of Greenfield airports. Moreover, this has led to a dramatic
improvement in the flyer experience through modern amenities, state-of-
the-art infrastructure and efficient handling of passenger and cargo. As
per a report by the Working Group on Civil Aviation Sector (National
Transport Development Policy Committee, 2012), in the financial year
2011, 75 per cent of domestic and 72 per cent of international cargo was
handled by the joint-venture international airports with fewer infrastruc-
ture woes. The share of private sector investment in airport infrastructure
increased to 69.85 per cent in the 11th five-year plan (2007 to 2012) from
43.36 per cent in the 10th plan. Revenues generated at airports in Delhi and
Mumbai increased dramatically after their privatization. In the financial

M4371-FINGER_9781786431851_t.indd 104 14/11/2017 11:14

180 168.9
160
139.3
140 121.5 116.4 122.3
120 105.5
100 87.1 89.4
77.3
80 70.65
54.7
60 51.04
43 46.6 50.8
39.9 37.9 40.8
40 23.13 25.61 29.8 31.6 34.4
19.45 22.25
20
0 16.6
4
6
–0
–0
–0
–0
–0
–0
–1
–1
–1
–1
–1
–1
–1
03
04
05
06
07
08
09
10
11
12
13
14
15
20
20
20
20
20
20
20
20
20
20
20
20
20
International passenger traffic (million)
Domestic passenger traffic (million)
Figure 6.4 Passenger traffic in Indian airports
year 2014–15, close to half of the $321.14 million bottom-line came from
these airports (Swarajya, 2016).
Greenfield airports constructed through the PPP model will prove
essential in building connectivity in underserved Tier II and Tier III
cities. Expert private players can help to build new capacity in cities where
current airports are over-utilized. The government has decided to increase
the number of airports to 500 in the next 20 years, with 367 of these being
Greenfield airports. As part of the latest National Civil Aviation policy, the
government has proposed the development of ‘no-frills’ airports at desti-
nations with low air traffic in the central and north-eastern parts of the
country. Airlines operating from these airports will be given concessions
on landing and parking charges. To boost traffic at these airports the ticket
prices have been capped at about $50 for a one-hour flight.
AIR TRAFFIC MANAGEMENT
Air Traffic Control (ATC) forms an important component of airport infra-

structure to manage the flow of air traffic, prevent collisions and provide

M4371-FINGER_9781786431851_t.indd 105 14/11/2017 11:14

support to pilots. The Airports Authority of India provides communica-

tion, navigation and surveillance infrastructure to ensure safe civilian
operations at airports in India. The DGCA forms regulations and proce-
dures and issues directions concerning Aeronautical Telecommunication
facilities to be complied with by the Authority, airlines and airports.
India’s airspace covers about 9.5 million square kilometres, of which
35 per cent is reserved for military use (Centre for Asia Pacific Aviation,
2015). This poses some challenges for routings, as direct flights have to
circumvent the restricted areas. In 2015 the AAI, along with the Defence
Ministry, tested flight routes for the flexible use of airspace. If institution-
alized, this could result in significant savings in travel time, fuel costs and
emissions for airlines.
As airspace is a limited resource, it should be used efficiently to sustain
growth in air traffic. Performance-based navigation improves airspace
efficiency and safety, as it is required to conform to the specified level of
performance for the operation rather than the performance of the naviga-
tion signal or equipment (Directorate General of Civil Aviation India,
2012). Area navigation or RNAV permits aircraft operation on any desired
flight path when the aircraft is within the limit of station-referenced navi-
gation or self-contained aids (International Civil Aviation Organisation,
2014). In August 2002, multiple parallel RNAV-10 routes were imple-
mented under the EMARSSH project to improve safety and efficiency of
aircraft operations and Air Transport Management (ATM) services. The
standard vertical separation required between aircraft was reduced with
the implementation of Reduced Vertical Separation Minimum (RVSM) in
2003 (Airports Authority of India, 2009) in the Indian Flight Information
Region. This has resulted in higher capacity of the airspace by increasing
the number of aircraft that can safely fly in a given volume of airspace,
provided the aircraft has specially certified altimeters and autopilots. To
facilitate timely seamless ATM throughout the Asia Pacific region, the
International Civil Aviation Organisation (ICAO) constituted the Asia
Pacific Seamless ATM Planning Group in 2011 (Airports Authority of
India, 2014). The air navigation systems strategic plan outlines the com-
mission of planned infrastructure in a timely manner to meet the demand
of rising air traffic while ensuring safety over the Indian continental
airspace. India became the fourth country to offer satellite navigation
services, with the launch of the GPS-Aided Geo Augmented Navigation
system in 2015 (The Hindu, 2015). This system is expected to benefit 50
airports in India, with withdrawal of ground aids, precision guidance for
landing and increased capacity by reducing separation between aircraft.
In a booming airline market, India could borrow from countries such as
the US and China with regard to implementation of Performance-based

M4371-FINGER_9781786431851_t.indd 106 14/11/2017 11:14

Navigation and Flexible Use of Airspace. The government should consider

corporatization of air traffic control and navigation facilities by involving
private participation in activities such as setting up training centres for Air
Traffic Control Officers to address the current shortage of trained person-
nel. In view of increasing air traffic and privatization of Indian airports,
India might have to restructure ATC services to unbundle ATC from
privatized airport services, modernize infrastructure and bring flexibility
in procurement decisions. The Naresh Chandra Committee (Ministry of
Civil Aviation, 2003b) proposed that ATC services should be provided
by a separate corporate entity (or joint venture) and separated from the
control of AAI, in line with international trends. Technological upgrading
of air control and navigation is a dynamic process, and AAI will have to
be proactive in ensuring that the growth in air traffic is met with sustained
improvement in infrastructure in the en-route and terminal airspace.
ENVIRONMENTAL IMPACT
Just as every creation is accompanied by destruction of nature, the devel-

opment of airports and increase in aircrafts has had adverse impacts on
agricultural land, forests, water and air. According to the DGCA’s carbon
footprint report (Directorate General of Civil Aviation India, 2013), in
2013, the CO2 generated by the Indian and foreign passenger movement to
and from domestic and foreign destinations was 15.6 million tons, a 1.57
per cent increase from 2012. This increase follows the growth in passenger
numbers between the two years. The CO2 emission (in kg) per revenue
transport kilometre of 0.96 is marginally higher than the global average.
According to the same report, 95 to 98 per cent of aviation’s emissions are
emitted as a result of the burning of aviation fuel, and the remainder is
from airport activities. Airports emitted 0.78 million tons of CO2 in 2013,
with electricity consumption being the major source of airport emissions.
Destruction of land for construction of Greenfield airports leads to loss
of habitat and reduction in biodiversity. Particulate matter in the air is
increased due to the burning of worn-out tyres during landing and take-
off (Anushua, 2015). If unchecked, emissions from the aviation sector are
projected to reach 29 million tonnes by 2020.
According to the Aviation Environment Circular 2 of 2013 (DGCA
Technical Centre, 2013), scheduled airlines and large airports will set up
their own Aviation Environment Cell, develop their own carbon footprint,
monitor emission trends and submit data to the DGCA on an annual
basis. Airlines such as Jet Airways, IndiGo and SpiceJet have put in place
fuel management and weight reduction systems to optimize aviation fuel

M4371-FINGER_9781786431851_t.indd 107 14/11/2017 11:14

usage. Joint venture airports at Bangalore, Delhi, Hyderabad and Mumbai

have taken the lead to participate in the Airport Carbon Accreditation – a
carbon management certification standard for airports. While the avia-
tion sector brings economic and social benefits, sustainable growth can be
achieved only through cautious consideration of limited environmental
resources.
FUTURE OUTLOOK
It is beyond doubt that the Indian airline industry has seen tremendous
growth in passenger traffic in the last 30 years. The first wave of liber-
alization in the 1990s brought back private players to the Indian skies,
and the twenty-first century saw the entry of low-cost carriers. Today,
passengers are beneficiaries of the price wars between various public and
private sector airlines and a wider airline network. This is largely due to
privatization and globalization of the airline industry. However, to unleash
the growth potential of the market, enhance the ease of doing profitable
business, build a global aviation hub and make air travel a reality for the
common man in India, positive regulatory measures will have to be revised
and made relevant to the current scenario.
The National Civil Aviation Policy, 2016 (PIB, 2016) outlines steps in
this regard. The policy scrapped the 5/20 rule, allowing start-up airlines
to fly to international destinations without requiring them to be operating
in the domestic circuit for a minimum of five years, provided 20 aircraft
or 20 per cent of capacity (whichever is higher) is utilized for domestic
operations. This allows new airlines to provide services on profitable inter-
national routes and gives Indian travellers more options to fly abroad.
The government will enter into reciprocal open sky agreements with coun-
tries located beyond 5000 km from Delhi and remove restrictions on the
number of seats, flights and airports.
Additionally, regional connectivity has been given a boost by capping
the charge for a one-hour flight at Rs. 2500 (approx. 50 USD) on regional
routes (North East, Jammu and Kashmir, etc.). Liberal measures include
reduced airport charges, service tax, value added tax and excise duty on
fuel used on these routes. The government will also provide financial assis-
tance to carriers flying on unprofitable regional routes.
While the Indian government has enabled private participation in
airport infrastructure, the growth in capacity seems to be lagging behind
the demand. Major airports at Mumbai, Delhi and Bangalore continue
to struggle with passenger traffic and over-utilized runways. The Policy
envisages development of additional airports through the PPP route

M4371-FINGER_9781786431851_t.indd 108 14/11/2017 11:14

and increase in non-aeronautical revenue to cross-subsidize aeronautical

charges. The government needs to take a long-term view and come up with
a strategy spread over the next couple of decades to ensure airport capac-
ity grows ahead of the rising demand in India. In the short to medium
term, there are efforts to improve infrastructure and reduce operational
inefficiencies stemming from policy-related hurdles. For example, plans
are under way to open alternative airports at metropolitan cities such
as Mumbai, where current airports are saturated. Domestic scheduled
airlines will be allowed to carry out self-handling at all airports, bringing
efficiency in ground handling and faster processing of passengers. An
incentive structure is being put in place to bring maintenance, repair and
overhaul – which is mostly carried out overseas – into India by providing
incentives in the form of reduced VAT, airport charges, customs duty and
ensuring easy availability of land for such activities.
The process of air transport liberalization in India has fostered growth
in air traffic and has led to the entry of new players into the market. Air
travel has also helped to boost tourism and generate employment while
bringing investment from foreign sources. While many of the current
players make huge losses in the course of their operations, an open
market with fewer restrictions on pricing, air fuel and route decisions will
enable operators to achieve maximum efficiency. Policy changes aimed
at increasing airport capacity, facilitating maintenance, repair and over-
haul services and generating investment in infrastructure will bring more
private players on board. Low-cost airlines will play a key role in making
air travel accessible to the masses as well as contributing towards effective
capacity utilization. Liberal policies in line with international benchmarks
can help India leverage its strategic location and become a global aviation
hub.
REFERENCES
Administrative Staff College of India (ASCI) (2009), Competition Issues in the Air
Transport Sector in India, Khairatabad.
Aeronautical Information Circulars (1992), AIC No. 18/1992, Director General of
Civil Aviation, Government of India, New Delhi.
Aeronautical Information Circulars (2003), AIC No. 03/2003, Director General
of Civil Aviation, Government of India, New Delhi.
Aeronautical Information Circulars (2013), AIC No. 12/2013, Director General of
Civil Aviation, Government of India, New Delhi.
Air Transport Action Group (2014), Aviation Benefits beyond Borders, Geneva.
Air Transport Inquiry Committee (ATIC) (1950), Report of the Air Transport
Inquiry Committee, University of California.

M4371-FINGER_9781786431851_t.indd 109 14/11/2017 11:14

Airports Authority of India (2009), Performance Based Navigation (PBN)

Implementation Road Map Version 1.0, Directorate of Air Traffic Management,
New Delhi.
Airports Authority of India (2014), Strategic Plan for Air Navigation Services
2014–2018, New Delhi.
Anushua (2015), Eco-friendly airports in India – treat to eco-conscious travelers,
Green Clean Guide.
Association of Private Airport Operators (2013), Chronology of Events of Indian
Civil Aviation Sector, New Delhi.
Aviation Safety Network (1990), Indian Airlines – Accidents and Incidents,
Alexandria: Flight Safety Foundation.
Bangalore Aviation (2008), Regional airlines take to Indian skies, 21 May.
Business Standard (2015), 40 years ago ... and now: Air travel – fixed fares to
dynamic pricing, 8 February.
Centre for Asia Pacific Aviation (2014), India Airports Public Private Partnership
Model is Transformational but Key Lessons to be Learned, Sydney.
Centre for Asia Pacific Aviation (2015), Unique CAPA Report: The ATM System
and AAI’s Plans for Integration, Sydney.
Directorate General of Civil Aviation India (2008), Bilateral Air Service Agreements,
New Delhi.
Directorate General of Civil Aviation India (2012), Civil Aviation
Requirement, Section 8 – Aircraft Operations, Series ‘O’, Part VI, Issue I, New
Delhi.
Directorate General of Civil Aviation India (2013), Carbon Footprint of Indian
Aviation 2013, New Delhi.
Directorate General of Civil Aviation (DGCA) Technical Centre (2013), Aviation
Environment Circular 2 of 2013, New Delhi.
Economic Times (2007), Government to open skies further, 12 November.
European Commission (2008), EU–India summit signs civil aviation agreement,
European Commission press release database, Brussels.
Government of India (2008), Greenfields Airports Policy, New Delhi: Government
of India.
Hooper, P. (1997), Liberalisation of the airline industry in India, Journal of Air
Transport Management, 3, 115–23.
iGovernment (2013), India approves Greenfield airports policy, 12 December.
Indian Express (2014), In search of open skies, 1 March.
International Civil Aviation Organisation (2003), India’s Open Sky Policy on Air
Cargo, Montreal.
International Civil Aviation Organisation (2014), Performance-based navigation,
frequently asked questions, Montreal.
International Civil Aviation Organisation (2015), Public Private Partnership (PPP)
– Case Study, Geneva.
Joshi, V. and I.M.D. Little (1996), India’s Economic Reforms, 1991–2001, Oxford:
Oxford University Press.
Krishnan, R.T. (2008), The Indian airline industry in 2008, Working Paper series,
IIM Bangalore.
Mehta, D.C. (2011), 100 Years of Aviation in India, Delhi: Airports Authority of
India.
Ministry of Civil Aviation (MoCA) (1997), Policy on Airport Infrastructure, New
Delhi.

M4371-FINGER_9781786431851_t.indd 110 14/11/2017 11:14

Ministry of Civil Aviation (MoCA) (2003a), The Airports Authority of India Act,
1994, New Delhi.
Ministry of Civil Aviation (MoCA) (2003b), Report of the Committee on a Road
Map for the Civil Aviation Sector, New Delhi.
Ministry of Tourism, Government of India (2014), India Tourism Statistics, New
Delhi.
Nathan Economic Consulting India (2012), Research Study of the Civil Aviation
Sector in India, Chennai.
National Transport Development Policy Committee (2012), Report of Working
Group on Civil Aviation Sector, New Delhi: Ministry of Civil Aviation.
National Transport Development Policy Committee (2013), Civil Aviation, New
Delhi.
Oxford Economics (2011), Economic Benefits from Air Transport in India, Oxford.
Parliament of India (1994), The Air Corporations (Transfer of Undertakings and
Repeal) Bill, 1994, New Delhi.
Press Information Bureau (PIB) (2016), National Civil Aviation Policy, 2016:
Salient Features, 15 June.
Saranga, H. and R. Nagpal (2016), Drivers of operational efficiency and their
impact on market performance in the Indian airline industry, Journal of Air
Swarajya (2016), FDI alone won’t help – major restructuring needed for AAI take-
off, 22 June.
The Hindu (2004), Air traffic boom expected from Kochi, 11 January.
The Hindu (2006), Flying high, 5 November.
The Hindu (2013), A year of surprises for aviation industry, 29 December.
The Hindu (2015), Satellite-based navigation system launched, 14 July.
Varkkey, B. and G. Raghuram (2001), Public private partnership in airport develop-
ment: governance and risk management implications from Cochin International
Airport Ltd, India Infrastructure Report 2002: Towards Better Governance for
Commercialization, Oxford: Oxford University Press.

M4371-FINGER_9781786431851_t.indd 111 14/11/2017 11:14

7. Air transport development:
a comparative analysis of
China and India
Yahua Zhang and Anming Zhang
INTRODUCTION
Since 2007 China has been the second largest aviation market in the world
(after the United States) in terms of volume of passengers and air cargo
moved in its domestic market. In 2015 the whole industry handled 435.6
million passengers and 6.3 million tons of air cargo, an increase of 11.1
per cent and 5.2 per cent from the previous year. China’s airline market is
a growing market underpinned by a huge population and rapid economic
growth. The International Air Transport Association (IATA) forecasts
that China will overtake the US as the largest air passenger market by
around 2030 as measured by traffic to, from and within a country.
India’s civil aviation industry is also on a high-growth trajectory.
Passenger traffic in 2015 increased from 67.4 million in 2014 to 81.1
million the next year, at a rate of 20.3 per cent. It is expected that India
will become the third largest aviation market by 2020. Both India and
China started the deregulation process in the aviation sector in the 1980s.
However, 30 years on, the outcomes of the two aviation markets are stun-
ningly different. In particular, in China, most of the aviation reforms have
put the interests of the state-owned airlines first, which results in the four
state-owned airline groups dominating the domestic market with an aggre-
gate market share of 90 per cent in 2015. In contrast, in India the private
airlines commanded a market share of more than 80 per cent in recent
years, with low-cost carriers (LCCs) commanding a market share of about
60 per cent.
This chapter attempts to outline the differences between the two markets
and evaluate airline efficiencies in the two countries, which will help build
the momentum for further reforms in the aviation sector.
112

M4371-FINGER_9781786431851_t.indd 112 14/11/2017 11:14

Air transport development: China and India compared 113
LITERATURE REVIEW
China and India are two of the fastest-growing economies in the world,
operating under very different political systems. India is the world’s largest
democracy, while China fails to develop Western-style democratic institu-
tions and practices, but has been pursuing the establishment of a market
economy. Comparative studies between China and India are abundant in
academic research and even more popular in public media, given that they
are the two most populated countries in the world (Liu and Jayakar, 2012).
There has been much research examining the air transport develop-
ment in the two countries, including the topics of aviation reforms and
their impacts. In the case of China, a pioneer work by Zhang (1998) first
described China’s airline deregulation process and documented its impact
on traffic growth, market structure, and airline operation and competi-
tion. This study was extended by Zhang and Chen (2003), who argued
that any reforms targeting the following four constraints would acceler-
ate the air traffic growth in China: (1) the protection of non-competitive
state-owned airlines; (2) low-income and restrictive policy on its citizens
travelling abroad; (3) underdeveloped infrastructure including airport
capacity constraints and air traffic control inefficiencies; and (4) the
lack of human capital and management skills. Zhang and Round (2008)
provided a detailed analysis on the driving forces that led to the airline
consolidations in China’s aviation markets in 2002, and argued that the
aviation market concentration is a natural response to the uncertainties
and changes that accompanied airline deregulation in China. Wang et al.
(2016) also agreed that the aviation market has become more concentrated
as a result of airline deregulation. Airline concentration and market power
issues in China’s airline market were further studied by Zhang and Round
(2009, 2011) and Zhang (2012). The general conclusion is that airline con-
solidations in the early 2000s did not grant China’s three airline groups
(China Eastern, China Southern and Air China) any significant market
power in both the short and longer term, largely owing to the implementa-
tion of other forms of deregulation in the last two decades – including the
relaxation of entry to and exit from markets, inviting private capital into
the aviation sector and the emergence of LCCs. However, after examining
the merger case between China Eastern and Shanghai Airlines in 2009,
which resulted in a rise in departure-day prices by 22 per cent, Zhang
(2015) warned that China’s anti-trust authorities should remain vigilant in
handling airline mergers when numerous parallel routes are involved. The
market power issue was also noted in Zhang et al. (2014), whose Lerner
indices for the period 2010–11 confirmed the existence of a certain degree
of market power in China’s airline markets.

M4371-FINGER_9781786431851_t.indd 113 14/11/2017 11:14

Yuen and Zhang (2009) investigated China’s airport policy reforms,

including partial privatization and public listing, and revealed their posi-
tive impacts on the airport efficiency and productivity. Fan et al. (2014)
assessed China’s airport performance following privatization, commer-
cialization and localization. They discovered that overall average efficien-
cies of China’s 20 major airports steadily improved during the 2006–09
period, and the operating performance of international hub airports was
significantly superior to that of non-hub airports. Fu et al. (2015a) noted
that there is a two-way relationship between government policy and market
performance in China. On the one hand, air transport reforms have been
constantly adjusted in the last two decades to reflect the changes in macro-
economic environment and industry conditions; on the other hand, these
reforms have had a significant influence on the market equilibrium and
substantially shaped the landscape of the industry.
Wang et al. (2014) evaluated the performance of China’s major airlines
between 2001 and 2010. Chinese airlines’ performance indicators, includ-
ing productivity, yield, cost competitiveness and input prices were bench-
marked against major international airlines. The findings showed that
Chinese airlines’ productivity has improved significantly over time, but
still lags behind that of the world’s leading airlines. Although Chinese air-
lines enjoyed higher profitability in recent years due to lower input prices,
further liberalization in this industry is needed to sustain their long-term
growth.
The emergence of LCCs and high-speed rail (HSR) services has posed
a threat to China’s airline industry. Zhang and Lu (2013) examined pas-
senger traffic flows on the routes in and out of Shanghai, and their study
found that the presence of Spring Airlines – an LCC in China – on a
domestic route has contributed to an increase in passenger volume by 23
per cent, other factors holding constant. The role of LCCs in promoting
domestic traffic flows was further confirmed in Zhang and Zhang (2016),
although the LCC market share is still relatively small. Fu et al. (2015b)
also noted that China’s LCCs are still not a game changer and their entry
decision was not affected by the competition from full-service airlines or
HSR services. Fu et al. (2012) showed that the HSR will be more competi-
tive in terms of network connectivity, travel time and cost efficiency, which
will force Chinese airlines to consider developing hub-and-spoke networks
and increase their competitiveness in international markets to achieve
sustainable growth in the future.1 The suppressing effect of the presence
of HSR on airline city-pair traffic could, on average, be as high as 53 per
cent in China’s aviation market (Zhang and Zhang, 2016). In extreme
cases, the launch of HSR services have resulted in the suspension of air
services on some routes. Wan et al. (2016) investigated, based on panel

M4371-FINGER_9781786431851_t.indd 114 14/11/2017 11:14

data, the impact of the commencement of HSR services on airlines’ avail-

able domestic seats on affected routes in China. By using the propensity
score matching method to pair HSR affected routes with routes without
HSR services, they found strong negative impacts on short-haul routes, but
that medium-haul routes would lose seat capacity only in the second year
after HSR entries; they also found that there was little impact on long-haul
routes.
Case studies on India’s aviation sector can be found in Hooper (1998),
Saraswati (2001), Findlay and Goldstein (2004), O’Connell and Williams
(2006), Jain and Natarajan (2015) and Saranga and Nagpal (2016). The
earlier works described the inconsistent and cumbersome aviation policies
in India and highlighted the directions for change. Recent studies, such as
Jain and Natarajan (2015) and Saranga and Nagpal (2016), focused on the
efficiency analysis for Indian airlines. Hooper (1998) first documented the
changes in the regulatory system and analysed the strategies adopted by
Indian airlines. The author concluded that inappropriate aviation policies,
such as the requirement for airlines to allocate their capacity on a mix of
profitable and unprofitable routes, have constrained the development of
the industry. Hooper (1998) further noted that government guidelines were
often changed at short notice, which created an uncertain environment in
which the industry operated. Saraswati (2001) observed inconsistent poli-
cies in India’s aviation market and argued that political and bureaucratic
hurdles impeded the privatization process in this sector. For example,
Findlay and Goldstein (2004) pointed out that the bureaucratic processes
in seeking strategic partners for India’s national airlines turned out to be
lengthy, inefficient and a failure. Taneja (2004) held a similar view. Zhang
and Findlay (2010) also mentioned that the government had long planned
to privatize the national airline, but this goal has never been achieved mainly
because for political reasons and because of opposition from trade unions.
The entry of the private carriers with better service and on-time perfor-
mance posed a serious challenge to Indian Airlines’ dominant status in the
domestic market (O’Connell and Williams, 2006). The government-owned
Indian Airlines lost not only consumers to their private counterparts, but
also a great number of pilots and engineers, who received better pay for
working with the new airlines. Mazumdar (2008) claimed that deregula-
tion in the aviation sector did not bring an immediate benefit to Air India,
which mainly provided service in the international market.
The effect of the presence of LCCs was first examined by O’Connell
and Williams (2006), who noted that LCCs have also made a significant
contribution to lowering airfares in India. India’s first LCC, Air Deccan,
was established in 2003 and provided many low-income Indians with the
first opportunity to fly. Its rapid growth has sparked the entry of many

M4371-FINGER_9781786431851_t.indd 115 14/11/2017 11:14

other low-cost carriers in India including SpiceJet, IndiGo, GoAir and

Paramount. Jain and Natarajan (2015) investigated the technical and scale
efficiencies of the airlines operating in India across service types, sizes and
ownership structures. The authors reported that newly established LCCs
were more efficient than the full-service airlines with private ownership.
Even though all the public sectors suffered losses during the period under
study, they were relatively efficient. Scheraga (2004) contended that relative
technical efficiency does not imply superior financial performance. Saranga
and Nagpal (2016) argued that the Indian airline industry is characterized
mainly as a low-cost industry; their study reports that the national carrier
Air India was among the most technically efficient airlines during 2005–07,
but both technical and cost efficiency dropped after the 2007 merger
between Air India and Indian Airlines. The technical efficiency scores of
the LCCs such as SpiceJet, Go Air and IndiGo were consistently high and
close to the frontier, but the cost efficiency scores were comparatively low
for many LCCs, except for Air India Express, Alliance Air and IndiGo.
We can identify two gaps from the brief literature review above. First,
there has been no scholarly literature comparing the development of air
transport policies and market outcomes for China and India. Such com-
parison will be of value to both regulators and operators in this industry,
given the many similarities between the two nations and the challenges
and opportunities that the air transport sector currently faces. Second,
although the performance of Chinese and Indian airlines as measured by
efficiency and productivity has been assessed in different studies, direct
comparison of their performance in one study is still lacking. This chapter
aims to fill these two gaps.
This research is organized as follows. In the next section, the evolution
of the air transport industry in China and India will be discussed. The
following section will compare the outcomes of the two markets, includ-
ing the efficiency performance of the airlines in the two countries and the
policy environment in which they operate. The last section presents the
conclusions.
THE EVOLUTION OF THE AIRLINE INDUSTRY IN

CHINA AND INDIA
The Case of China
Before 1980, China was politically and economically isolated from the
outside world. The airline industry was a semi-military organization under
the dual leadership of the Air Force and the State Council, engaging in

M4371-FINGER_9781786431851_t.indd 116 14/11/2017 11:14

only limited commercial operations. The Civil Aviation Administration of

China (CAAC) not only acted as a government department and a regula-
tor of the civil aviation, but also as the operator of airline and airport.
The combination of government function and enterprise function existed
in many Chinese industries during the period of the centrally planned
economy. The airline industry was regulated in every aspect of air services
provision, including market entry and exit, frequency, pricing, aircraft
purchase and even who was eligible to fly (Zhang, 1998).
The market-oriented reforms beginning in the late 1970s sought to sepa-
rate government function from enterprise function, and this policy guided
the industry evolution in the following three decades (Le, 1997; Zhang,
1998; Zhang and Round, 2008). In 1980, the CAAC was separated from
the Air Force and placed under the direct supervision of the State Council.
In the late 1980s, the CAAC’s governmental, administrative and regula-
tory role was separated from the direct management of the day-to-day
operations of commercial airlines and airports. By the mid-1980s, ordinary
citizens were allowed to use air services as long as they could afford the
airfare, which resulted in high demand for air travel, putting tremendous
pressure on the CAAC’s infrastructure and internal systems of operation
(Dougan, 2002).
The high demand for air services and the ever-increasing financial loss
of CAAC led to the dissociation of the governmental and company func-
tions. Between 1987 and 1991, six trunk airlines based in the regional
capital cities were established: Air China (based in Beijing), China Eastern
(Shanghai), China Northwest (Xi’an), China Northern (Shenyang), China
Southwest (Chengdu) and China Southern (Guangzhou). There was a
growth in the number of local airlines during this period, which were
usually established by local governments or jointly with the CAAC. Some
of the local airlines established in this period – such as Xiamen Airlines
and Hainan Airlines – are still in operation today, but most others have
ceased to exist in the last two decades, the majority being taken over by the
major airlines.
Until 1997 there was only limited competition between Chinese airlines
as the government still regulated the airfare and market entry. The relaxa-
tion of the Chinese government’s control of airfares in 1997 might have
been the most influential act in shifting the way that the carriers compete.
As a result of this deregulation, airfares became cheaper and more flex-
ible. From 1997 on, Chinese airlines experienced strong challenges from
both home and abroad, including continuous price wars in the domestic
markets, and constantly lost customers to aggressive foreign airlines that
had better services and safety records in the international markets. As
a result, most Chinese airlines suffered huge financial losses in the late

M4371-FINGER_9781786431851_t.indd 117 14/11/2017 11:14

1990s and early 2000s (Zhang and Round, 2008). This has led to the
airline consolidations that occurred in October 2002. China’s nine state-
owned airlines merged into three airline groups: the Air China Group,
the China Eastern Group and the China Southern Group. These mergers
faced no anti-trust challenge at the time due to the absence of any effective
anti-trust laws. Under a constant expansion policy, Hainan Airlines had
become the fourth largest group in China in the early 2000s.
In the years following the 2002 consolidations, competition in the
markets associated with Beijing, Guangzhou and Shanghai remained
strong; these markets are of critical importance to Chinese airlines due
to their large market potential (Fu et al., 2015b; Zhang et al., 2014). The
routes in and out of these three cities are a significant source of revenue
for almost all domestic airlines. Airlines never stopped increasing frequen-
cies and launching new routes to these cities. As a result, price wars broke
out regularly, which was one of the reasons for the merger between China
Eastern and Shanghai Airlines in 2009: both airlines were in deep financial
distress prior to the merger (Zhang, 2015).
Entry to and exit from a route for existing airlines has been easier after
2006. Apart from the routes associated with the eight busiest airports,
prior approval for entry and exit is no longer required: a simple notifica-
tion by an airline to the regional civil aviation bureau prior to an action
is sufficient (Zhang and Round, 2008). Further deregulation of the route
control occurred in 2010, with prior approval only being needed when
airlines operate a flight in and out of Beijing, Shanghai and Guangzhou.
Although airfares had been largely deregulated since the late 1990s, to
prevent ‘irrational’ pricing behaviour the government promulgated ‘The
Scheme of Domestic Airfare Reform’ in 2004. This reform scheme set a
benchmark price (0.75 yuan per kilometre)2 and imposed a price ceiling
and a price floor for domestic airfares. However, from the day that this
scheme took effect, the price floor limit was largely ignored and competi-
tion in prices remained strong as the state-owned airlines had gained more
autonomy and power in doing business. There were no effective measures
to enforce the pricing rules. The government realized this and publicly
noted in 2006 that the price control would be abandoned in the next few
years. The 2004 scheme and all restrictions on pricing were formally abol-
ished in 2013.
China’s private airlines emerged in 2005, with more than ten being
licensed in that year following the release of the ‘Regulation on Domestic
Investment on Civil Aviation’ in 2004, which allowed private sector
participation in the civil aviation industry – including setting up new
airlines. Shortly prior to this, the limit on foreign ownership participa-
tion in Chinese airlines was raised from 35 per cent to 49 per cent. As a

M4371-FINGER_9781786431851_t.indd 118 14/11/2017 11:14

result of these liberalization moves, three private carriers (Okay Airways

based in Tianjin, United Eagle Airlines in Chengdu and Spring Airlines
in Shanghai) launched their maiden flights in 2005. Spring Airlines posi-
tioned itself as an LCC and believed that this model would help them
secure a slice of the market dominated by their state-owned counterparts
(Zhang and Lu, 2013). In 2006 two other private carriers (Juneyao in
Shanghai and East Star Airlines in Wuhan) began operating. By 2007
some 20 new private airlines had been approved in China. The expansion
in the number of new airlines led to the Civil Aviation Authority’s decision
to suspend approval of new domestic entrants until 2010. Given safety
concerns following the crash of an aircraft of a local airline in 2010, the
government extended this policy until 2013. Subsequently, another wave of
private airlines has emerged since 2013: more than ten have been approved,
including Ruili Airlines, Zhejiang Loong Airlines and Qingdao Airlines,
and more are in the application process.
Some of the private airlines established in the first wave around 2005
quickly failed due to the lack of experienced pilots and skilled personnel,
along with the high costs and taxes associated with aircraft purchases, jet
fuel and airport charges (Zhang et al., 2008; Zhang and Lu, 2013). For
example, in 2009 East Star Airlines entered bankruptcy, and financially
troubled United Eagle Airlines was taken over by state-owned Sichuan
Airlines. Spring Airlines has been one of the most successful private air-
lines, and has achieved profitability from the start (Zhang et al., 2008).
It maintains a high load factor of 95 per cent, well above the industry
average of 70 per cent. Strict cost control has been the key to its success.
For example, by using its own computer reservation system and online
sales, sales costs are 82 per cent lower than those of traditional airlines
(Li, 2014). In 2015, Spring launched an initial public offering (IPO) on the
Shanghai Stock Exchange to finance its rapid fleet expansion.
The policy of relaxing the restrictions on market access and the estab-
lishment of new carriers around 2005–06 also encouraged state-owned
airlines to expand their capacities. Record plane orders were placed in
2005 and 2006. At the end of 2005 the whole industry only had 863 air-
craft, while in 2015 this number had increased to 2650. At the end of 2015,
China had 55 commercial airline companies, including 41 state-owned
airlines and 14 privately-owned or controlled airlines, with Air China,
China Eastern, China Southern and Hainan Airlines and their subsidiar-
ies carrying about 88.4 per cent of the traffic measured in ton-kilometres.
Table 7.1 summarizes the major milestones in the development of China’s
aviation industry.

M4371-FINGER_9781786431851_t.indd 119 14/11/2017 11:14

Table 7.1 Evolution of China’s civil aviation industry
Year Major Milestones

1949 Civil aviation came under the leadership of the Air Force, the Central Military
Commission of China.
1962 Civil aviation came under the dual leadership of the State Council and Air
Force, the Central Military Commission of China.
1980 Civil aviation was separated from the Air Force and operated in the name of
CAAC (Civil Aviation Administration of China).
1987 The State Council ratified the ‘Report on Civil Aviation Reform Measures and
Implementation’, and separated the CAAC’s government, administrative and
regulatory roles from the direct management of the day-to-day operations of
commercial airlines and airports.
1988–1991 Six trunk airlines based in regional capital cities established: Air China (based
in Beijing), China Eastern (Shanghai), China Northwest (Xi’an), China
Northern (Shenyang), China Southwest (Chengdu) and China Southern
(Guangzhou).
1993 Hainan Airlines launched commercial services.
1995 Civil Aviation Law of China was enacted.
1997 China Eastern was partly privatized and went public on the Hong Kong Stock
Exchange.
1997 Price discrimination towards non-Chinese citizens was removed and the same
price applied to all passengers. CAAC promulgated the policy of ‘one class with
multiple discounts’, allowing airfare discounts by airlines.
1998 CAAC issued an urgent notice to airlines, capping the airfare discount at 20 per
cent.
1999 Shanghai Pudong International Airport was put into use.
2000 CAAC forced 25 domestic airlines to participate in the ‘revenue pooling’
programme, an explicit price-fixing agreement, to avoid destructive
competition.
2002 CAAC repealed the ‘revenue pooling’ programme. Nine airlines were regrouped
into the ‘big three’: Air China Group, China Eastern Group and China
Southern Group. ‘New Regulations for Foreign Investment in the Civil Aviation
Industry’ came into effect, allowing (up to) 49 per cent foreign ownership of
Chinese airports and airlines.
2004 ‘The Scheme of Domestic Airfare Reform’ was promulgated to regulate airline
pricing.
2005 ‘Regulation on Domestic Investment on Civil Aviation’ was promulgated,
resulting in the establishment of more than ten private carriers in the following
years, including Spring, Dongxin, Okay, etc.
2006 ‘Regulation on Operation of Chinese Civil Aviation Domestic Routes and
Flights’ relaxed the entry to and exit from a route.
2007 CAAC suspended approval for new airline applications.
2009 China Eastern took over Shanghai Airlines.
2013 CAAC resumed the issuing of licences to new airlines, resulting in the birth
of a second wave of private airlines. Restrictions on airfares were completely
removed.
Source: Based on various sources.

M4371-FINGER_9781786431851_t.indd 120 14/11/2017 11:14

The Case of India
Tata, renamed ‘Air India’ in 1946, was the most successful private airline
in India before 1953. Since the independence of India, the government
has maintained a heavy presence in the aviation activities, with many
cumbersome rules and regulations imposed on almost every sector of
the industry. In 1947 the Indian Government acquired 49 per cent of Air
India’s shares. In 1953, under the Air Corporation Act (1953), the whole
industry was nationalized. At the same time, the government established
Indian Airlines by merging 11 domestic carriers on the grounds that there
were too many airlines in the domestic market. After that and until 1994
the operation of scheduled air transport services was largely dominated by
the two government-owned corporations. In the intervening years between
1953 and 1994, Indian Airlines mainly served the domestic routes, while
Air India primarily served the international market. The government con-
trolled airfares and regulated the carriers’ entry to and exit from the routes.
India initiated a move towards liberalization in the 1980s. The Minister
of Tourism and Civil Aviation announced that private airlines were
allowed to operate charter and non-scheduled services under the Air Taxi
scheme in 1986. This move came at a time when the poor performances
of Air India and Indian Airlines were blamed for stagnant international
tourism (Hooper, 1998). However, this initial liberalization was inade-
quate, as many conditions were imposed on the private airlines: they could
not publish time schedules or issue tickets to passengers, and they could
only use 15- to 50-seat aircraft. The government then declared in 1989
that the air taxi operators would eventually be given the right to provide
scheduled services. East West Airlines, established in 1992, was the first
private airline to become a substantial competitor against Indian Airlines.
Together with Jet Airways, Damania Airways and ModiLuft, this airline
added capacity and ‘stole’ the national airline’s market share (Hooper,
1998).
Deregulation in air transport was part of a broader economic liberali-
zation agenda in India in the early 1990s that opened many industries to
private domestic and foreign investors. The Air Corporations Act 1953
was repealed in 1994 and replaced with the Air Corporations Act of the
same year. The 1994 reform corporatized Air India and Indian Airlines,
and allowed the new private companies to provide services as fully-fledged
airlines (Findlay and Goldstein, 2004; Hooper, 1998). Six air taxi opera-
tors were given permits to operate scheduled services in October 1994.
However, the government’s policy at the time did not intend to remove
restrictions on private investment in airlines. For example, Tata’s proposal
for launching a domestic carrier was rejected in 1996–98. In 1997, Kuwait

M4371-FINGER_9781786431851_t.indd 121 14/11/2017 11:14

Airways and Gulf Air had to sell out their shares in Jet Airways to a non-
resident Indian as required by the domestic service ownership rules.
In 2003, the Civil Aviation Ministry set up a high-level committee
headed by the former Cabinet Secretary Shri Naresh Chandra to prepare
a roadmap for the civil aviation sector; this roadmap was to provide the
basis of a new National Civil Aviation Policy. This milestone report,
released by the Naresh Chandra Committee, recommended that foreign
entities be allowed to pick up to a 49 per cent equity stake in the domestic
and international scheduled air transport services with the approval of the
Foreign Investment Promotion Board (FIPB). This recommendation was
immediately adopted by the government in 2004, with the limit on foreign
equity in airline companies being raised from 40 per cent to 49 per cent.
However, equity from foreign airlines was not allowed – either directly or
indirectly – in domestic transport services.
In 2004 the Indian government introduced the 5/20 rule, which stated
that new airlines (mainly private airlines) are permitted to fly international
routes only if the airline has five years of domestic flying experience and at
least 20 aircraft in its fleet. Despite the restrictions of serving international
markets, the private carriers have flourished in the domestic markets. In
1992 Indian Airlines commanded a market share (domestic) of 95 per cent
(Jain, 2006); only one year later, its market share dropped to 76 per cent. In
2003 there were only three private carriers in India: Jet Airways, Air Sahara
and Air Deccan, but the private airlines had already carried more than
half of the passenger traffic in the domestic market. The private airlines
have not only taken market share, but their presence is also associated with
lower prices. They have led the significant change in competition in India’s
domestic market. In 2015, major private carriers present in India’s domes-
tic market included Jet Airways, SpiceJet, IndiGo, Vistara, AirAsia India
and Go Air. By the end of 2015 their market share had increased to 84 per
cent, according to the Directorate General of Civil Aviation.3
Low-cost airlines have also made a significant contribution to lower-
ing airfares. Established in 2003, Air Deccan was India’s first low-cost
airline based at Bangalore, famous for its software exports. O’Connell and
Williams (2006) noted that Air Deccan’s strategy for low fares provided
many low-income Indians with their first opportunity to fly. Deccan also
established an effective distribution system by taking advantage of India’s
software technology, thereby substantially lowering its distribution cost
(O’Connell and Williams, 2006). The airlines’ rapid growth sparked the
entry of many other LCCs in India between 2005 and 2007, including
SpiceJet, IndiGo, GoAir and JetLite. However, the quick expansion was
not accompanied by healthy financial positions. Deccan was acquired
by Kingfisher Airlines in 2008; Kingfisher failed in 2012 due to financial

M4371-FINGER_9781786431851_t.indd 122 14/11/2017 11:14

problems, and subsequently its domestic and international flight entitle-

ments were withdrawn by the government. IndiGo seized the opportunity,
and has become the largest Indian airline since 2012 in terms of passenger
market share. In 2015 it commanded a market share of 36.7 per cent, fol-
lowed by Jet Airways’ 19.2 per cent.
The reform to improve the profitability of the government-owned air-
lines has been an ongoing task. As early as 1995, the government set up
a commission to diagnose the problems of Air India and Indian Airlines
(Findlay and Goldstein, 2004). It seems that no effective measure has been
taken in the last two decades, even though the prime problem facing these
airlines is very clear to the industry leaders. Saraswati (2001) cited former
Air India Chairman Mr Modi: ‘Air India is not profitable because govern-
ment is not allowing the board to manage the airline in the right way’.
Privatization seemed to be one of the means that could be used to drive
the airlines to improve. However, the bureaucratic processes in seeking
strategic partners for the two airlines turned out to be lengthy, inefficient
and uncertain (Findlay and Goldstein, 2004). The industry union and
the local media also played a significant part in preventing the govern-
ment from seeking privatization. On 15 July 2007, Air India and Indian
Airlines merged into one airline: a new company – called the National
Aviation Company of India Limited (NACIL) – was created as the holding
company, and the merged entity operated under the name of Air India. In
the following year, Air India continuously lost ground to the aggressive
private carriers and LCCs. Its market share shrank to 16 per cent in the
domestic market in 2015; the company is currently the third-largest carrier
in India, after IndiGo and Jet Airways, in terms of the number of pas-
sengers carried. Table 7.2 summarizes the major milestones in the develop-
ment of India’s aviation industry.
A COMPARISON OF THE TWO MARKETS
An Overview
Over the past decade the growth of revenue passenger kilometres (RPK)
in the airline industry has remained at an annual rate of about 5 per cent
among the advanced economies such as the US, Australia, Canada and the
European Union. On the other hand, the emerging and developing econo-
mies (EDEs) such as Russia, Brazil, China and India have enjoyed a higher
RPK growth rate of about 10 per cent. Overall, these eight economies
accounted for 71.6 per cent of the world’s RPK in 2014, and of this, the
US, Australia, Canada and the EU accounted for 54.5 per cent, whereas

M4371-FINGER_9781786431851_t.indd 123 14/11/2017 11:14

Table 7.2 Evolution of India’s civil aviation industry
Year Major Milestones

1953 All private airlines were nationalized through the Air Corporation
Act.
1986 Private airlines were permitted to operate as air taxi operators.
1989 The planning commission (Transport Division) proposed that a
phased deregulation policy be adopted.
1990 Air cargo ‘open skies’ policy was adopted.
1993 Gulf Air and Kuwait Airways acquired 40 per cent stake in Jet
Airways.
1994 Air Corporation Act was repealed and private airlines were allowed
to operate scheduled services; Route Dispersal Guidelines were
introduced, forcing carriers to deploy part of their capacity on less
commercially viable routes.
1995 Jet, Sahara, Modiluft, Damania and East West were granted
scheduled airline status.
1997 Kuwait Airways and Gulf Air sold their stakes in Jet Airways to a
non-resident Indian, as required by the domestic service ownership
rules.
2003 The Civil Aviation Ministry set up a high-level committee, headed
by the former Cabinet Secretary Shri Naresh Chandra, to prepare a
roadmap for the civil aviation sector; Air Deccan started its service as
LCC.
2004 Airlines with a minimum of five years of continuous operations and
a minimum fleet size of 20 aircraft were permitted to fly international
routes (‘5/20 rule’); Jet Airways launched its international services.
2005 Kingfisher, SpiceJet, IndiGo, GoAir and Paramount commenced
operations.
2006 The government approved the restructuring of Mumbai Airport and
Delhi Airport through public–private partnership.
2007 Airline consolidation occurred: Air India took over Indian Airlines;
Jet Airways acquired Sahara; Kingfisher acquired Air Deccan.
2009 The Airports Economic Regulatory Authority (AERA) Bill was
passed to regulate the economic aspects of airports.
2010 SpiceJet started its international services.
2011 IndiGo started its international services.
2016 Foreigners are allowed to own 100 per cent of Indian airlines, but
airlines based outside of India can own no more than 49 per cent.
The 5/20 rule is modified and all airlines can commence international
operations if they have 20 aircraft or 20 per cent of total capacity,
whichever is higher for domestic operations.
Source: Based on ICRA (2012).

M4371-FINGER_9781786431851_t.indd 124 14/11/2017 11:14

25%
20%
15%
10%
5%
0%
–5%
–10%
–15%
–20%
1990 1995 2000 2005 2010 2015
US AU CA E6 RU BR CN IN
Notes:
1. E6 5 the UK, France, Germany, Netherlands, Italy and Spain; E6 account for 55 per cent
of the EU’s RPK.
2. Because of the disintegration of the USSR, USSR’s RPK data is used for Russia before
1992.
Source: ICAO.
Figure 7.1 Ten-year geometric average growth rate of revenue passenger

kilometres (RPK)
Russia, Brazil, China and India accounted for only 17.1 per cent. China’s
RPK share was far greater than that of India.
Figure 7.1 depicts the ten-year geometric average growth rates of RPK
for the eight economies. The aviation industries of the EDEs have experi-
enced much faster growth than that of the advanced countries in recent
years, and growth has been particularly impressive in China and India. As
shown in Figure 7.1, the growth of China’s RPK has been maintaining a
rate of over 10 per cent. India has had a similarly fast growth pattern since
2002.
Tables 7.3 and 7.4 provide data on air passenger traffic carried by the
major players in the two countries’ domestic aviation markets from 2001.
It can be seen that LCCs, India’s IndiGo and China’s Spring, enjoyed the
fastest growth from 2007, although SpiceJet and GoAir also grew fast.
However, it should be noted that even the largest Indian carrier, IndiGo, is
smaller than China’s fourth largest airline group, Hainan Airlines, in terms
of the volume of passengers carried. In fact, compared with their Chinese
counterparts, most Indian carriers operate at a much smaller scale. China’s

M4371-FINGER_9781786431851_t.indd 125 14/11/2017 11:14

Table 7.3 Passengers carried by Indian airlines
Air India Alliance Air Jet Airways Air Sahara/JetLite SpiceJet IndiGo GoAir Kingfisher
M4371-FINGER_9781786431851_t.indd 126
2000–01 3 288 000 1 670 000 5 925 000 688 000
2001–02 3 126 000 1 490 000 5 820 000 613 000
2002–03 3 394 000 1 420 000 6 402 000 1 302 000
2003–04 3 772 000 1 550 000 6 898 000 1 924 000
2004–05 4 391 000 1 450 000 9 015 000 2 606 000
2005–06 4 362 000 1 010 000 9 104 000 2 767 000 1 108 000 166 700
2006–07 4 342 000 820 000 10 739 373 3 180 694 2 598 479 912 962 1 263 048 3 260 714
2007–08 3 839 000 540 000 11 411 494 3 296 680 4 094 632 3 890 112 1 806 121 5 785 627
2008–09 2 834 000 320 000 11 030 474 3 369 012 4 099 458 4 917 360 1 128 475 8 651 773
126
2009–10 3 424 000 455 000 12 184 743 3 596 664 5 692 049 6 554 908 2 438 167 11 063 400
2010–11 3 843 000 503 000 14 643 462 4 328 524 7 384 010 9 466 897 3 393 984 12 011 807
2011–12 13 614 000 467 000 17 282 287 4 791 324 9 572 842 13 094 766 3 785 881 10 649 941
2012–13 14 183 200 393 000 16 818 737 3 867 556 11 681 066 16 887 969 4 386 914 1 211 925
2013–14 15 405 800 363 000 17 175 949 3 509 646 13 537 000 19 568 603 5 252 416
Annual growth 6.4% since −11.0% 6.9% 1.4% 26.6% 54.9% 22.6%
since 2007 2011–12
Notes:
1. For the 2011–12 period, the figures are from the merged data of Air India and Indian Airlines.
2. Air Sahara was taken over by Jet Airlines in 2007 and renamed Jetlite, which was rebranded as JetKonnect in 2012.
3. Kingfisher ceased operation in 2013 due to financial difficulties.

Source: Directorate General of Civil Aviation (DGCA).
14/11/2017 11:14
Table 7.4 Number of passengers carried by Chinese airlines
Air China China Hainan Shenzhen Sichuan Xiamen Shandong Spring

China Eastern Southern Airlines Airlines Airlines Airlines Airlines
M4371-FINGER_9781786431851_t.indd 127
2001 9 288 941 10 371 399 19 120 855 3 728 786 1 589 814 1 708 794 1 527 746
2002 10 586 674 10 385 521 14 074 768 3 706 595 2 560 566 2 037 485 4 737 062 2 172 512
2003 18 053 749 10 534 242 12 850 280 3 217 348 3 502 843 2 221 933 4 906 639 2 278 144
2004 24 499 955 15 273 214 17 675 604 4 357 095 4 818 808 3 655 026 6 232 405 3 009 114
2005 27 694 749 2 672 112 31 510 998 5 503 823 5 733 814 4 790 071 6 921 895 3 800 996 180 533
2006 31 504 292 29 707 707 34 641 367 7 298 060 7 178 267 5 881 210 7 784 997 5 040 415 1 137 062
2007 34 841 224 33 233 100 41 562 421 8 877 748 9 518 511 6 755 946 9 249 413 5 359 869 2 353 059
127
2008 34 256 741 31 686 627 43 967 758 8 976 447 11 951 918 6 701 470 9 618 860 5 420 085 2 943 775
2009 39 830 010 39 961 339 49 416 431 11 907 599 15 120 153 9 164 113 11 125 700 6 621 557 4 312 889
2010 46 245 017 43 063 011 56 636 882 11 508 391 16 487 503 10 667 547 13 561 359 8 071 913 5 859 697
2011 48 672 216 45 704 878 58 504 693 13 780 307 18 280 805 12 948 760 15 316 509 9 887 030 7 150 814
2012 49 334 466 48 068 308 61 809 891 15 238 122 19 805 010 14 445 233 16 843 870 11 246 342 9 110 993
2013 52 019 775 52 073 143 64 510 555 18 515 494 21 356 761 16 716 342 18 572 251 12 855 107 10 550 788
2014 54 637 234 54 275 647 70 611 294 20 937 194 23 214 385 19 191 395 20 380 159 14 125 476 11 446 984
Annual 6.6% 7.3% 7.9% 13.0% 13.6% 16.1% 11.9% 14.8% 25.4%
growth
since 2007
Source: Statistical Data on Civil Aviation of China.

14/11/2017 11:14
three major state airlines – Air China, China Eastern and China Southern
– recorded a steady growth rate in the last 15 years, while Air India, owned
by the Indian Government, was stagnant for many years and has only
managed to achieve a stable growth in recent years.
An Assessment of the Efficiency Performances of Airlines in China and

India
Data envelopment analysis (DEA) is a widely used non-parametric linear

programming technique for measuring the production efficiency of a
decision-making unit. Berger and Humphrey (1997) reviewed 130 studies
from 21 countries and concluded that the results from non-parametric
approaches, such as DEA, and parametric approaches, such as stochastic
frontier analysis, are similar, but the non-parametric approach generates
lower mean efficiency scores and has greater dispersion of efficiency scores
than a parametric approach. In addition, non-parametric approaches do
not require an a priori assumption on functional form specification, which
may restrict the frontier shape. Instead, they rely on linear programming to
identify a benchmark practice and measure the efficiency of other firms in
relation to this. Here, the DEA approach is used to assess the efficiency of
Chinese and Indian airlines.
The envelopment surface of DEA is constructed based on the constant
return to scale (CRS), the variable return to scale (VRS), and non-increasing
returns to scale (NIRS) assumptions. Saranga and Nagpal (2016) argued
that the VRS assumption is more realistic in the Indian airline industry,
as regulatory constraints, budgetary restrictions and mergers may lead
airlines to operate on an inefficiently small scale. Technical efficiency under
the CRS assumption measures the overall technical efficiency (OTE),
which reflects an airline’s ability to transform multiple resources into mul-
tiple air services. It can be decomposed into pure technical efficiency (PTE)
and scale efficiency (SE). The PTE is generated by estimating the efficient
frontier under the assumption of VRS. It can be used as a measure captur-
ing a firm’s managerial performance. SE is the ratio of OTE to PTE. The
scale efficiency score indicates whether a firm operates at the most pro-
ductive scale. If SE is less than one, scale inefficiency occurs, the direction
of which can be determined by running the DEA model under the NIRS
assumption and comparing the efficiency scores with those estimates for
the CRS-based technical efficiency measure. When the NIRS and CRS
measures are equal but differ from the VRS measure, increasing returns to
scale exist for that firm. However, if the VRS and NIRS measures are equal
but differ from the CRS measure, the case of decreasing returns to scale
applies at the relevant point on the frontier (Ray, 2004).

M4371-FINGER_9781786431851_t.indd 128 14/11/2017 11:14

The two outputs used are passenger traffic and cargo, measured by
RPK and revenue ton-kilometres. The inputs included in this research
are the number of employees and the number of aircraft, reflecting the
fact that the airline industry is both labour-intensive and capital-intensive.
The choice was also determined by the availability of data. The input and
output data for Indian airlines were drawn from air transport statistics
provided by the Directorate General of the Ministry of Civil Aviation of
the Government of India, while the data for Chinese carriers come from
the yearbooks of Statistical Data on Civil Aviation of China.
Much of the work measuring airline efficiency uses the input-oriented
model, given that airlines have a greater influence on the inputs than on
the outputs, as macroeconomic conditions tend to have a strong impact on
consumer demand for air services (Merkert and Hensher, 2011). This seems
appropriate because a carrier’s management has control over how inputs
are used to generate outputs in most instances. The rule of thumb when
establishing a sample size in DEA studies is that it be at least equal to the
product of input and output or three times the sum of inputs and outputs
(Cooper et al., 2001). The two inputs and two outputs used here satisfy the
minimum requirement, as there are 14 airlines in the sample.
The technical efficiency scores as a percentage for 2007–14 under
various assumptions are seen in Tables 7.5 and 7.6. The main findings are:
●● India’s state-owned, full-service carrier Air India, and Spring

Airlines, a private LCC in China, exhibited technical efficiency con-
sistently throughout the study period. This is consistent with Jain
and Natarajan (2015) and Saranga and Nagpal (2016), who found
that Air India was among the most technically efficient airlines in the
periods 2005–07 and 2005–10. Spring has also been widely regarded
as a successful carrier that delivered consecutive net profits since it
commenced operations in 2005.
●● China Eastern has been relatively inefficient among the four major
airline groups in China, although there has been a slight improve-
ment after its merger with Shanghai Airlines in 2009. Hainan
Airlines’ performance was the best of the four. However, the provin-
cial carrier Sichuan Airlines achieved much higher OTE for most of
the time – except in 2009, probably due to the global financial crisis.
The regional carrier China Express has shown consistently low OTE
but high PTE, indicating that the inefficiency mainly came from
scale inefficiency. In fact, a comparison of the VRS and NIRS meas-
ures suggests that this regional carrier is in the stage of increasing
returns to scale. Therefore, its technical efficiency could be improved
through an expansion of its size.

M4371-FINGER_9781786431851_t.indd 129 14/11/2017 11:14

Table 7.5 Technical efficiency under CRS and VRS assumptions
M4371-FINGER_9781786431851_t.indd 130
2007 2009 2011 2013 2014
OTE PTE OTE PTE OTE PTE OTE PTE OTE PTE
(CRS) (VRS) (CRS) (VRS) (CRS) (VRS) (CRS) (VRS) (CRS) (VRS)
Air India 100 100 100 100 100 100 100 100 100 100
Jet Airways 83.26 83.3 90.98 100 71.3 100 100 100 93.49 100
SpiceJet 99.27 100 91.76 92.75 100 100 74 78.07 80.85 82.67
IndiGo 56.86 73.92 100 100 100 100 100 100 100 100
130
GoAir 84.95 100 42.01 64.45 75.26 95.61 95.78 100 91.87 100
Air China 69.25 100 64.37 100 65.74 100 79.48 100 80.38 100
China Eastern 49.91 69.71 47.95 73.9 51.88 78.24 55.89 82.51 55.26 70.19
China Southern 48.06 87.3 46.01 72.63 51.76 100 65.82 100 68.17 100
Hainan Airlines 76.82 93.32 71.76 94.26 65.01 76.88 87.94 100 100 100
Sichuan Airlines 100 100 65.2 100 98.68 100 90.64 100 90.14 100
Spring Airlines 100 100 100 100 100 100 100 100 100 100
Juneyao Airlines 63.74 88.84 67.35 75.73 75.06 81.87 79.79 84.12 84.93 90.93
Okay Airways 54.32 81.88 44.38 68.47 69.56 88.34 51.87 79.62 50.71 85.39
China Express 17.81 100 16.66 100 23.95 100 19.04 100 30.75 100

14/11/2017 11:14
Table 7.6 Scale efficiency and technical efficiency under the NIRS assumption
M4371-FINGER_9781786431851_t.indd 131
2007 2009 2011 2013 2014
CE TE CE TE CE TE CE TE CE TE
(NIRS) (NIRS) (NIRS) (NIRS) (NIRS)
Air India 100 100 100 100 100 100 100 100 100 100
Jet Airways 99.95 83.26 90.98 100 71.3 100 100 100 93.49 100
SpiceJet 99.27 99.27 98.93 91.76 100 100 94.8 74 97.8 80.85
IndiGo 76.92 56.86 100 100 100 100 100 100 100 100
131
GoAir 84.95 84.95 65.19 42.01 78.72 75.26 95.78 95.78 91.87 91.87
Air China 69.25 100 64.37 100 65.74 100 79.48 100 80.38 100
China Eastern 71.59 69.71 64.88 73.9 66.31 78.24 67.73 82.51 78.73 70.19
China Southern 55.05 87.3 63.35 72.63 51.76 100 65.82 100 68.17 100
Hainan Airlines 82.32 93.32 76.13 94.26 84.57 76.88 87.94 100 100 100
Sichuan Airlines 100 100 65.2 100 98.68 100 90.64 100 90.14 100
Spring Airlines 100 100 100 100 100 100 100 100 100 100
Juneyao Airlines 71.74 63.74 88.94 67.35 91.67 75.06 90.78 62.56 93.41 84.93
Okay Airways 66.34 54.32 64.82 44.38 78.74 69.56 76.8 60.95 59.39 50.71
China Express 17.81 17.81 16.66 16.66 23.95 23.95 24.31 24.31 30.75 30.75

14/11/2017 11:14
●● IndiGo, the largest Indian carrier, and Jet Airways have been operat-
ing at a very efficient level in recent years in using their resources. In
general, these private airlines outperformed their Chinese counter-
parts such as Okay and Juneyao in terms of OTE and PTE.
●● In 2014 Air China, China Southern, Sichuan Airlines and Jet
Airways exhibited diminishing returns to scale, implying that they
would be more technically efficient if they could shrink their size.
Apart from Air India and Spring Airlines, the other carriers were at
the point where increasing returns to scale applied in 2014.
Comparison of the Policy Environments in which Private Airlines and

LCCs Operate
The analysis has shown that private airlines are dominant players in the
Indian airline market, and that they were generally more efficient than
their Chinese counterparts. State-owned Air India, despite being more
efficient than Chinese state-owned carriers, constantly lost market shares
to private airlines. This is not surprising, as India opened the aviation
sector to private investors in the mid-1990s, whereas China only did this
halfway through 2005. In India, airfares have been largely determined by
market forces since the 1990s. In contrast, China’s first private airline did
not appear until 2005, and the government had attempted to influence
the price-setting in various ways until 2013, when the 2004 Scheme of
Domestic Airfare Reform was formally repealed. Although price discounts
were allowed as early as 1997, the CAAC constantly sought to re-regulate
airfares to avoid price wars and industry losses by putting a limit on the
maximum discounts. The Chinese government also introduced a ‘revenue
pooling’ programme and guided the airline consolidations to restrict com-
petition from the late 1990s to the early 2000s (Zhang and Round, 2008).
Additionally, before China introduced its anti-trust laws in 2008, price-
fixing was tacitly allowed and sometimes encouraged by the CAAC.
Zhang and Round (2011) noted that airfare collusion in China was not a
secret and had been widely reported by newspapers. Major Chinese airline
groups such as Air China, China Southern and China Eastern held talks
from time to time to prevent airfares sliding down to train-fare levels.
Zhang and Round (2011) reported that since the deregulation of airline
prices in 1997, Chinese passengers have constantly witnessed overnight
across-the-board increases in airfares. Although most of the collusive
agreements were short-lived, some markets had consistently higher prices
for a relatively long time, especially on some business routes associated
with Beijing, Guangzhou and Shanghai, which are vital to the profits of
the state-owned airlines. This was still the case even after the introduction

M4371-FINGER_9781786431851_t.indd 132 14/11/2017 11:14

of China’s Anti-Monopoly Law in 2008, as the new anti-trust enforcement

agencies had limited resources and little experience in dealing with anti-
trust cases. For example, the merger between China Eastern and Shanghai
Airlines in 2009 was not challenged, nor were any conditions that were
imposed, such as giving up some slots at both Shanghai Pudong and
Shanghai Hongqiao airports, resulting in substantial airfare increase in the
following years (Zhang, 2015). This was especially so when private airlines
were taken over by their state-owned counterparts, such as the takeover of
the largest private carrier, Shenzhen Airlines, by Air China in 2010. United
Eagle Airlines, one of the first private carriers, was also taken over in 2009
by state-controlled Sichuan Airlines without any investigation by the anti-
trust authorities. It appears that China’s state-owned airlines could enjoy
a certain degree of market power either through explicit and implicit col-
lusion or mergers and acquisitions to defend their market share and elimi-
nate potential competition from private carriers. Therefore, Zhang and Lu
(2013) claimed that the lack of effective enforcement of the anti-trust laws
in the aviation markets might be an important factor that has impeded the
development of LCCs and private carriers.
As with other private carriers, the successful Spring Airlines also faced
the restrictions on entry into some lucrative routes and ideal time slots.
The Shanghai-based LCC had to wait for six years before it was allowed to
fly the Shanghai–Beijing route in 2011 (Zhang and Lu, 2013). However, in
2015, the CAAC announced that it would reform flight-time slot allocation
at Guangzhou Baiyun Airport and Shanghai Pudong Airport, with an aim
to promote transparency, efficiency and equity. The previous allocation
system favoured the state-owned airlines and led to rampant corruption
and bribery. The proposed new allocation methods included auction and
lottery. Slots are valid for three years and can be exchanged, transferred
and sold. Although the new system seems to be fairer to the private air-
lines, in the first round of slot auction at Baiyun Airport, all the ‘nice’
slots were claimed by the four major airline groups and their subsidiaries.
A senior executive from a private carrier told China Daily that the auction
was too expensive for them (Wang, 2015).
Airport slot allocation in India seems much more transparent and fairer
compared with the case of China. A guideline on slot allocation issued
by the Ministry of Civil Aviation has been in place for a long time, and
has been subject to review and revision from time to time to make it more
transparent and equitable. The allocation is based on ‘Grandfather Rights’
and ‘Use it or Lose it’ in the case of mergers and acquisitions. After the
allocation following these two rules, 50 per cent of the leftover slots will be
allocated to new airlines. Although there is no doubt that the ‘Grandfather
Rights’ rule favours the incumbents and creates a barrier for new airlines,

M4371-FINGER_9781786431851_t.indd 133 14/11/2017 11:14

the guideline still explicitly considers the interests of the new carriers,
which is lacking in China’s system of slot allocation.
Private airlines in India were also subject to some constraints. For
example, the five-year operation requirement prior to being eligible for
operating international services protected the interests of the experienced
airlines such as Air India and Jet Airlines, but is unfair for the newly
established low-cost carriers which could not launch international services
until five years later. This policy closed the door for new airlines to grow
through operating international flights, even though they may have met
all the security and safety standards. There have been constant calls for an
equal treatment for private and state-owned airlines (e.g. Nathan, 2012),
while such voice is much weaker in China.
India’s state-owned carriers appeared to have served more social goals
in addition to commercial performance. For example, the government
required that all scheduled carriers flew at least 10 per cent of their capac-
ity on trunk routes in the north-eastern sector, Jammu and Kashmir,
Andaman and Nicobar Islands and Lakshadweep (Category II routes).
However, the state-owned national airline was forced to put 17 per cent
of its flights on these routes, causing huge losses every year. On Category
III routes (non-trunk and non-Category-II routes, mainly associated
with small cities), the airlines were mandated to deploy 50 per cent of
the number of aircraft deployed on Category I routes. Again the national
airline was forced to undertake a higher percentage of its flights in these
markets than the private airlines. It is inefficient for the public carrier to
operate with such constraints (Jain, 2006). This might be one of the factors
that led to Indian Airlines constantly losing its domestic market share to
private competitors.
CONCLUSION
China and India have attracted the attention of the world in the last two
decades as powerhouses of economic growth. There has been a plethora
of comparative studies on various aspects of the economies between China
and India; however, research into the development of air transport poli-
cies and market outcomes for the two countries is lacking. This study has
reviewed the evolution of the air transport industry in the two countries,
and then assessed and compared the efficiency performances of the state-
owned and privately owned airlines in the two markets. It has found that
Indian carriers tend to be more efficient than their Chinese counterparts.
Air India and Spring Airlines are the leaders in terms of pursuing techni-
cal efficiency, and the dominant status of the private airlines in the Indian

M4371-FINGER_9781786431851_t.indd 134 14/11/2017 11:14

market does not come by chance. The analysis of this research shows
that China’s aviation policy has long been hostile to the private carriers
and overly protective of the state-owned airlines. Various obstacles have
constrained the expansion and growth of the private airlines. In contract,
Indian private airlines, especially the LCCs, operate in a relatively more
liberal environment. With the private carriers dominating the market, they
could exercise a stronger influence on government policy, which would in
turn promote and accelerate their growth.
NOTES
1. This intuition was formalized in a recent theoretical paper by Jiang and Zhang (2016).
2. One US$ is about 6.6 yuan (Chinese currency, also known as RMB) as of August 2016.
3. See http://dgca.nic.in/reports/Traffic-ind.htm.
REFERENCES
Berger, A.N. and D.B. Humphrey (1997), Efficiency of Financial Institutions:

International Survey and Directions for Future Research, European Journal of
Operational Research, 98, 175–212.
Cooper, W.W., S. Li, K. Tone, R.M. Thrall and J. Zhu (2001), Sensitivity and
Stability Analysis in DEA: Some Recent Developments, Journal of Productivity
Analysis, 15, 217–46.
Dougan, M. (2002), A Political Economy Analysis of China’s Civil Aviation
Industry, New York, USA and London, UK: Routledge.
Fan, L.W., F. Wu and P. Zhou (2014), Efficiency Measurement of Chinese Airports
with Flight Delays by Directional Distance Function, Journal of Air Transport
Findlay, C. and A. Goldstein (2004), Liberalization and Foreign Direct Investment
in Asian Transport Systems: The Case of Aviation, Asian Development Review,
21, 37–65.
Fu, X., A. Zhang and Z. Lei (2012), Will China’s Airline Industry Survive the
Entry of High Speed Rail?, Research in Transportation Economics, 35, 13−25.
Fu, X., T.H. Oum, R. Chen and Z. Lei (2015a), Dominant Carrier Performance
and International Liberalization: The Case of North East Asia, Transport Policy,
43, 61−75.
Fu, X., Z. Lei, K. Wang and J. Yan (2015b), Low Cost Carrier Competition and
Route Entry in an Emerging but Regulated Aviation Market: the Case of China,
Transportation Research Part A: Policy and Practice, 79, 3–16.
Hooper, P. (1998), Airline Competition and Deregulation in Developed and
Developing Country Contexts: Australia and India, Journal of Transport
Geography, 6, 105–16.
ICRA (2012), Indian Aviation Industry: Through Turbulent Times, FDI
Relaxation Alone not a Game Changer, available at http://www.icra.in/Files/
ticker/Indian%20Aviation%20Industry%20(NEW).pdf.

M4371-FINGER_9781786431851_t.indd 135 14/11/2017 11:14

Jain, R.K. and R. Natarajan (2015), A DEA Study of Airlines in India, Asia Pacific
Management Review, 20, 285–92.
Jain, S. (2006), Competition Issues in Transportation Sectors, in P.S. Mehta
(ed.), Towards a Functional Competition Policy for India, Haipur: Academic
Foundation in Association with Consumer Unity and Trust Society (CUTS
International).
Jiang, C. and A. Zhang (2016), Airline Network Choice and Market Coverage
Under High-speed Rail Competition, Transportation Research, Part A: Policy
and Practice, 92, 248–60.
Le, T.T. (1997), Reforming China’s Airline Industry: From State-owned Monopoly
to Market Dynamism, Transportation Journal, 36, 45−62.
Li, M. (2014), Spring Airlines Approved for IPO After Ten Years of Strict Cost
Control. Jiemian, available at http://www.jiemian.com/article/202427.html.
Liu, C. and K. Jayakar (2012), The Evolution of Telecommunications Policy-
making: Comparative Analysis of China and India, Telecommunications Policy,
36, 13–28.
Mazumdar, A. (2008), Deregulation of the Airline Industry in India: An Analysis
of the Government’s Policy, Rationale and Strategy, PhD thesis, Department of
Political Science, Miami University.
Merkert, R. and D.A. Hensher (2011), The Impact of Strategic Management and
Fleet Planning on Airline Efficiency – A Random Effects Tobit Model Based on
DEA Efficiency Scores, Transportation Research Part A: Policy and Practice, 45,
686–95.
Nathan (2012), Research Study of the Civil Aviation Sector in India, Nathan
Economic Consulting India, submitted to the Ministry of the Corporate Affairs,
Government of India.
O’Connell, J.F. and G. Williams (2006), Transformation of India’s Domestic
Airlines: A Case Study of Indian Airlines, Jet Airways, Air Sahara and Air
Deccan, Journal of Air Transport Management, 12, 358–74.
Ray, S.C. (2004), Data Envelopment Analysis: Theory and Techniques for Economics
and Operations Research, Cambridge: Cambridge University Press.
Saranga, H. and R. Nagpal (2016), Drivers of Operational Efficiency and its
Impact on Market Performance in the Indian Airline Industry, Journal of Air
Saraswati, S.K. (2001), Operating Environment for a Civil Aviation Industry in
India, Journal of Air Transport Management, 7, 127–35.
Scheraga, C.A. (2004), Operational Efficiency Versus Financial Mobility in the
Global Airline Industry: A Data Envelopment and Tobit Analysis, Transportation
Research Part A, 38, 383–404.
Taneja, N. (2004), Simpli-flying, Optimizing the Airline Business Model, Aldershot:
Ashgate Publishing.
Wan, Y., H-K. Ha, Y. Yoshida and A. Zhang (2016), Airlines’ Reaction to High-
speed Rail Entries: Empirical Study of the Northeast Asian Market Working
paper, Hong Kong Polytechnic University.
Wang, J., D. Bonilla and D. Banister (2016), Air Deregulation in China and its
Impact on Airline Competition 1994–2012, Journal of Transport Geography, 50,
12–23.
Wang, K., Q. Gong, X. Fu and X. Fan (2014), Frequency and Aircraft Size
Dynamics in a Concentrated Growth Market: The Case of the Chinese Domestic
Market, Journal of Air Transport Management, 36, 50–58.

M4371-FINGER_9781786431851_t.indd 136 14/11/2017 11:14

Wang, W. (2015), Big Carriers Dominate First Airport Slot Auction, China
Daily USA, 31 December, available at http://usa.chinadaily.com.cn/epaper/201
5–12/31/content_22885816.htm.
Yuen, A. and A. Zhang (2009), Effects of Competition and Policy Changes
on Chinese Airport Productivity: An Empirical Investigation, Journal of Air
Zhang, A. (1998), Industrial Reform and Air Transport Development in China,
Journal of Air Transport Management, 4, 155−64.
Zhang, A. and H. Chen (2003), Evolution of China’s Air Transport Development
and Policy towards International Liberalization, Transportation Journal, 42,
31−49.
Zhang, A., S. Hanaoka, H. Inamura and T. Ishikura (2008), Low Cost Carriers in
Asia: Deregulation, Regional Liberalization and Secondary Airports, Research
in Transportation Economics, 24, 36–50.
Zhang, Q., H. Yang, Q. Wang and A. Zhang (2014), Market Power and its
Determinants in the Chinese Airline Industry, Transportation Research Part A:
Policy and Practice, 64, 1−13.
Zhang, Y. (2012), A Retrospective Analysis of the Effects of Airline, Mergers on
China Eastern’s Lifeline Routes: A Difference-in-differences Approach, Journal
of the Asia Pacific Economy, 17, 464−75.
Zhang, Y. (2015), Merger Between Airlines in Financial Distress: Does the Merger
Save Them?, Competition and Regulation in Network Industries, 16, 66−81.
Zhang, Y. and C. Findlay (2010), Air Transport Liberalization, in E. Ghani (ed.),
The Service Revolution in South Asia, New York: Oxford University Press.
Zhang, Y. and Z. Lu (2013), Low Cost Carriers in China and its Contribution to
Passenger Traffic Flow, Journal of China Tourism Research, 9, 207−17.
Zhang, Y. and D.K. Round (2008), China’s Airline Deregulation Since 1997 and
the Driving Forces Behind the 2002 Airline Consolidations, Journal of Air
Transport Management, 14, 130−42.
Zhang, Y. and D.K. Round (2009), The Effects of China’s Airline Mergers on
Prices, Journal of Air Transport Management, 15, 315−23.
Zhang, Y. and D.K. Round (2011), Price Wars and Price Collusion in China’s
Airline Markets, International Journal of Industrial Organization, 29, 361–72.
Zhang, Y. and A. Zhang (2016), Determinants of Air Passenger Flows in China
and Gravity Model: Deregulation, LCCs, and High-speed Rail, Journal of
Transport Economics and Policy, 50, 287–303.

M4371-FINGER_9781786431851_t.indd 137 14/11/2017 11:14

8. European market: present and
future
Volodymyr Bilotkach
INTRODUCTION
The creation of the single airline market in the European Economic Area
(EEA) is a major achievement of the European Union. The establish-
ment of this market has created incentives for new airlines to appear and
grow, made air travel affordable to millions of people, and has (generally)
contributed to deeper integration of the European countries. This brief
chapter will review the main changes that have occurred in European com-
mercial passenger aviation since the 1990s (with a more detailed focus on
the last decade), compare outcomes of the European liberalization to what
happened in the United States, and outline the threats and opportuni-
ties for the future of European aviation, especially in light of the United
Kingdom’s forthcoming departure from the EU.
Overall, the main developments in the European passenger airline
market since deregulation have been the following. First, we have witnessed
an explosive growth of a new breed of airline – the so-called low-cost
carrier or LCC. While industry observers and experts would not, gener-
ally speaking, be able to point to the threshold in, say, cost per available
seat mile (CASM), below which an airline becomes an LCC, the business
models of these airlines do share several common traits. Interestingly, the
LCCs that emerged in Europe post-liberalization are different in a number
of ways from the airlines that are considered to be their counterparts in the
United States. The key hereto unexplored question related to these airlines
is whether the LCCs pose a genuine threat to the so-called legacy carri-
ers, or whether they simply expand the market to the segments the legacy
airlines did not cover. The second important development is consolidation
of the European legacy carriers. I believe that we are on the path towards
a set-up that includes three large EU network carriers. Interestingly, this
setting mirrors what is happening in the United States, transatlantic and
mainland China markets. Last but not least, while the market within the
EEA is truly liberalized and open; the same cannot be said about most of
138

M4371-FINGER_9781786431851_t.indd 138 14/11/2017 11:14

European market: present and future 139
the Europe–rest-of-the-world market segments. The EU’s effort to create

a network of open skies agreements with the rest of the world have so far
yielded few successes, the most important of which is the Transatlantic
Open Aviation Area. I believe that the limited liberalization of the airline
markets between Europe and the rest of the world does not allow full ben-
efits of airline liberalization within Europe to be realized.
European aviation has thrived since liberalization. While the process
was not without its losers (several legacy carriers have left the market in
the 21st century), overall air travel in Europe is more available and afford-
able than it has ever been – although mainly for people who live in major
metropolitan areas. Yet, a number of challenges remain. Airport capacity
and (to a larger degree) fragmentation of air navigation services create
important bottlenecks in the system. Most importantly, very recent politi-
cal events – such as the UK Government’s apparent decision to leave not
only the European Union, but also the Single Market, and the outcome of
the 2016 Presidential election in the United States – pose serious threats to
European aviation. Last but not least, development of the Europe–Asia
market poses both threats and opportunities to European airlines.
Unfortunately, academic research on development of the European
airline markets is scarce, especially if one compares it with the wealth of
studies on US airlines. This is related predominantly to the relative lack
of data on the European side. Nevertheless, a few studies are available.
One strand of literature on European aviation deals with the issue of price
setting and price dispersion, using samples of fare quotes collected online.
Notable papers here include Giaume and Gillou (2004), Bachis and Piga
(2011), Bilotkach et al. (2015a), Obermayer et al. (2013), and Bergantino
and Capozza (2014). An estimation of the market conduct parameters in
the Spanish airline industry is offered by Fageda (2006). A series of papers
have examined the issue of airline network development and network
structure in Europe (Burghouwt and De Wit, 2005; Dobruzskes, 2006,
2013). Calzada and Fageda (2012) offered an analysis of public service
obligations in Spain. Bilotkach et al. (2014) analysed the impact that Malev
Hungarian Airlines’ bankruptcy had on consumer welfare. Analysis of
competition between airlines and high-speed rail in the European context
was offered by Behrens and Pels (2012), Dobruzskes (2011), and Albalate
et al. (2015). With regard to other aviation sectors in Europe, the follow-
ing studies are of note. Bel and Fageda (2010) and Bilotkach et al. (2012)
analysed the effects of airport deregulation and privatization on aeronau-
tical charges. Analyses of the efficiency of Europe’s air navigation service
providers have been made by Button and Neiva (2014) and Bilotkach et al.
(2015b).
The rest of this chapter is organized as follows. The next section will

M4371-FINGER_9781786431851_t.indd 139 14/11/2017 11:14

discuss the EU airline market liberalization and key subsequent develop-

ments on the market. This will be followed by a more general discussion
of the nature and role of the airlines that are conventionally called ‘low-
cost carriers’. The final section discusses the current and future challenges
for European aviation, in light of the current situation and recent global
developments.
EU MARKET AND ITS GRADUAL LIBERALIZATION
Before we continue, some terms should be defined. When discussing the

European airline market, it is important to distinguish between the EU,
the EEA, and the Schengen Area. The European Union (the EU) currently
includes 28 countries. On 23 June 2016, an advisory referendum in the
United Kingdom gathered a narrow majority support for the UK to leave
the European Union, and the UK Government appears determined to act
upon this referendum outcome. However, the UK is likely to remain an
EU member state until about March 2019. I will discuss the implications
of this event for the airline market later on. The EU member states are:
Austria, Belgium, Bulgaria, Croatia, Republic of Cyprus, Czech Republic,
Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland,
Italy, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, Poland,
Portugal, Romania, Slovakia, Slovenia, Spain, Sweden and the UK. The
European Economic Area (the EEA), also known as the Single Market,
is the common trade area, which includes the European Union, as well
as Norway, Lichtenstein and Iceland. The EEA members subscribe to
the four freedoms of movement within the Single Market, which include
freedom of movement of goods, people, services and capital. Switzerland
is neither an EU nor an EEA member; its access to the EU Single Market
is governed by a series of treaties with the Union. The Schengen Area
consists of 26 European states that have agreed to abolish passport
control for travel between them. While EU countries comprise the core
of the Schengen Agreement (of the EU states, only Cyprus, Bulgaria,
Romania, Croatia, Ireland and the United Kingdom are not currently in
the Schengen Area), a number of non-EU states (Norway, Switzerland,
Lichtenstein and Iceland) are also part of the agreement.
For the purposes of this chapter, I will define the European airline
market as the market for scheduled commercial passenger airline services
within the EEA and Switzerland. However, examining this market over
time is a somewhat tricky affair. Since the 1990s, the EU/EEA has enlarged
considerably. Austria, Finland and Sweden joined the Union in 1995. A
considerable enlargement took place in 2004, when 10 countries became

M4371-FINGER_9781786431851_t.indd 140 14/11/2017 11:14

members, including five Eastern European Countries, three Baltic States,

as well as the island nations of Malta and Cyprus. Bulgaria and Romania
joined in 2007. The most recent European country to be admitted into the
EU was Croatia in 2013. For most of the discussion, I will focus on the
data since the 2004 enlargement.
Prior to liberalization, the EU commercial passenger airline industry
operated on the basis of a network of restrictive bilateral air service agree-
ments, typical of international aviation in general. Accordingly, air travel
within Europe was inconvenient and expensive. I have heard anecdotes
about a European organization finding it more cost-effective to hold a
meeting of its members from various countries in Washington, DC than
anywhere in Europe, and of a business traveller from North America
who had to visit two European destinations within a short period of time
finding it cheaper to book two transatlantic round-trips than an open-jaw
transatlantic itinerary combined with a one-way flight within Europe.
Things started changing in December 1987, when the First Aviation
Liberalization Package was agreed, which essentially removed most capac-
ity restrictions from the existing bilateral air service agreements, automati-
cally granted EU carriers ‘fifth freedom’ rights within the Union (so that,
for instance, Lufthansa could carry passengers from France to Italy via
Germany), and also removed the national governments’ right to block dis-
counted airfares if the airlines decide to offer those.
The Second Aviation Liberalization Package was agreed in July 1990.
It built on the First Package by introducing an element of ‘double disap-
proval’ for fares, stipulating that a fare set by an airline for a route between
Member States would be permitted unless both States disapproved of it.
The Second Package also opened up routes between almost all European
Community airports, relaxed restrictions on fifth freedom services, and
eased restrictions on multiple designation of airlines on particular routes.
Full liberalization of the European commercial passenger airline market
has been achieved only with the Third Package, which was approved in
June 1992. This package essentially established a truly open single market
within the European Economic Area. Henceforth, any airline that regis-
tered within the EEA and was capable of offering flights between any two
airports within the area has been permitted to do so. Essentially, the only
aspects of the airline market that remained subject to regulations are safety
and consumer protection (plus the obvious fact that the employers are to
treat their workers in accordance with the applicable labour laws; this is a
tricky issue in the airline industry).
Expansion of the European Union after the Third Liberalization
Package necessitated expansion of its single airline market. This posed
significant challenges to the former flag carriers, especially those from

M4371-FINGER_9781786431851_t.indd 141 14/11/2017 11:14

Eastern European countries. Those airlines did not have significant (if
any) experience of competing in a liberalized market, and they suddenly
ended up playing against both the larger network airlines that have been
working in a deregulated environment, sometimes for over a decade, as
well as against the new breed of point-to-point low-cost carriers that have
only known a competitive environment and have never been under any
regulatory protection. While not all of the Eastern European flag carri-
ers have fared well in this new environment, it would be incorrect to call
European market liberalization a failure for the region. I will now examine
the outcome of the EU market deregulation process in more detail.
LIBERALIZATION OUTCOMES
The goal of any market liberalization is to bring in the forces of competi-

tion, which should increase consumer welfare, yield new market entry,
lower prices, and generally bring the product mix offered on the market
more in line with what consumers are willing and able to pay for. While
data availability and the length of this chapter do not allow me to offer a
proper counterfactual analysis of the effects of airline market liberaliza-
tion in Europe (as Morrison and Winston (1986) have done for the US
market), some trends can be clearly outlined. Furthermore, it behoves us to
offer some comparison of the European market post-liberalization to the
outcomes of the US airline market deregulation in 1978. The two markets
are of comparable size. While Europe’s population (more than 500 million
people) exceeds that of the United States (approximately 300 million), gen-
erally longer distances between the metropolitan areas in the United States
and better developed surface transport travel alternatives in Europe could
even out the demand levels.
The main similarities between the EU and the US markets post-
liberalization include fast growth in passenger traffic; substantial reduc-
tion in average airfares; emergence of a new breed of airlines, commonly
known as low-cost carriers (LCCs); and, more recently, a trend towards
market consolidation. In both markets, the general trend has been towards
reduced levels of in-flight amenities. In the pre-liberalization airline
markets, price competition was restricted by either regulation (in the
United States) or bilateral air services agreements (on the European side),
prompting the carriers to use service quality levels to compete for passen-
gers. New entrants have discovered that the customers are willing to forgo
amenities such as meals, blankets and even free checked luggage in return
for a low enough price. As price competition intensified post-liberalization,
the incumbent carriers found themselves under pressure to reduce their

M4371-FINGER_9781786431851_t.indd 142 14/11/2017 11:14

costs, and also started following suit by reducing in-flight service quality
on short-haul intra-European flights (service adjustments on long-haul
flights have been less notable: if anything, premium class services on long-
haul flights have actually improved over the last two decades).
Market consolidation seems to be a natural outcome of airline market
liberalization. The United States has seen two merger waves since deregula-
tion (the first occurred in the 1980s and the second in the new millennium).
The Chinese domestic market has also seen some consolidation recently.
Mergers on the European market have generally been less conspicuous, as
they rarely result in disappearance of the airline brands. It would be more
correct to say that mergers between legacy carriers from different coun-
tries have not yet resulted in the airline brands disappearing. Air France
and KLM are now parts of one company; British Airways, Iberia and
Aer Lingus form the entity called the International Airline Group (IAG);
and Lufthansa owns Austrian Airlines, Swiss International Airlines and
Brussels Airlines. Consolidation events involving the new low-cost carri-
ers, however, have always resulted in one of the merger partners’ brands
disappearing. This somewhat unusual arrangement is partly driven by the
regulatory structure of most of the EU-outside-Europe markets. While
the European Union is striving to conclude multilateral air services agree-
ments with the outside countries, little progress has been made on this
front. With the exception of the United States and several North African
countries, air travel between the EU countries and outside states is still
driven by bilateral treaties, typically containing nationality clauses. This
necessitates retaining the airlines’ brands after the mergers, which is obvi-
ously not an issue for the United States airline industry.
Both markets are currently experiencing the emergence of a more or
less similar structure. Following the last US merger wave and consolida-
tion events in Europe, both sides of the Atlantic are left with three large
network carriers and a number of so-called low-cost airlines, which gener-
ally focus on point-to-point rather than connecting traffic. Interestingly,
there is a similar structure on the domestic market in Mainland China. Of
course, bankruptcies are as much a feature of liberalized markets as the
entry of new players and mergers between the incumbent firms. Notable
bankruptcies in the European airline industry have included the disappear-
ance of Swissair and Sabena (the former Belgian flag carrier) in 2001, as
well as Malev Hungarian Airlines and Spanair, both within one week in
February of 2012. If we trace the bankruptcy events on both sides of the
Atlantic since 2000, we will notice that bankruptcies are more common on
the US market. In fact, most of the large US network carriers have gone
bankrupt at least once since 2000 (with Southwest Airlines being a notable
exception). At the same time, larger airline brands on the US market

M4371-FINGER_9781786431851_t.indd 143 14/11/2017 11:14

(Northwest, Continental, and US Airways) disappeared via mergers rather

than bankruptcies. The US Bankruptcy Code provides for the option
to request bankruptcy protection under Chapter 11 – something that is
not generally available in Europe. Under this arrangement, a firm that is
unable to cover its current payments can continue operating while under-
going restructuring under the supervision of a Bankruptcy Court judge.
However, the performance of the European market does differ from
that of the US airline industry in several important ways. First, unlike
their US counterparts, European incumbent carriers have not moved to
adjust their network structure. Establishment of hub-and-spoke networks
has been one of the major innovations introduced by the US airlines
post-deregulation. In Europe, however, the incumbent airlines’ network
structure has not changed much. The carriers have essentially expanded
their networks from the same key airports as they did prior to liberaliza-
tion. As most of Europe’s incumbent carriers have traditionally been the
respective countries’ flag carriers, they have naturally developed their hubs
in their countries’ capital cities (with the exception of the German carrier
Lufthansa, which developed its hubs at Frankfurt and Munich airports).
Interestingly, none of Europe’s legacy carriers has been able to establish
a base outside of its home country (or region, in the case of SAS). The
only such attempt was made by Lufthansa, which set up a subsidiary in
Italy (the carrier, named Lufthansa Italia, used Milan Malpensa airport
as its base of operations, following Alitalia’s closure of its hub at this
gateway). However, Lufthansa’s Italian venture lasted for only three
years.
The second important difference between outcomes of the EU and the
US market liberalization is in the way the low-cost airlines’ strategies have
developed. For starters, there are many more low-cost carriers in Europe
than in the United States (there are generally more airlines in Europe than
in the US market). Most of the European LCCs have pursued strategies
involving exclusive reliance on point-to-point operations, the absence
of customer loyalty programmes, and a general focus on leisure traffic
(although this has been changing recently). The use of secondary airports
is another strategy typically associated with low-cost airlines – airport
charges tend to be significantly lower at those airports, which enables the
airlines to keep costs in check. Also, small remote gateways tend to be less
congested, allowing for better on-time performance, an important factor
for airlines trying to maximize their aircraft utilization and achieve aircraft
turnaround times as low as 20 minutes. Key LCC players in the United
States (Southwest and JetBlue) tend to offer better amenities to their pas-
sengers than their European counterparts; at times their offerings might be
even better in some dimensions than those of large US network airlines.

M4371-FINGER_9781786431851_t.indd 144 14/11/2017 11:14

For instance, it was American Airlines, not Southwest or JetBlue, that first
introduced fees for a passenger’s first checked bag. In fact, Southwest cur-
rently has the most generous checked baggage policy on the US market:
the airline does not impose charges on the first two checked bags. Also, all
US low-cost carriers not only offer point-to-point, but also connecting ser-
vices, and even the self-proclaimed ultra-low-cost Spirit Airlines operates
a frequent flier programme (something that none of the top-five European
LCCs offer).
We will now examine the dynamics of the European airline industry over
recent years. Passenger volume in the European Union has recovered from
the slump caused by the 2008 financial crisis. Overall, about 918 million
passengers travelled on commercial passenger flights within the 28 EU
member states in 2015, according to Eurostat. This represents a 15 per cent
increase compared to 2008.
Traffic growth (both over the last dozen years and since the latest finan-
cial crisis) has been uneven across the European countries, as Figure 8.1
clearly demonstrates. The cumulative growth reported here is for passenger
traffic within the EU/EEA.
Looking at the time since the 2004 EU enlargement, four Eastern
European countries (Latvia, Poland, Lithuania and Romania) emerge
as clear leaders in air passenger traffic growth, with Iceland and Estonia
being the two other countries that have been able to more than double
their passenger numbers between 2004 and 2015. However, it would not
140
120
100
80
60
40
20
0
Iceland
Lithuania
Romania
Luxembourg
Poland
Malta
Croatia
Portugal
Belgium
Latvia
Norway
Switzerland
Netherlands
Greece
Sweden
Denmark
Hungary
Italy
Estonia
Bulgaria
Finland
Germany
France
Austria
United Kingdom
Spain
Cyprus
Ireland
Czech Republic
–20
Slovenia
Slovakia
–40
Figure 8.1 Cumulative percentage growth in air passenger traffic, 2008–15

(Eurostat data)

M4371-FINGER_9781786431851_t.indd 145 14/11/2017 11:14

be entirely correct to say that Eastern European countries have clearly

out-performed their Western counterparts as far as passenger air traffic
growth is concerned: Hungary, Slovenia and Czech Republic have exhib-
ited below-median cumulative growth.
The 2008 financial crisis has also affected the European countries dif-
ferently. Interestingly, of all the EU/EEA states, Iceland showed the most
impressive cumulative growth in air traffic over the 2008–15 period. Apart
from Iceland, only four other states (Lithuania, Romania, Luxembourg
and Poland) managed to achieve over 50 per cent cumulative passen-
ger growth over the same time period. Four countries (Ireland, Czech
Republic, Slovenia, and Slovakia) have demonstrated negative cumulative
growth. Interestingly, Greece’s air traffic has not been affected as much,
despite lingering economic woes, as the tourism industry has remained
relatively strong in that country.
Even within different groups of countries, we sometimes see diverging
trends in air passenger traffic. While the largest Western European markets
have demonstrated generally similar growth patterns (Figure 8.2), the
situation is different for Eastern Europe. As can be seen from Figure 8.3,
Hungary, Poland, and Czech Republic had comparable passenger volumes
when the countries entered the European Union. What followed was rapid
growth in the Polish market, with modest development in the other two
countries. As a result, by 2015, Poland – which had smaller passenger
volumes in 2004 than Czech Republic and Hungary – had more than
twice the passenger traffic of those other two countries. Figure 8.4 clearly
250
200
150
Millions
100
50
0
2002 2004 2006 2008 2010 2012 2014 2016
Germany Spain France Italy United Kingdom
Figure 8.2 Passenger dynamics, largest Western European markets

(Eurostat data)

M4371-FINGER_9781786431851_t.indd 146 14/11/2017 11:14

35
30
25
20
Millions
15
10
0
2002 2004 2006 2008 2010 2012 2014 2016
Czech Republic Hungary Poland
Figure 8.3 Passenger dynamics, largest Eastern European markets

(Eurostat data)
4
Millions
0
2002 2004 2006 2008 2010 2012 2014 2016
Estonia Latvia Lithuania
Figure 8.4 Passenger dynamics, Baltic States (Eurostat data)
shows that in the three Baltic States (Latvia, Estonia and Lithuania),
the divergence across the countries has been even more dramatic. The
countries entered the EU in 2004 with nearly equal annual passenger
volumes. Eleven years later, Latvia’s traffic volume nearly quintupled and
Lithuania’s quadrupled, while Estonia’s ‘merely’ doubled.
What are the key players on the European airline market, and has any-
thing changed on this scene over the last decade? The short answer is that
while not much has changed, there have been some quite tectonic shifts.
Indeed, the top five European airlines in 2016 are the same as in 2006. At

M4371-FINGER_9781786431851_t.indd 147 14/11/2017 11:14

140
120
100
80
60
40
20
0
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Ryanair Lufthansa IAG Air France-KLM easyJet
Note: Passenger volume is in millions per annum and only includes intra-European traffic.
Figure 8.5 Top European airlines by passenger volume (Eurostat data)
the same time, only one carrier in the top five preserved the same rank in
2016 as it had in 2006. Air France-KLM dropped from the top spot in 2006
to fourth in 2016, while Ryanair rose from fourth to first. Lufthansa and
the IAG (with British Airways a key airline here) swapped second and third
positions, while easyJet remained Europe’s 5th largest carrier. Figure 8.5
illustrates this situation and also demonstrates the extent to which Europe’s
largest carriers have grown over the last decade. In 2006, none of the top
airlines carried over 80 million passengers. In 2016, however, all but one
exceeded 90 million.
While Ryanair and easyJet – Europe’s leading low-cost airlines – firmly
occupy spaces in the top five, with the former having become Europe’s
largest carrier in 2016, the big three network carriers have not lost their
place on the market, managing healthy growth rates. Among the next five
largest airlines (Figure 8.6), we see two legacy network carriers (SAS and
Alitalia) and three airlines that are normally associated with the low-cost
business model (Norwegian, Air Berlin and Wizz Air). At the same time,
the business models of Norwegian and Air Berlin tend to be somewhat
different from that of Wizz Air, as the former operates more of a hybrid
network (offering limited connecting services while still mostly relying
on point-to-point traffic). Note also that the airlines in this group carry
about one-third of the passenger traffic as compared to the top five carri-
ers. Overall, the picture is clear: the development of the European airline
market post-liberalization has seen explosive growth of LCCs; however,
it appears that the legacy carriers have been able to find ways to compete

M4371-FINGER_9781786431851_t.indd 148 14/11/2017 11:14

35
30
25
20
15
10
0
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Norwegian Air Shuttle Air Berlin Group SAS Group

Alitalia Wizz Air
Note: Passenger volume is in millions per annum and only includes intra-European traffic.
Figure 8.6 Top 6–10 airlines by passenger volume (Eurostat data)
with them. We will take a closer look at the LCCs’ business model in the
next section.
LOW-COST CARRIERS: WHAT ARE THEY ANYWAY?
Explosive growth of the so-called low-cost carriers (LCCs) worldwide

raised the question of whether this new breed of airlines will dominate the
skies in the future. As a matter of fact, the airlines commonly included in
the LCC group have achieved an approximate 35 per cent passenger share
on the European market. I am deliberately using somewhat sceptical lan-
guage when talking about this category of airlines due to the fact that we
do not actually have a clear definition of what an LCC is. No expert will
be able to point to the value of cost per available seat mile (a conventional
cost measure in the airline industry), which would separate the LCCs
from the non-LCCs. For me, a classification of airlines by their network
strategy (I tend to distinguish between network, point-to-point and hybrid
carriers) seems most appropriate. Moreover, the airline’s choice of whether
to operate a proper network or focus on point-to-point services only can
explain part of the cost differences between the carriers.
So what are the sources of LCC cost advantages? First of all, possibili-
ties for fuel savings are limited, other things being equal. In other words,

M4371-FINGER_9781786431851_t.indd 149 14/11/2017 11:14

the fuel consumption of a Boeing 737–800 does not depend on which

airline is operating the aircraft. Using newer aircraft can help, however, as
these airplanes are more fuel efficient. Operational savings due to techno-
logical advances are available and accessible to any airline. However, full
service carriers’ (FSCs) operations are considerably more complex than
those of LCCs. The main difference is the fact that FSCs operate networks
that require more sophisticated technology to ensure that passengers and
their luggage are able to connect between flights in the airline’s network,
whereas most airlines classified as LCCs only perform point-to-point
operations.
Network complexity also necessitates using different aircraft types,
which implies that FSCs will have higher maintenance costs than LCCs,
other things being equal. Use of a single aircraft type will generally limit
the kinds of markets an airline will be able to serve: if your fleet only
includes Boeing-737 aircraft, for instance, you will likely be unable to serve
smaller cities profitably.
Airlines that have been operating in the pre-liberalization era tend also
to have more complex and rigid labour relationships than younger carri-
ers. This will mean higher labour cost and/or less flexibility when an FSC
has to come up with ways to generate savings in this category. Established
network carriers often have to deal with strong labour unions, which some-
times do not hesitate to use industrial actions to obtain a better deal with
their airline.
Some budget airlines tend to fly into secondary airports to save on
airport charges. However, the downside of this strategy is that flying into
secondary airports is less attractive to business travellers.
Ticket distribution is another area where LCCs generate some cost
savings. By relying predominantly on their websites to sell their tickets,
LCCs obviously save on commission to global distribution systems and
travel agents. Moreover, a prospective customer checking price quotes on
an airline website will not immediately see what the competitors might
offer – something an airline can use to its advantage.
I think the main conclusion from this discussion of cost differences
between LCCs and FSCs is that the LCCs’ lower costs are driven predomi-
nantly by the kind of product they offer and the market segment they serve
with their product. Not having to operate a complex network comes with
cost savings, but makes the product less attractive to price-insensitive busi-
ness passengers. Furthermore, the main reason why Lufthansa will not be
able to lower its operating costs to match those of easyJet is that doing so
would require the German carrier effectively to abandon its most lucrative
market segment. On the other hand, if easyJet wanted to really compete
with Lufthansa for its business traffic, the LCC would have effectively to

M4371-FINGER_9781786431851_t.indd 150 14/11/2017 11:14

increase the quality of its product. Would easyJet be able to run a proper
network at a lower cost than Lufthansa? It probably would. However, the
difference in conventional cost measures between the two carriers would
definitely become less dramatic.
LOOKING AHEAD
At first, my intention for this part of this chapter was to present my
vision of the European airline industry 10–15 years from now. However,
the events of the second half of 2016 brought about some more urgent
concerns. Specifically, on 23 June 2016 voters in the United Kingdom sup-
ported the idea of their country leaving the European Union. While the
referendum was an advisory one, the UK Government looks determined
not only to act upon the recommendation, but also to take the country out
of the EU Single Market. As of the time of this writing, it appears that the
United Kingdom will leave the EU by the spring of 2019. This has created
significant challenges for the European market and beyond.
Essentially, if the UK decides not to leave the Single Market, nothing
will change for UK–Europe (and the transatlantic) traffic. Consequences
of the United Kingdom’s departure from not only the European Union
but also from Europe’s common aviation market will, of course, depend
on whether and what kind of agreement the UK will be able to reach
with the EU. In the worst-case scenario, we can return to the network of
bilateral air service agreements between the UK and the corresponding
European countries. These agreements will likely come with nationality
clauses, creating difficulties for the point-to-point LCCs and their custom-
ers. As I noted above, former flag carriers have not been able to establish
an appreciable presence outside of their home countries. Thus most, if not
all, of the flights by those carriers are operated from their home countries,
which means that satisfying the nationality clauses of any future bilateral
agreements should not be a problem for them. The situation will be even
easier if the UK and EU manage to agree an open-skies deal similar to the
one the EU currently has with some countries in North Africa (such as
Morocco).
However, the situation with the point-to-point LCCs can get rather
messy. For instance, Ryanair is an Irish-registered airline. So, without
an open-skies deal with the EU, the airline might be restricted to flying
between the UK and the Republic of Ireland. Ryanair has already
announced that it has no intention of registering a UK subsidiary, so
even if a UK–EU open skies deal happens, Ryanair will most likely dis-
continue serving the UK domestic market (unless, of course, the UK

M4371-FINGER_9781786431851_t.indd 151 14/11/2017 11:14

Government decides to apply loose foreign ownership restrictions post-

Brexit and allows airlines that are majority-owned by foreign citizens to
operate domestic routes). The situation with easyJet is unclear as well.
While this carrier is registered in the United Kingdom (easyJet also has
a Swiss-registered subsidiary – recall that Switzerland is not a part of the
EU Single Market), the airline’s founder holds dual nationality (Cyprus
and the UK). As a result, determining whether the carrier will satisfy
nationality clauses and the EU foreign ownership restrictions is not a
straightforward matter. Moreover, if Ryanair’s ownership structure does
not change, the carrier risks being in breach of the EU foreign ownership
restrictions following the UK’s departure from the EU and the common
market (jointly, UK and non-EU shareholders owned more than 50 per
cent of this airline at the time of this writing). It is currently unclear how
this situation will be resolved.
The future of the IAG could also be jeopardized by the UK’s departure
from the European Union. The IAG was formed after British Airways’
merger with Iberia (Spain’s former flag carrier). Aer Lingus and Vueling
(airlines registered in Ireland and Spain, respectively) are currently also
IAG subsidiaries. Since British Airways owns 55 per cent of IAG, the
airline will not in its present form be able to operate intra-EU flights
should the United Kingdom leave the EU common market. It is cur-
rently unclear how this issue will be resolved. While intra-EU mergers are
not uncommon (as I noted above, Air France and KLM operate as one
company, and Lufthansa owns a number of smaller European carriers),
and ownership of European carriers by non-EU investors (within the 49
per cent EU ownership restriction) is also commonplace, a merger involv-
ing an EU and a non-EU airline is not feasible in the current regulatory
environment. One potential institutional arrangement that the IAG could
pursue is to form a metal-neutral joint venture similar to those operating
on the transatlantic market (this venture would only cover UK–Spain and
UK–Ireland flights); however, the scale of operations does not appear suf-
ficient for such an arrangement to be viable.
The potential effects of Brexit outside the European market have to do
with the fact that departure of the UK from the European common airline
market will also disrupt the transatlantic Open Aviation Area that has
been in place since 2008. This area is essentially an open skies arrangement
between the EU and the USA, which replaced a set of respective bilateral
air service agreements. Negotiations that preceded the Open Aviation Area
revolved largely around the US–UK segment of the transatlantic market.
In fact, over one-third of the passengers boarding transatlantic flights in
the United States land at UK airports, with Heathrow taking in the lion’s
share of this traffic. Whether the US and the UK will be able to come up

M4371-FINGER_9781786431851_t.indd 152 14/11/2017 11:14

with an open-skies-type arrangement for air services between their coun-

tries or whether they will revert to the previous agreement (with restricted
access to Heathrow airport) is anyone’s guess. In either case, however, one
European airline currently building up its presence on the transatlantic
market will be adversely affected should the UK leave Europe’s common
market. Norwegian is preparing to launch a significant number of new
transatlantic services from both the United Kingdom and Ireland. As the
new legal entity that will operate these services will be registered in Ireland,
it will not be able to operate under the new US–UK institutional arrange-
ments, which will most likely include the nationality clause in some form. I
am generally more optimistic about the survival of the transatlantic open
aviation area post-Brexit, despite the strong protectionist stance of the
current US Administration.
Overall, the European airline industry remains an excellent example of
successful liberalization of international airline markets within a region. I
hope that this will serve as an example for other regions, leading to more
places where the forces of competition will be unleashed to the benefit of
the travelling public. Over the course of two decades or so, the European
airline industry evolved from a set of markets with restricted competition
and high fares to a vibrant industry with a diverse set of competitors that
offer various travel options to their passengers. Network carriers connect
Europe’s regions with frequent flights and convenient connections at their
hubs, offering premium services to their most loyal customers. On the other
hand, point-to-point low-cost carriers offer affordable no-frills options.
Challenges posed to European aviation by the impending departure of the
United Kingdom from the EU are not trivial, but are not critical either.
Overall, my assessment is that UK aviation has more to lose from Brexit
than the overall European airline industry.
If the EU is to remain otherwise intact, European regulators will need to
address the following key issues as far as further development of European
civil aviation is concerned. First, European skies are a fragmented mess
in need of consolidation. The idea of creating Functional Airspace
Blocks to replace the current set-up of a couple of dozen air navigation
service providers will likely meet stiff resistance from the relevant interest
groups. Second, the EU will need to work towards further liberalization
of EU–rest-of-the-world markets. This involves establishing multilateral
open-skies-type agreements between the EU and other countries, similar
to the transatlantic open aviation area currently in place. Progress towards
achieving this goal has been painfully slow, and I am afraid that delays
have sometimes been due to intra-European issues. For instance, it is
ridiculous that the issue of Gibraltar appears to be hampering progress
towards the EU–Ukraine open skies deal. This is a symptom of a more

M4371-FINGER_9781786431851_t.indd 153 14/11/2017 11:14

general problem with EU bureaucracy. Last but not least, for a liberalized
market to thrive, everyone must play by the same rules. Therefore, the
European Union must continue working to ensure a level playing field for
all competitors. This includes absence of subsidies and adherence to fair
competition practices by all market players.
REFERENCES
Albalate, D., G. Bel and X. Fageda (2015), Competition and cooperation between
high-speed rail and air transportation services in Europe, Journal of Transport
Bachis, E. and C. Piga (2011), Low-cost airlines and online price dispersion,
International Journal of Industrial Organization, 29, 655–67.
Behrens, C. and E. Pels (2012), Intermodal competition in the London–Paris pas-
senger market: high-speed rail and air transport, Journal of Urban Economics,
71, 278–88.
Bel, G. and X. Fageda (2010), Privatization, regulation and airport pricing: an
empirical analysis for Europe, Journal of Regulatory Economics, 37, 142–61.
Bergantino, A.S. and C. Capozza (2014), Airline pricing behaviour under limited
intermodal competition, Economic Inquiry, 53, 700–713.
Bilotkach, V., J.A. Clougherty, J. Mueller and A. Zhang (2012), Regulation, privati-
zation, and airport charges: panel data evidence from European airports, Journal
of Regulatory Economics, 42, 73–94.
Bilotkach, V., A. Gaggero and C. Piga (2015a), Airline pricing under differ-
ent market conditions: evidence from European Low-Cost Carriers, Tourism
Bilotkach, V., S. Gitto, R. Jovanovic, J. Mueller and E. Pels (2015b), Cost efficiency
benchmarking of air navigation service providers, Transportation Research Part
A, 77, 50–60.
Bilotkach, V., J. Mueller and A. Nemeth (2014), Estimating the consumer welfare
effects of de-hubbing: the case of Malev Hungarian Airlines bankruptcy,
Transportation Research Part E, 66, 51–65.
Burghouwt, G. and J. de Wit (2005), Temporal configurations of European airline
networks, Journal of Air Transport Management, 11, 185–98.
Button, K. and R. Neiva (2014), Economic efficiency of European air traffic
control systems, Journal of Transport Economics and Policy, 48, 65–80.
Calzada, J. and X. Fageda (2012), Discounts and public service obligations in
the airline market: lessons from Spain, Review of Industrial Organization, 40,
291–312.
Dobruzskes, F. (2006), An analysis of European low-cost airlines and their net-
works, Journal of Transport Geography, 14, 249–64.
Dobruzskes, F. (2011), High-speed rail and air transport competition in Western
Europe: a supply-oriented perspective, Transport Policy, 18, 870–79.
Dobruzskes, F. (2013), The geography of European low-cost airline networks: a
contemporary analysis, Journal of Transport Geography, 28, 75–88.
Fageda, X. (2006), Measuring conduct and cost parameters in the Spanish airline
market, Review of Industrial Organization, 28, 379–99.

M4371-FINGER_9781786431851_t.indd 154 14/11/2017 11:14

Giaume, S. and S. Gillou (2004), Price discrimination and concentration in

European airline markets, Journal of Air Transport Management, 10, 305–10.
Morrison, S. and C. Winston (1986), The Economic Effects of Airline Deregulation,
Washington, DC: Brookings.
Obermayer, A., C. Evangelinos and R. Pueschel (2013), Price dispersion and com-
petition in European airline markets, Journal of Air Transport Management, 26,
31–4.

M4371-FINGER_9781786431851_t.indd 155 14/11/2017 11:14

9. Latin America and the Caribbean,
thirty-plus years of lukewarm
liberalization of air transport markets
Henry L. Vega
INTRODUCTION
Demand for air transport in the Latin America and Caribbean (LAC)
region has increased significantly since the 1990s, and even more so in the
twenty-first century. The increase has had its basis in the internationaliza-
tion of production, economic growth and, possibly, efforts to deregulate
air transport markets throughout the region. This chapter is concerned
primarily with an assessment of evidence in existing academic studies and
industry reports on the effects of deregulation of air transport markets in
the region. To narrow the geographic scope, only developments in sover-
eign countries are highlighted. This list is included in Table 9.1.
With the exception of Chile, LAC countries have found air transport
liberalization challenging, even after having been long exposed to the US’s
trade policies in the region. Indeed, the LAC’s most important trading
Table 9.1 Sovereign states of Latin America and the Caribbean
Antigua and Barbuda Dominica Nicaragua

Argentina Dominican Republic Panama
Bahamas Ecuador Paraguay
Barbados El Salvador Peru
Belize Grenada Saint Kitts and Nevis
Bolivia Guatemala Saint Lucia
Brazil Guyana Saint Vincent and the Grenadines
Chile Haiti Suriname
Colombia Honduras Trinidad and Tobago
Costa Rica Jamaica Uruguay
Cuba Mexico Venezuela
Source: Organization of American States.
156

M4371-FINGER_9781786431851_t.indd 156 14/11/2017 11:14

Latin America and the Caribbean 157
partner, the US, has pushed air markets’ liberalization in the form of Open
Skies Agreements (OSA) since the early 1990s. To illustrate the slow pace
of liberalization in the region, only 12 OSAs were signed between the US
and the 33 LAC sovereign countries in the Americas by September 2009;
by 2016 this had only increased to 18 OSAs.
OSAs in the region and other efforts to liberalize the provision of air trans-
port services have helped remove economic regulations, but the details of this
are poorly documented. Using a case study approach, this chapter consoli-
dates the information available to expose the implications of liberalization of
the LACs’ transport markets. Descriptive statistics are used to assess the state
of these markets in the region. The focus is mostly on market and economic
regulations. Less emphasis is given to the liberalization of technical and
safety oversight because these functions have largely remained in the hands
of the governments. Mexico, Chile and Colombia – partly because of the
scale of their markets – are used as the case studies. Additionally, in selecting
these case studies, consideration was given to the traffic composition split
between domestic and international passengers, as was the fact that Chile
was a pioneer in liberalizing its air transport sector whereas Mexico and
Colombia have had a lukewarm relationship with liberalization.
OVERVIEW OF THE AIR TRANSPORT MARKET IN

THE REGION
Despite the deregulation of the airline industry in regions around the

world, access to aviation markets in the LAC region remains a contentious
issue. Although LAC countries have faced similar fiscal constraints they
have often differed in their approach to liberalization of aviation markets.
Decades after the 1944 Chicago Convention, most South American coun-
tries still restrict access to their airspace, particularly for non-scheduled
services. Dozens of bilateral air service agreements (ASAs) have been
signed between LAC countries, and many others between trading groups
such as the Andean Community of Nations (CAN) or the Southern
Common Market (MERCOSUR). Dismally, the scope of the permission
of access to countries’ airspaces has been narrow and subject to which
‘freedoms of the skies’ are included in the region’s numerous bilateral
ASAs, and most importantly, subject to political circumstances and sec-
toral interests often intended to protect national carriers and increase
government revenues.
Regarding traffic of passengers, Table 9.2 presents a picture of the evo-
lution in the size of this market in the region. As can be seen, the Brazilian
market was about 2.5 times that of Mexico in 2015, although much of this

M4371-FINGER_9781786431851_t.indd 157 14/11/2017 11:14

Table 9.2 Air transport, passengers carried, selected LAC countries, 1985–2015
Country Name 1985 1995 2005 2015 Annual Growth (%)

Brazil 13 402 900 20 196 100 37 661 733 102 039 359 22.0
M4371-FINGER_9781786431851_t.indd 158
Mexico 15 364 400 14 968 800 21 857 658 45 560 063 6.6
Colombia 5 737 300 7 863 300 9 984 424 30 742 928 14.5
Chile 825 200 3 197 200 5 939 020 15 006 762 57.3
Argentina 4 713 200 6 641 700 6 938 436 14 245 183 6.7
Peru 1 563 900 2 508 300 4 332 223 13 907 948 26.3
Panama 390 300 661 200 1 795 958 12 018 103 99.3
Venezuela 4 966 500 4 445 000 5 043 028 6 456 854 1.0
Ecuador 664 200 1 670 900 2 011 004 5 762 485 25.6
Trinidad and Tobago 1 299 500 1 727 000 1 055 106 2 617 843 3.4
158
El Salvador 395 000 1 698 000 2 540 564 2 597 649 18.6
Bolivia 1 343 300 1 223 700 1 892 343 2 578 960 3.1
Costa Rica 310 000 870 400 953 217 1 617 076 14.1
Cuba 894 400 823 700 812 781 1 294 458 1.5
Antigua and Barbuda 500 000 1 050 000 778 260 1 039 810 3.6
Belize 935 604 110.0
Bahamas, The 860 500 936 900 1 020 000 587 517 −1.1
Paraguay 212 000 105 200 445 854 452 004 3.8
Suriname 108 000 161 800 315 410 259 682 4.7
Honduras 451 000 251 150 −1.5
Guatemala 108 400 300 000 93 129 −0.5
Total 54 110 000 71 049 200 105 377 019 260 064 567

Source: World Bank’s World Development Indicators database.
14/11/2017 11:14
was to link its sparsely inhabited areas with its major economic centres (US
Department of Commerce, 2006).
As in other regions of the world, the data presented in Table 9.2 exposes
the key role that air transport plays in facilitating population movements
and the integrating of people in spatially dynamic regional and global
labour markets. As pointed out by Button and Vega (2008), air transport
has not only made long-distance initial permanent migration less costly,
but has also contributed to limit social costs by allowing regular visits to
friends and family.
The markets for air transport services have developed rapidly since
the mid-1990s, to the point where the provision of air services is a major
enabler of global supply chains. Connectivity of employees operating
from different hubs around the world and the integration of the electron-
ics, high-tech goods, pharmaceuticals, fashion products, perishables and
exotics (P and E) are quite unimaginable without access to air transport.
Vega (2008 and 2010) investigated in detail trade flows and air transport
costs of high-tech goods and P and E and provided definitions and char-
acteristics of their commerce. The list of exotics is extensive and includes
the legal trade of items such as tropical fruits and vegetables, ornamentals,
fresh herbs and spices, wild mushrooms, fish and seafood.
In LAC countries, because of economies of scope, airfreight plays a role
in allowing trade among countries with different capacities for production
but with similar tastes. For instance, air transport has allowed countries
such as Colombia and Ecuador, Chile and Peru to engage in the trade
of exotics and other perishable products in which these countries have a
comparative advantage. Similarly, in MERCOSUR countries, Martínez-
Zarzoso and Nowak-Lehmann (2004) have found that air transport offsets
the effect of geographic distance having only a low or statistically non-
significant effect on these countries’ exports of fish, meat, vegetables and
fruits to the European Union (EU). Evidently, this is partly also the result
of more diverse populations, mainly in the US and the EU, demanding and
willing to pay for more expensive exotic groceries such as starfruit, jicama,
pitahaya, tropical flowers and so on.
Large islands and landlocked countries have benefited from air trans-
port, as they have seen their production feasibility frontiers expand with
access both to remote suppliers, domestic and international, and to cus-
tomers. Table 9.3 shows the magnitude of the flows of goods that are
moved by air from the LAC region to the EU.
As expected, Brazil and Mexico, the LAC region’s largest economies,
register the highest freight exchanges with the EU. Argentina, the region’s
third largest economy appears in fifth place. Colombia and Ecuador, due
to their exports of flowers and other P and E, appear in third and fourth

M4371-FINGER_9781786431851_t.indd 159 14/11/2017 11:14

Table 9.3 LAC–EU freight and mail, loaded and unloaded, metric tons
Year 2005 2007 2009 2011 2013 2015

Brazil 188 897 214 968 182 061 284 208 264 655 256 527
Mexico 87 388 96 117 102 689 152 562 150 486 166 059
Colombia 37 294 31 467 51 197 52 940 48 267 62 963
Ecuador 29 101 37 719 55 206 52 132 58 579 54 220
Argentina 51 875 61 223 52 824 57 238 57 709 48 491
Chile 32 755 32 820 36 227 38 840 38 805 38 918
Peru 23 663 24 942 26 637 32 812 30 209 31 107
Dominican Republic 18 550 20 355 19 836 21 174 19 107 27 998
Cuba 11 126 9634 9467 11 799 10 482 14 773
Venezuela 23 165 22 363 20 299 18 733 18 611 12 956
Panama 2614 2942 9112 8861 9912 12 523
Jamaica 5004 5949 4171 4641 4631 5271
Costa Rica 4913 5295 4411 6639 4511 5024
Uruguay 4644 5850 5695 6034 5149 4987
Barbados 14 823 16 817 6451 6252 6816 4869
Suriname 4083 3656 4252 4734 4246 4821
Trinidad and Tobago 3462 1887 3204 3113 3526 4704
Guatemala 1286 243 371 1683 2952 4458
Saint Kitts and Nevis 23 43 448 752 852 1042
Antigua and Barbuda 404 418 538 707 565 681
Grenada 180 249 656 766 274 598
Bolivia 41 443 372 486 293 428
El Salvador 0 0 11 1695 0 318
Saint Lucia 1483 1366 725 798 395 216
Paraguay 0 0 0 0 0 142
Honduras 0 0 1 9 98 109
Bahamas 86 82 56 0 308 94
Haiti 29 63 38 211 207 44
Nicaragua 0 0 28 0 0 38
Dominica 26 29 28 31 42 25
Total 546 915 596 940 597 011 769 850 741 687 764 404
Source: Eurostat.
place, respectively. From the data, it also seems that the annual growth rate
in freight levels was linked to the region’s economic expansion with the
commodity boom after the year 2005, followed by collapsing levels related
to the financial crisis of 2008.
High freight costs are often the result of imbalanced freight flows, since
a relevant factor determining airfreight rates is the peak load factor. While
passenger business is generally bidirectional, cargo is not. This implies

M4371-FINGER_9781786431851_t.indd 160 14/11/2017 11:14

that when transporting goods from point A to point B, the freight rate
charged must also cover the return trip from B to A. When the demand for
transport services is unidirectional, freight rates are simply higher on one
direction, as the shipper pays for forgone capacity on either the inbound
or outbound flight. When the trade imbalance is high (import/export
ratio higher than one), transport costs for exports tend to be higher. LAC
countries have struggled to unmask the factors behind the region’s notori-
ously high airfreight rates. Vega (2010), for example, found that there is
not a clear relationship between distance and the cost of airfreight ser-
vices. Vega (2010) also found that liberalization of air transport markets,
through OSAs, reduced airfreight charges for countries exporting P and E.
Most interestingly, the costs of air shipping varied widely across countries,
regardless of proximity to the US market. Previously, Vega (2008) had
found that neighbours Ecuador and Colombia had very different freight
rates for their exports of fresh flowers: transport costs tended to be 10 to
20 per cent higher for Ecuador compared to Colombia.
In the LAC region, for economic growth reasons, reducing air transport
costs should have been a policy priority many years ago. Unfortunately, it
has not and this has had a negative effect on the movement of P and E,
which make up a high proportion of air cargo exports from South America
to the US (347 000 tons in 2006) and to the EU (100 000 tons in 2006)
(Vega, 2008).
INSTITUTIONAL CONSTRAINTS OF THE MARKET

IN THE LAC REGION
Overall, government interventions in the air transport market can be

grouped into two approaches: protectionist and liberalized. Protectionism
is often justified because governments need to prevent foreign airlines
from taking over the domestic market and assume, from an economic per-
spective, that air transport provision is a public utility. However, Hanlon
(2007) has explained that air transport, strictly speaking, is not a public
utility. Public utilities are considered those exhibiting the characteristics of
a natural monopoly, in which the barrier imposed by size is so great that
a service can only be provided at least cost if it is supplied by one – and
only one – firm. The advent of low-cost carriers (LCCs) in developed and
developing countries provides strong evidence suggesting the opposite. A
liberalized approach refers to absence of government intervention in the
regulation of airport infrastructure, allocation of slots, routes, airlines’
ownership and barriers to entry, among other elements of the aviation
market.

M4371-FINGER_9781786431851_t.indd 161 14/11/2017 11:14

AIRPORTS IN THE LAC REGION
The lack of airport infrastructure is a concern in the LAC region, which

has experienced rapid economic growth rates in the twenty-first century.
As in other developing regions, the difficulty in maintaining an adequate
supply of infrastructure might be related to what Forsyth (2007) described
as the short-run problem, as it is difficult to expand airport capacity in
short periods of time. In opposite circumstances, excess capacity has also
become a problem in the region. Expanding infrastructure to deal with the
limitations on the number of slots has at times been pursued on economic
development grounds. Often, long-term infrastructure projects have been
in disconnect with policy-makers’ priorities. At their worst, these efforts
have failed to attract new customers, leading to excess infrastructure and
abuse of the system by carriers. In Brazil, for instance, excessive ware-
housing infrastructure at São Paulo-Guarulhos International Airport was
being used at very low cost for months as temporary storage by free-riding
shippers.
Table 9.4 lists the LAC region’s top airports by passenger movements,
aggregating both domestic and international flights, as well as separating
the two categories. The importance of Brazil is highlighted when it comes
Table 9.4 Top 15 airports in the LAC region, all flights and international
flights only (2015)
Airport City Country Total Flights International

Flights Only
1 MEX Mexico City Mexico 388 732 109 696
2 BOG Bogota Colombia 278 438 78 992
3 GRU São Paulo Brazil 271 324 76 480
4 CGH São Paulo Brazil 173 882
5 BSB Brasilia Brazil 157 390
6 LIM Lima Peru 150 004 64 390
7 PTY Panama City Panama 133 394 131 860
8 CUN Cancun Mexico 131 826 86 658
9 SJU San Juan Puerto Rico 128 732 43 274
10 GIG Rio de Janeiro Brazil 122 494
11 VCP São Paulo Brazil 122 118
12 SCL Santiago Chile 121 936 55 754
13 AEP Buenos Aires Argentina 113 600 55 226
14 CNF Belo Horizonte Brazil 109 650
15 SDU Rio de Janeiro Brazil 106 940
Source: Adapted from ALTA (2016).

M4371-FINGER_9781786431851_t.indd 162 14/11/2017 11:14

Table 9.5 Private participation in airport infrastructure (1991–2015)

(US$ million)
1991–95 1996–2000 2001–05 2006–10 2011–15 Total

Argentina 1614 42 725 2381
Bolivia 17 17
Brazil 2 7 28 254 28 262
Chile 8 286 22 44 964 1322
Colombia 145 201 1000 139 1484
Costa Rica 161 34 195
Dominican Republic 85 85
Ecuador 665 665
Honduras 120 120
Jamaica 175 175
Mexico 1146 595 1576 3317
Peru 8 202 220 257 687
Uruguay 31 164 83 278
Total 184 3638 1866 3605 29 697 38 989
Source: World Bank’s Private Participation in Infrastructure Database. Accessed 6

December 2016 at https://ppi.worldbank.org/data.
to total passengers, but, as can be seen, when separating out international

passengers, much of this is domestic.
Regarding foreign carriers, a constant complaint in the region has been
discrimination in the cost of fuel charged to foreign airlines compared to
the subsidized cost that national carriers receive. Also, Domenech and
Montalvo (2006) have reported that preferential treatment decisions over
ground handling are usually enjoyed by the dominant carrier. Incumbent
carriers are also reportedly at an advantage when it comes to allocation of
slots, since slots are authorized under a process that favours incumbent
carriers through ‘grandfather rights’, ‘use-it-or-leave-it’ principles or night
curfews, particularly at major gateway airports in the LAC region.
Table 9.5 includes the amounts of private investment in airport infra-
structure in the LAC region for the period between 1991 and 2015. Brazil
is the country which has received the largest sums of private investment to
improve its airport infrastructure: about $28.3 billion from 1991 to 2015.
Airport investments are forecast to continue increasing over the next
two decades. It is estimated that for the period 2011 to 2040, these invest-
ments will reach at least $30 billion (in 2009 dollars), about 0.4 per cent of
investments on infrastructure in the LAC region, under a business-as-usual
scenario. Under accelerated growth resulting from overall better prospects
in the global economy, investments on airports could reach $74 billion,

M4371-FINGER_9781786431851_t.indd 163 14/11/2017 11:14

or about 0.6 per cent of total infrastructure investment (Kohli and Basil,
2011).
Airports are largely run by the central government in the LAC region.
They are at times in poor condition and provide inadequate commercial
air transport services. Many have runways that are short or not properly
asphalted, lack adequate passenger and cargo handling facilities, and
provide only limited control and landing aids. In other cases, airports
are used to meet wider government objectives such as a source of foreign
revenue or even to fund general expenditures.1 The case of Chile, where
airport charges were used for general expenditures, is one case cited in
the academic literature. Chile increased passengers’ boarding charges for
international flights to finance campaigns for promoting the country as a
tourist destination and to support a ‘Fight Against Poverty’ fund (Gómez-
Lobo and González, 2008).
To lessen the problems, a common practice in running airports in LAC
countries in recent years has been to hire a concessionaire to manage
them, particularly large ones. In general, concessions have contributed to
improvements in specific aspects of operating an airport, often only the
financial aspects – as a third party is arguably more agile at collecting fees
than governments. In Argentina, for instance, the concessionary system
proved defective, with a single bidder that offered an unrealistic contract
that needed to be restructured afterwards (Lipovich, 2008). Ultimately,
the government of Argentina was forced to commit additional resources
to subsidize the management of smaller rural airports. Another case that
deserves attention is that of the Mexican airports, where only those air-
ports with high passenger and cargo volumes and the best growth expecta-
tions were selected for privatization.2
Regarding productivity, Perelman and Serebrisky (2010) found that in the
LAC region the largest airports increased their productivity between 1995
and 2007, especially those airports handling between 7.5 and 10 million pas-
sengers per year, which posted an annual growth of 5.4 per cent. The authors
also reported that public airports have fared better, on average, than private
ones, experiencing an increase in productivity of 2.9 per cent compared to
0.7 per cent of private airports. However, when the authors looked at the
period between 2003 and 2007, both public and private airports experienced
positive productivity increases. On a more recent variation of the study of
airports’ productivity, Perelman and Serebrisky (2012) found that private
airports performed better than public airports in the period between 2000
and 2007, experiencing an increase in productivity of 2.8 per cent per year.
On the other hand, during the same period publicly-owned airports experi-
enced a decrease in productivity of 0.9 per cent. In contrast to their earlier
research results, the authors also found that it was medium-sized airports –

M4371-FINGER_9781786431851_t.indd 164 14/11/2017 11:14

those handling between 5 and 8 million passengers per year – that experi-
enced the fastest growth in total factor productivity: 1.5 per cent on average.
Another aspect of the concession process over which no definite
outcome can be expected is that of an independent vs. non-independent
regulatory authority. Serebrisky et al. (2011) have reported that evi-
dence on the functioning for independent vs. non-independent airport
regulators did not provide definite conclusions, with each model having
its advantages and disadvantages. Independent authorities scored better
in the transparency of the rule-making and the establishment of a quality
bureaucracy. When it came to autonomy, some government departments
performed better than independent regulatory agencies.
A final issue that has been a major challenge in the history of conces-
sions is renegotiation. Sirtaine et al. (2005) have stated that ‘the short
interval between the granting of a concession and its renegotiations,
about two years, and the outcome of the renegotiation process, makes
the resulting regime a hybrid of price caps (high-powered contracts) and
rate of return (lower-powered contracts)’. More recently, Moore et al.
(2014) have investigated the contracting conditions in the LAC region,
presenting evidence of how private concessionaires ultimately maximize
their profits by increasing their leverage in the regulatory process of the
concession.
AIRLINES IN THE LAC REGION
There is evidence that LAC airlines have performed reasonably when

compared to airlines around the world. For example, Lin (2012), using
data envelope analysis where unity indicates the highest efficiency of the
group, looked at the financial performance and customer service of 38
international airlines and found the following values: Avianca 0.75, Lan
1.00, Aeroméxico 0.93 and Mexicana 0.91. On a more negative note,
Wanke and Barros (2016), using a slightly different technique, found
generally low efficiency across Latin American airlines, which was partly
attributed to the region’s preference for using smaller aircraft with high
operational costs. Speculating about the reasons for this preference, the
authors pointed to causes ranging from insufficient traffic density within
a given route to the scale of the network, which in the case of the LAC
countries corresponds to an insufficient hub-and-spoke system. The
authors also found that public ownership is related to higher levels of
efficiency.
With a few exceptions, LAC countries still impose strict limits on foreign
ownership of airlines. However, a major development in the LAC region

M4371-FINGER_9781786431851_t.indd 165 14/11/2017 11:14

has occurred. A few governments have started requiring airlines to have

only their principal base of business in the designating state instead of
imposing strict quotas on the percentage of nationals owning equity or
serving on the board of directors of an airline (World Economic Forum,
2016). Chile, Costa Rica and El Salvador have adopted this model. Chile
in particular has had success in convincing partners to use a principal place
of business in its bilateral ASAs.
OPEN SKIES AGREEMENTS IN THE REGION
Full liberalization has been a utopia in the negotiations for air access
between LAC countries, and especially between LAC countries and their
northern neighbour, the US. Even though bilateral ASAs must be anchored
in reciprocal rights, unfortunately in the North–South trade, reciprocity is
not perceived as a negotiating point, since one party may be required to
make more concessions than the other simply because air transport indus-
try sizes vary across countries. Policy-makers in LAC countries have often
referred to this as an asymmetry of negotiation condition, an asymmetry
in which they base arguments against liberalization with politically charged
statements such as ‘air carriers in developing countries cannot compete
with those of developed ones’.
Within the LAC region, the CAN adopted a decision to integrate its
air transport sector back in 1991. Members of the Caribbean Community
(CARICOM) signed a multilateral ASA (MASA) in 1996, followed by
additional negotiations. Members of the MERCOSUR signed the Sub-
regional ASA (Fortaleza Agreement) in 1996. In all cases, however, only
limited gains have been achieved, mainly due to the relatively small size
of the intraregional aviation market. Often, new routes made possible by
these agreements do not have the potential to generate economically viable
traffic to guarantee economically sound operations.3 Overall, efforts to
liberalize the airspace of member countries of the CAN, and to a lesser
degree MERCOSUR countries, have gone unnoticed or have been halted
since individual country members are allowed to request exemptions or
can apply their own rules in accordance with other policy priorities, such
as the negotiation of a bilateral trade agreement with a third country.
At the bilateral level between LAC countries and the US, the most liberal
approach has been the negotiation of Open Skies. A major development
of this negotiation is the optional provision for the so-called 7th Freedom
of the Skies or all-cargo rights.4 Overall, OSAs are a welcome development
for users of air services in a region that depends on air transport for its
exports of P and E. Vega (2010), for example, found that developing coun-

M4371-FINGER_9781786431851_t.indd 166 14/11/2017 11:14

Table 9.6 US Open Skies sovereign partner countries as of 14 November

2016
Partner Application Date Concluded All-cargo 7ths*

Costa Rica In Force 8 May 1997 –
El Salvador** In Force 8 May 1997 Yes
Guatemala** In Force 8 May 1997 Yes
Honduras Provisional 8 May 1997 Yes
Nicaragua In Force 8 May 1997 Charter Only
Panama** In Force 5 Aug. 1997 Yes
Chile** In Force 28 Oct. 1997 Yes
Peru** In Force 10 June 1998 Yes
Jamaica In Force 30 Oct. 2002 –
Uruguay** In Force 20 Oct. 2004 Yes
Paraguay** In Force 2 May 2005 Yes
Trinidad and In Force 1 May 2010 Yes
Tobago**
Barbados** In Force 1 July 2010 Yes
Colombia In Force 11 Nov. 2010 –
Brazil N/A 3 Dec. 2010 –
St Kitts** In Force 28 Nov. 2011 Yes
Suriname In Force 21 June 2012 –
Guyana** In Force 25 March 2013 Yes
Notes:
* Optional 7th Freedom All-Cargo Rights provides authority for an airline of one country
to operate all-cargo flights between the other country and a third country, via flights that
are not linked to its homeland.
** Open Skies Agreements currently enforced that include 7th Freedom all-cargo rights.
Source: US State Department.
tries that have negotiated an OSA with the US benefit from an increase
in their volume of exports of P and E of 85.7 per cent, while the value of
their exports increases by 46.9 per cent. In the LAC region, 18 out of 33
countries have signed bilateral Open Skies agreements with the US as of 14
November 2016. Of those 18 countries, 11 have an Open Skies agreement
fully in force that also includes 7th Freedom all-cargo rights, as shown in
Table 9.6. The OSA between the US and Brazil has yet to go into effect.
In contrast to some advances in liberalizing air markets in the Americas,
progress in liberalizing intercontinental markets between the Americas
and other continents is basically nil. Perhaps the European Commission
has made marginal progress in negotiating horizontal aviation agreements
with LAC countries.5 Progress in negotiating comprehensive aviation

M4371-FINGER_9781786431851_t.indd 167 14/11/2017 11:14

a greements, on the other hand, has been met with multiple obstacles; no
such agreement currently exists between any LAC country and the EU.6
As an alternative, the EU has promoted horizontal aviation agreements
in the region, but they do not replace the existing bilateral agreements in
place between EU member states and a third country. Chile was the first
country to successfully negotiate a horizontal agreement with the EU,
in September 2005; however, negotiations of a comprehensive aviation
agreement, that were expected to start right afterwards, have not reached
any major milestone. Uruguay was the second country in the region to
sign a horizontal aviation agreement with the EU, in November 2006. In
February 2007, Paraguay became the third LAC country to sign a horizon-
tal aviation agreement with the EU. Peru and the EU started negotiations
in September 2009. More recently, there was much expectation about the
negotiations of the first comprehensive ASA agreement between the EU
and a South American country, Brazil, in March 2011. This agreement
would have allowed all ‘EU airlines to operate direct flights to any des-
tination in Brazil from any point in the EU (and vice versa for Brazilian
carriers) – without restrictions on routes, prices or the number of weekly
flights’ (European Union, 2017). According to press reports, the agreement
is currently being renegotiated.
SAFETY AND SECURITY REGULATIONS
International agreements and ASAs have assigned the responsibility of

guaranteeing safety and security to national governments.7 Responsibility
is shared with private parties such as airlines, aircraft manufacturers
and freight forwarders. New threats as well as new standards in secu-
rity and safety impose additional pressure on air transport costs. In the
LAC region, the US government – and to a lesser extent the European
Commission – has been keen on requiring compliance with ICAO’s safety
standards. The results of the latest IASA rating of LAC countries are
included in Table 9.7.
Complying with ICAO standards is not easy, and LAC countries have
made significant progress in meeting these standards. As of 2016, the US
Federal Aviation Administration (FAA) has recorded only two countries,
Barbados and Uruguay, in the LAC region as Category 2 for not comply-
ing with the ICAO Standards. On the issue of adoption of ICAO stand-
ards, Button et al. (2004) have performed statistical analysis to examine
what factors influence whether countries comply, as approximated by the
FAA’s International Aviation Safety Assessment’s rating, with ICAO safety
obligations. The explanatory variables that more largely contribute to a

M4371-FINGER_9781786431851_t.indd 168 14/11/2017 11:14

Table 9.7 FAA Flight Standards Service, International Aviation Safety

Assessment programme
Country Category Country Category

Argentina 1 Nicaragua 1
Bahamas 1 Organization of Eastern 1
Barbados 2* Caribbean States – Eastern
Belgium 1 Caribbean Civil Aviation
Bolivia 1 Authority members:
Brazil 1 Antigua and Barbuda,
Chile 1 Dominica, Grenada, St
China 1 Lucia, St Vincent and The
Colombia 1 Grenadines, St Kitts and
Costa Rica 1 Nevis 1
Dominican Republic 1 Panama 1
Ecuador 1 Peru 1
El Salvador 1 Suriname 1
Guatemala 1 Trinidad and Tobago 2*
Jamaica 1 Uruguay 1
Mexico 1 Venezuela
Notes:
* Countries not serving the US at the time of the assessment.
As of 8 March 2013, countries are removed from the list after four years if they do not
provide air transport service to the US, have no code-share arrangements with US air
carriers, and have no significant interaction with the FAA.
Source: FAA, 15 August 2016.
country being rated 1 include its ‘economic performance’ measured as its

GDP, imports and exports; ‘aviation’-related variants such as airports, reg-
istered aircraft, international airports, airline passengers; and the existence
of an OSA with the US. With respect to LAC airlines flying to the EU,
the region’s carriers have fared well. The EU has been particularly fond
of using a blacklisting approach to address safety issues following airline
accidents. As of December 2016, only Suriname’s Blue Wing Airlines was
on the blacklist.
Regarding air cargo security, some of the most important multilateral
instruments addressing aviation security, hijacking and terrorism have
been adopted under the ICAO’s framework. Part of this framework is
the so-called ‘Standards and Recommended Practices’ (Annex 17 to the
Chicago Convention), which includes considerations designed to prevent
explosives and incendiary devices from being placed on board aircraft.

M4371-FINGER_9781786431851_t.indd 169 14/11/2017 11:14

Following the events of 11 September 2001, air cargo security has

become one of the major global security concerns in LAC countries given
the vulnerabilities that make air cargo a possible target for terrorists. In
the LAC countries, the major concerns currently surround full compliance
with safety and security standards, which are very costly, considering that
most of the region’s air cargo is made of low-value-to-weight perishables.
Usual limitations in the LAC region have persisted in what has to do with
radar coverage, route coverage, instrumental landing services and air
traffic control services (Domènech and Montalvo, 2006), issues required
by ICAO and the US Federal Aviation Administration (FAA).
CASE STUDY: MEXICO
Mexico has made sizeable investments in airport infrastructure and

has used long-term concession models as an approach to privatize the
operation of this infrastructure. However, it is still a highly-regulated
market: entry of airlines and routes require considerable authority over-
sight – for example, with the Secretaría de Comunicaciones y Transporte
SCT (Mexican Ministry of Transportation and Communications) as the
responsible agency. In consequence, airfares have remained high despite
the entry of LCCs. Mexico first attempted to liberalize its aviation market
in 1988 with the signing of a bilateral ASA with the US.
Market Regulation: Open Skies Agreements
Mexico has not been active at negotiating OSAs. For example, although
the traffic between Mexico and the US is the most important traffic for
Mexico, Mexico has used a restrictive policy towards opening its air market
to competition from US carriers. In 2015, both countries concluded nego-
tiations of a lightweight OSA that entered force in August 2016. The main
development is the removal of the limitation of two carriers per route, with
the aim of increasing competition. The previous bilateral ASA between
the countries dated from 1960: it took these neighbours roughly 55 years
to reach some level of liberalization in their joint air transport market.
Although some features of OSAs are part of the deal, overall the agree-
ment is not an OSA.
The agreement will allow scheduled airline flights between the two coun-
tries to travel from any point in one country, via an intermediate country,
to any point in the other country and beyond. However, cabotage will
still not be allowed. There will be no restrictions to schedule cargo flights
for any number of airlines in any city pairs. Airlines from both countries

M4371-FINGER_9781786431851_t.indd 170 14/11/2017 11:14

will be able to freely establish airfares. This new agreement includes an

obligation for each country to grant the airlines of the other country most-
favoured-nation treatment in terms of airport user fees; that is, airport user
fees shall not exceed amounts necessary to cover the full cost of facilities
and services at the airport plus a reasonable return on assets, after depre-
ciation (Fernández-Briseño and López Scherer, 2016). Finally, limits on
the number of airlines that can operate flights between the two countries
have been lifted.
Market Regulation: Airlines
From 1920 to 1988, the Mexican federal government imposed strict price
controls and entry barriers to Mexico’s airlines market. The consequence
of that was the persistence of legal government-owned monopolies for
decades. Valdés and Ramírez (2011) have illustrated how the liberalization
process that started in 1988 – aimed at encouraging regional airlines to
serve as feeders of passenger traffic to national route carriers – never took
off.
As of today, Mexico’s two main airlines, Aeroméxico and Mexicana,
have been privatized. The story of privatization goes back to 1988, when
the Government of Mexico publicly announced that Aeroméxico was
bankrupt. The company Aerovías de México subsequently bought it
out. The same year, Mexicana entered into a public–private model of
ownership in which the Mexican government owned about 40 per cent
of the capital. By 1989, the Mexican government had authorized private
capital investments in Aeroméxico and Mexicana de Aviación. In 1993,
Aeroméxico acquired 55 per cent of the shares of Mexicana. According
to Vargas-Hernandez and Noruzi (2010), at the time the merged company
represented two-thirds of the Mexican air travel market. In 1995, the
Mexican government released its stake in Mexicana. It has been argued
that the privatization process of the two Mexican legacy carriers was
simply a change in the ownership of a monopoly, an action which provided
only limited opportunities to introduce competition and enhance the com-
petitiveness of the broader Mexican economy. The introduction of LCCs
took place in 1991, partly triggering the financial debacle of Mexicana and
Aeroméxico which ended up in these two airlines being statized once again
in 1994–95 (Valdés, 2013).
In July 1991, new air carriers were authorized to serve national routes,
and control over tariffs was eliminated in routes with more than one com-
petitor. The 1995 Civil Aviation Law formalized the liberalization process.
Valdés and Ramírez (2011) have assessed the impact of deregulation on
the performance of the Mexican airline industry using long sets of data

M4371-FINGER_9781786431851_t.indd 171 14/11/2017 11:14

available from 1980 to 2007. Specifically, the authors sought to isolate the
effect of different variables from the effect of deregulation on the number
of passengers and airfares using a series of econometric models. They
found that the effect of deregulation was only marginal and smaller than
statistically more significant variables such as increases in income per
capita. Ros (2011) has also suggested that annual growth in domestic pas-
senger traffic between 1989 and 2008, at a rate of 5.4 per cent, coincided
with a period when the Mexican economy was growing at approximately
3 per cent per year. Regarding the effect of deregulation on the price of
airfares, Ros (2011) did not find evidence of increases or decreases attrib-
utable to deregulation between 1996 and 2004. While there were periods
when competition seemed to lower the cost of flying, during other times
airfares remained high despite additional players entering the market; for
instance, LCCs. Regarding the effect on airfares, existing research has
found that between 1996 and 2004 there were periods with greater contest-
ability in airfares, but in other periods the opposite occurred. Thus, the
effects of deregulation on the Mexican airline industry have been mixed at
best. There is no evidence that the growth in the number of passengers can
be directly attributed to deregulation only. There is also no evidence of a
decrease in the cost of airfares.
Today, the Mexican government continues to set limits when it comes
to approving new routes or establishing limits on foreign ownership of
airlines. Nevertheless, there have been welcome market liberalization dis-
ruptions. For example, Aeroméxico is the last legacy airline in the Mexican
aviation market, as low-cost carriers have come to dominate the market.
Overall, it took Mexico’s institutions nearly 30 years to unenthusiastically
approach liberalization strategies.
Economic Regulation: Airports and Slot Allocation
Mexico’s Airports Law of 1995 allowed the possibility of granting airport

management to private concessionaires. By 1998, 35 of 58 airports had
been concessioned for 50 years with the aim of privatizing the provision
of airport services and infrastructure investments. Mexico concessioned
its airports in groups. Thirty-four small, medium-size airports and a large
airport were mixed into three groups: Center-North (OMA) (13); Pacific
(GAP) (12); and Southeast (ASUR) (9). The Mexico City International
Airport (AICM) remained under public control. The Mexican Secretariat
of Communications and Transportation established a price cap of aero-
nautical revenues.
The privatization process was two-staged. First, the control of the con-
cessionary entities and 15 per cent of their shares were sold to a strategic

M4371-FINGER_9781786431851_t.indd 172 14/11/2017 11:14

Table 9.8 Investor-owned Mexican airports, 2014
Airport Company Global Main 2014 Privatization

Rank Airport(s) Revenue Status
($ Million)
ASUR Aeropuertos del Sureste 59 Cancun 440 Full
GAP Grupo Aeroportuario del 64 Guadalajara, 415 Full
Pacífico Tijuana
Operadora Mexicana de 86 Acapulco, 256 Full
Aeropuertos Monterrey
Source: Adapted from Poole (2016).
partner. The strategic partner had to include a Mexican partner owner

with at least 25.5 per cent of the capital and an operator in charge of the
airport’s aeronautical operations. Second, under the 1995 Airport Law,
foreign investment in airport companies was capped to 49 per cent (ICAO,
2013b) with the caveat that the National Foreign Investments Commission
could grant a waiver for this cap.
Mexican airports, excluding the AICM, are now owned and operated
by the private sector (Poole, 2016). Mexico’s airports rank in the list
of the top 100 largest investor-owned airport companies, as shown in
Table 9.8.
For the assignation of slots, the law had established that the first criteria
would be to give preference to incumbent airlines based on their prior use;
that is, once an airline received a slot the authority could not take it away
unless it was not being used or due to delays (Valdés, 2013). In the case of
the AICM, there is evidence that this node of the system operates beyond
slot capacity while the rest of the system is underutilized. In consequence,
the AICM charges premium slot fees while other airports charge high fees
to be able to cover operational costs (Valdés, 2012). Slot allocation at the
AICM is reportedly a ‘major headache’ to be inherited by the new Mexican
competition authority, the Federal Commission of Economic Competition
(COFECE) in 2017.
The SCT is responsible for the rules regulating all airport services.
Concessions follow the price cap tariff scheme, and allow for inflation
adjustments, with a five-year review. The scheme is dual-till and non-
aeronautical activities are not subject to economic regulation (ICAO,
2013b). The SCT establishes the price caps (dual till); that is, the annual
average revenue per airport, infrastructure investment plans and desired
x-efficiency.

M4371-FINGER_9781786431851_t.indd 173 14/11/2017 11:14

CASE STUDY: COLOMBIA
Colombia started to liberalize its air transport market in 1991 under

a public policy known as ‘freedom under surveillance’. Through Law
80 for Public Procurement and Law 105 of National Transport from
1993, Colombia cemented the basis for exempting the public sector
from being the sole financier of air transport infrastructure. These laws
were complemented by the National Planning Department’s ‘Institutional
Rearrangement and Plan for the Expansion of the Airport System’ of
1994. The plan extended the authority for the government to go forward
with the concession of airports (Díaz Olariaga and López Rodriguez,
2016). The government transferred the direct management of the termi-
nals to the private sector, as well as responsibilities over runways, ramps,
visual aids and all other facilities with the additional requirement of
making investments for reconditioning and maintaining the facilities as
needed. The main obligation for a concession to be granted was that the
private concessionaire would return to the government coffers resources
earmarked for: (1) maintaining a fund to subsidize non-profitable airports;
(2) financing new investments in those airports; and (3) financing air
traffic control services and security. Aerocivil, the Colombian Ministry of
Transportation’s Special Administrative Unit of Civil Aviation, is respon-
sible for regulating, administering, surveilling and controlling the use of
the Colombian air space. However, its functions also include responsibili-
ties over fixing, gathering and collecting all charges, fees and dues for the
provision of both aeronautical and airport services, as well as the revenue
generated by concessions, authorizations, licences or any other type of
equity income.
Market Regulation: Open Skies
Colombia has reciprocity agreements with 13 EU countries, 10 coun-

tries in Central America and the West Indies, 10 countries in South
America, four countries in the Middle East, three countries in Asia,
two countries in North America and one country in Oceania (Martínez
and García, 2016). These agreements establish the frequencies of flights
between countries, the type of aircraft that parties can operate and the
commercial activities they can carry out, among others. Liberalization
through Open Skies agreements started in 1991, when Colombia
sought to open its freight market. Starting in 2009, Colombia started
an aggressive campaign to negotiate OSAs with those countries with
which connectivity was strong. Colombia and the US signed their OSA
in 2011. It should be noted though that Colombia has not achieved full

M4371-FINGER_9781786431851_t.indd 174 14/11/2017 11:14

liberalization (all freedoms of the air/cabotage rights) with any partner

country.
Full liberalization of operations for air carriers has still to occur in

Colombia. Díaz Olariaga (2016) defined the progress made in the last 25
years as one that has gone from a protectionist scheme to one of ‘freedom
under surveillance’. Until 2007, the government established the minimum
fare for domestic and international routes. Since then, there have been
isolated efforts to resume the full opening of the market. One of those
has had to do with allowing the entrance of low-cost-carriers (LCCs):
EasyFly started operations in 2007, and VivaColombia started in 2012. At
the beginning of 2017, both airlines are still operating. As limitations on
foreign ownership of airlines have been eliminated, Ryanair is expected to
take full control of VivaColombia. Nevertheless, there are two restrictions
to full liberalization still in place: (1) the maximum number of carriers in
passenger routes is established by the authority and (2) there is a cap on
airfare prices (Díaz Olariaga and Álvarez, 2015).
Economic Regulation: Airports
When Colombia opened itself up to investments in new infrastruc-

ture, using private participation as in other Latin American countries,
the challenge was how to design and implement effective and efficient
regulation (Serebrisky et al., 2011). These challenges have rapidly been
overcome in recent years. Thus, according to the Colombian Ministry of
Transportation, Colombia currently has 590 airports and airfields, of which
only 74 are owned by Aerocivil, 14 by the Departmental Governments, 94
by Municipalities, nine by the Military, 185 by fumigation companies and
214 are private (Ministerio de Transporte, 2016). Another success story is
that of Colombia’s largest airport, Bogotá El Dorado International, which
was concessioned to a consortium, Opain, in August 2006, and in 2016
handled 36 million passengers, about 50 per cent of the country’s traffic.
From an institutional perspective, Díaz Olariaga (2016) has reported
that Colombia’s approach to managing airports has been modelled on the
regional tendency of using the concession model, adding up to 18 conces-
sioned airports at the beginning of 2015. Colombia started using the con-
cession figure in 1995. Moreover, between 1995 and 2015 Díaz and others
have identified an evolution in the scope of the concession agreements (see
also Martínez and García, 2016).
The first concession agreement sought mainly to guarantee the con-

M4371-FINGER_9781786431851_t.indd 175 14/11/2017 11:14

cessionaire a minimum level of revenue, a scheme under which the con-

cessionaire assumed no risk. The conditions of this type of concession
included three major elements: operation, maintenance and transferring.
From 2000 to 2007, Aerocivil started demanding a share of fixed charges
from the concessionaire, supplemented by variable charges depending on
the total revenue collected. The conditions of the concession subsequently
changed to include construction, in addition to operation, maintenance
and transferring, as a push to improve existing airport infrastructure. The
change was motivated by insufficient resources available for Aerocivil to
continue expanding investments at other airports. In 2010, the combined
model of fixed and variable charges was replaced with a single charge as a
percentage of the concessionaire’s revenue.
In contrast to the successes in the deregulation of the Colombian air-
lines market, Aerocivil today continues to tightly regulate the prices of
aeronautical services for the airports of its property, the municipal ones
and for the concessionaries. These fees include landing or aerodrome fees,
flight protection services, navigation services, national and international
airport charges, fees for the use of boarding bridges and all other airport
services.
CASE STUDY: CHILE
Chile’s liberalization of the air transport sector started in 1979 with the
enactment of the Commercial Aviation Law, DL No. 2564. Chile has been
known as a pioneer in the deregulation of the aviation sector using an
approach known as ‘Open Skies with reciprocity’. There might be several
reasons for Chile being a pioneer, including its geography and political
history, but assessing those would be outside the scope of this chapter.
Since the focus here is regulation, it is instead more relevant to mention
the four major principles that guided the development of the Aviation Law
and which remain relevant today. These principles can be summarized as
follows: (1) expand OSAs with as many partners seeking free entry to the
market as possible; (2) grant freedom to set fares for air carriers; (3) mini-
mize intervention of the authority; and (4) liberalize ownership require-
ments and control over the air carriers.
Market Regulation: Open Skies
In the last ten years, Chile has pursued an ‘Open Skies with reciprocity’
policy that has meant that Chilean authorities have granted foreign car-
riers the same freedoms of the air that foreign countries have granted to

M4371-FINGER_9781786431851_t.indd 176 14/11/2017 11:14

Chilean carriers. Since 2011, with a new policy further liberalizing the
sector, cabotage rights were granted unilaterally to foreign airlines wishing
to operate domestic flight segments. According to Chile’s Civil Aviation
Board the objective of this latest wave of liberalization is to create more
competitive conditions through cabotage, facilitating the access of foreign
operators to the internal transport of passengers and cargo – that is, ‘that
all the companies of the world can enter the domestic Chilean market
without restrictions’. As expected, foreign airlines must still meet all the
established technical and insurance requirements as domestic carriers.
Villena et al. (2008) attempted to quantify the effects of market liberali-
zation through OSAs (Table 9.9). There seems to be evidence of a positive
impact on the flow of passengers and a tendency to market concentration
after a treaty is signed. Connectivity seems not to be affected. Regarding
concentration in the industry, there is a tendency for the Herfindahl–
Hirschman Index to increase following the finalization of an OSA.
Available academic and industry studies support the claim that Chilean
authorities have sought to give air carriers – both domestic and international
– freedom to set service frequencies, fares and capacity without prior
approval of the authority, except for technical aspects and insurance provi-
sions (Alvarado and Marcos, 2012). There are no legal restrictions with
respect to establishing an airline, its ownership or its management.
Economic Regulation: Airports
Chile’s Ministry of Public Works is primarily responsible for the design,

award and monitoring of airport concessions. The Ministry of Finance
needs to approve the concession. All concession bids are expected to
comply with Chile’s 1996 Infrastructure Concession Law and its regula-
tions. These legal documents govern the minimum rates, higher payments
for existing infrastructure, lower subsidies or a combination of technical
and economic considerations. Minimum income guarantee is possible in
airport contracts. Under this model, ‘if a concessionary bidder decides to
request a minimum income guarantee, it will have to share extra revenues
with the government if the collected revenues are higher than the thresh-
old established in the concession contract’. If the bidder decides not to
request this, it will have ‘to assume the whole traffic risk’ (ICAO, 2013a).
Reportedly, most bidders prefer the revenue guarantee model as it reduces
the financial cost and the risk of the projects. Air carriers are allowed a
maximum capital participation of 20 per cent in airport concessionaires.

M4371-FINGER_9781786431851_t.indd 177 14/11/2017 11:14

Table 9.9 Economic impact of Chile’s Open Skies policy
Partner Years Observed Passengers Herfindahl– Econometric Economic Connectivity

Country Effective Hirschman Index Analysis, Open Benefit,
Skies Effect 2001–06,
M4371-FINGER_9781786431851_t.indd 178
$million
Argentina 1990 Steady increase in Increase from 0.14 34 per cent additional 300.9 4 pairs in 1991 to
passengers 1990–99; in 1990 to 0.57 in passengers in 1991– 14 in 1995, 10
greater rate since 2003 2001 2006; 221 371 more in 2004–06
annual passengers
Brazil 1989, Large increase since 0.3 in 1989, 0.45 in 6 per cent extra 65.6 3 pairs in 1994, 8
1993, 2002 1992, 0.35 in 1995, passengers in 1998– in 2000, 3 in 2003,
1996 0.2 in 2006 2006, 21 038 more 7 in 2006
178
annual passengers
Spain 1991, Large increase since 0.3 in 1990, 0.5 in 54 per cent additional 110.6 1–3 pairs, no
1997, 1997 1995, 0.5 in 2004, passengers in 1997– apparent influence
2005 0.35 in 2006 2006, 79 662 more of the OSA
annual passengers
US 1999 Large decrease following 0.3 in 1999, 0.35 in Not estimated Not 5 pairs in 1992, 9
2001, recovery since 2006 estimated in 1996
2004
Peru 1997 Constant growth until 0.27 in 1997, 0.7 in 6 per cent additional 12.3 1 pair in 1995, 6 in
2000, recovery starting 2000, 0.5 in 2005, passengers in 1998– 1997, 1 after 1999
in 2004 0.7 in 2006 2006, 12 904 more when cabotage is
annual passengers enforced

Source: Based on Villena et al. (2008).
14/11/2017 11:14
Before 1996, Chile’s 1991 Infrastructure Concessions Law provided the

legal basis to attract private investments in airport projects. It aimed at
bringing in private sector experience into government-financed projects
at the same time it made it possible for users of the infrastructure to help
finance projects charging user fees. As an example, Paredes and Sánchez
(2004) have offered a brief history of how the Santiago International
Airport was concessioned with the Ministry of Public Works, the
General Directorate of Civil Aviation (DGAC) and the Ministry of
Finance participation in the process, with the DGAC taking the lead.
The concession contract was designed to last 15 years, increasing pas-
senger capacity by a factor of three and freight capacity by a factor of
1.5, with an investment of $100 million. The concessionaire was asked
to provide a series of services, subject to price regulation, including:
embarkation and disembarkation systems, airside platform service areas
and catering areas. Non-aeronautical services were also part of the con-
tract, and included: parking services, food, duty-free shops and public
transport. Domestic passengers were charged an airport tax of $8 while
international travellers paid $18. The DGAC received all the tax revenue
in addition to $2.5 million annually for the first five years and $2 million
annually thereafter.
Despite all the requirements included in the bidding process, the con-
cession of Santiago’s airport ranked poorly in terms of economic regula-
tion. For example, Paredes and Sánchez (2004) found that the winner was
selected based on non-economic variables, which are difficult to measure
objectively. This turned the bidding process into a ‘beauty-contest-type
allocation’, in which economic variables, such as time to make the project
profitable, were absent. Ultimately, Paredes and Sánchez argue that it
might have been against the self-interest of the DGAC to award the con-
cession based on economic variables, since doing so would have reduced
the DGAC’s rents by shortening the project’s lifespan. Therefore, the con-
cession rules ended up largely benefiting the DGAC, as they were designed
to make this entity’s budget sustainable over time, possibly at the cost of
consumers.
It remains to be seen if these issues were dealt with in 2014, when Chile’s
Ministry of Public Works concluded the concession of a $700 million
project adding a new international terminal and expanding the domestic
terminal at Santiago de Chile International airport. Poole (2016) has
reported that the triumphant bidding team, a consortium consisting of
Aéroports de Paris (45 per cent), Vinci (40 per cent) and Astaldi (15 per
cent), won the 20-year concession by offering the government 77.5 per cent
of the airport revenue.
The Chilean concession system follows the model of self-financing;

M4371-FINGER_9781786431851_t.indd 179 14/11/2017 11:14

that is, the system is financed through the collection of fees paid by air-
lines for use of the services provided by the DGAC, by passengers using
the network of air terminals and by aeronautical and non-aeronautical
concessions. The Aviation Law requires that the income obtained by the
collection of these rates be reinvested in the system itself (Agostini, 2008).
Through time, the Chilean concession model has been adjusted using a
case-by-case strategy.
CONCLUSIONS
The survey of the academic and industry literature on recent developments

in the aviation sector of LAC shows little conclusive evidence regarding the
economic impact of past liberalizations. To better understand aviation policy
in the region, policy-makers would benefit from additional analyses of avia-
tion markets under different future liberalization scenarios. This is an area
with great academic potential for aviation economists with an interest in the
region. To achieve that, a positive development would be a coordinated effort
by the countries of the LAC region, multilateral and regional organizations,
as well as industry to gather and disseminate data on the aviation sector.
LAC countries have made significant progress in improving the per-
formance of their airports’ operations through concessions, in providing
air navigation services and complying with ICAO safety standards. Some
LAC countries, such as Chile and Colombia, have been successful in
improving the efficiency of their airline industry through consolidation
and infusion of foreign investment. These developments have been made
possible through the elimination of entry barriers that have allowed for
cross-border consolidation.
In parallel, economic forces, globalization and changes in consumer
preferences have continued to shape aviation markets in the region, espe-
cially freight markets. For example, there is new, strong demand in the US,
the EU, and increasingly Asian countries for the wider variety of P and
E produced in LAC countries. Chile and Colombia have recognized the
value of air connectivity as an enabler of the success of these new industry
sectors that include flowers, fish, produce and exotic fruits. Both countries’
success seems, indeed, to be contributing significantly to other countries in
the region embracing liberalization more decisively.
Chile and Colombia may be exceptions in the region when it comes
to favouring liberalization of air transport markets. Most country regu-
lators have traditionally been more comfortable with approaches such
as Mexico’s lengthy and protectionist process to liberalize air transport
markets. This is unfortunate, since air transport does not require excessive

M4371-FINGER_9781786431851_t.indd 180 14/11/2017 11:14

investments in inflexible, fixed infrastructure, as is the case with roads or

rail. Long-term investments of less than 1 per cent of all expenses in infra-
structure are all what would be required for LAC countries to keep up with
future investment needs in air transport.
NOTES
1. Oum et al. (2006) provided evidence that airports with government majority ownership
and those with multi-levels of government involvement are significantly less efficient than
private airports.
2. In 1998 the Mexican government held 85 per cent of the shares, with a private cor-
poration holding the remaining 15 per cent. As of February 2006, the 85 per cent of
government shares had been auctioned in the domestic and international stock markets,
completing the privatization process (Rico Galeana, 2008). Overall, Mexican airports
have experienced steady growth of traffic. Smaller airports continue to be subsidized by
central and local governments.
3. A major accomplishment in Central America, although not an aviation market liberaliza-
tion, has been the structuring of the Central American Corporation for Air Navigation
Services, a regional consortium that provides air traffic services for its members.
4. This provision allows a domestic carrier to operate all-cargo services between the foreign
country and a third country, via flights that are not linked to its homeland.
5. Horizontal agreements are agreements negotiated by the European Commission on
behalf of EU member states, in order to bring all existing bilateral ASAs between EU
member states and a given third country in line with EU law, removing nationality restric-
tions of existing bilateral ASAs and therefore allowing European airlines to operate
flights between any EU member state and a third country.
6. According to the EU’s website, the so-called comprehensive aviation agreements that
the European Commission negotiates on behalf of the EU and its member states
are not limited to ‘Open Skies’ provisions only, but instead ‘the EU model also seeks
the establishment of a process of liberalization of ownership of airlines and a process
of regulatory convergence in matters of safety and security, competition, environment,
passengers protection, labor, etc. – which could not be obtained at national levels’.
7. The main distinction between failures in safety and security is the issue of intentional-
ity. While safety failures have unintentional causes, security failures are intentional in
nature.
REFERENCES
Agostini, C.A. (2008), ‘The industrial organization of air transport in Chile (La
organización industrial del transporte aéreo en Chile)’, Revista de Análisis
Económico, 23, 35–84.
ALTA (2016), The Latin American and Caribbean Capacity Analysis. Available at
https://www.alta.aero/la/upload/report/_9945.pdf.
Alvarado, A. and G. Marcos (2012), Evaluación del Impacto Económico de un
Acuerdo de Cielos Abiertos Chile-UE, Informe Final, Junta de Aeronautica Civil,
Santiago de Chile.
Button, K. and H. Vega (2008), ‘The effects of air transportation on the movement
of labor’, GeoJournal, 71, 67–81.

M4371-FINGER_9781786431851_t.indd 181 14/11/2017 11:14

Button, K.J., A. Clarke, G. Palubinskas, R. Stough and M. Thibault (2004),

‘Conforming with ICAO safety oversight standards’, Journal of Air Transport
Díaz Olariaga, O. (2016), ‘Análisis del desarrollo reciente del transporte aéreo en
Colombia’, Revista Transporte y Territorio, no. 14, 122–43.
Díaz Olariaga, O. and J.Á. Álvarez (2015), ‘Evolution of the airport and air trans-
port industry in Colombia and its impact on the economy’, Journal of Airline and
Airport Management, 5, 39–66.
Díaz Olariaga, O. and A.L. López Rodríguez (2016), ‘Comportamiento e inter-
relación del turismo y el transporte aéreo en Colombia’, Estudios y Perspectivas
en Turismo, 25, 36–56.
Domènech, A.B. and J.G. Montalvo (2006), ‘Free and non-discriminatory
access to airports: a proposal for Latin America’, International Journal of
Transport Economics/Rivista Internazionale di Economia dei Trasporti, 33,
211–55.
European Union (2017), ‘International aviation: Brazil’, accessed 20 February
2017 at http://ec.europa.eu/transport/modes/air/international_aviation/country_
index/brazil_et.
Fernández-Briseño, R. and S. López Scherer (2016), New Air Transport Agreement
Between the US and Mexico 2016, Mexico: White and Case LLP.
Forsyth, P. (2007), ‘The impacts of emerging aviation trends on airport infrastruc-
ture’, Journal of Air Transport Management, 13, 45–52.
Gómez-Lobo, A. and A. González (2008), ‘The use of airport charges for funding
general expenditures: the case of Chile’, Journal of Air Transport Management,
14, 308–14.
Hanlon, P. (2007), Global Airlines: Competition in a Transnational Industry,
Amsterdam: Elsevier.
ICAO (2013a), ‘Case study on commercialization, privatization and economic
oversight of airports and air navigation services providers: Chile’. Available at
http://www.icao.int/sustainability/CaseStudies/Chile.pdf.
ICAO (2013b), ‘Case study on commercialization, privatization and economic
oversight of airports and air navigation services providers: Mexico’. Available at
http://www.icao.int/sustainability/CaseStudies/Mexico.pdf.
Kohli, H.A. and P. Basil (2011), ‘Requirements of infrastructure investment in
Latin America under alternate growth scenarios 2011–2040’, Global Journal of
Emerging Market Economics, 3, 59–110.
Lin, W.C. (2012), ‘Financial performance and customer service: an examination
using activity-based costing of 38 international airlines’, Journal of Air Transport
Lipovich, G.A. (2008), ‘The privatization of Argentine airports’, Journal of Air
Martínez, A. and García, H. (2016), Competitividad en el transporte aéreo en
Colombia. Informe Final, June, Bogotá: Fedesarrollo, Fontur and MinCIT.
Martínez-Zarzoso, I. and F.D. Nowak-Lehmann (2004), ‘Economic and geograph-
ical distance: explaining Mercosur sectoral exports to the EU’, Open Economies
Review, 15, 291–314.
Ministerio de Transporte (2016), Information accessed on 30 December 2016 at
https://www.mintransporte.gov.co/loader.php?lServicio=FAQandlFuncion=vie
wPreguntas.
Moore, A., S. Straub and J.J. Dethier (2014), ‘Regulation, renegotiation and

M4371-FINGER_9781786431851_t.indd 182 14/11/2017 11:14

capital structure: theory and evidence from Latin American transport conces-
sions’, Journal of Regulatory Economics, 45, 209–32.
Oum, T.H., N. Adler and C. Yu (2006), ‘Privatization, corporatization, owner-
ship forms and their effects on the performance of the world’s major airports’,
Paredes, R.D. and J.M. Sánchez (2004), ‘Government concession contracts in
Chile: the role of competition in the bidding process’, Economic Development
and Cultural Change, 53, 215–34.
Perelman, S. and T. Serebrisky (2010), Measuring the Technical Efficiency of Airports
in Latin America, Policy Research Working Paper No. 5339, Washington, DC:
World Bank.
Perelman, S. and T. Serebrisky (2012), ‘Measuring the technical efficiency of air-
ports in Latin America’, Utilities Policy, 22, 1–7.
Poole Jr, R.W. (2016), Annual Privatization Report 2016: Air Transportation,
Washington, DC, Reason Foundation.
Rico Galeana, O.A. (2008), ‘The privatisation of Mexican airports’, Journal of Air
Ros, A.J. (2011), ‘The determinants of pricing in the Mexican domestic airline
sector: the impact of competition and airport congestion’, Review of Industrial
Organization, 38, 43–60.
Serebrisky, T., S.L. Azumendi and L.A. Andrés (2011), ‘Institutional design and
governance of airport regulators: the case of Latin America’, Journal of Air
Sirtaine, S., M.E. Pinglo, J.L. Guasch and V. Foster (2005), How Profitable are
Infrastructure Concessions in Latin America? Empirical Evidence and Regulatory
Implications, Washington, DC: World Bank.
US Department of Commerce (2006), Brazil, Washington, DC: Western
Hemisphere Diversification and Defense Market Guide. Accessed on 22 July
2006 at http://www.bis.doc.gov/DefenseIndustrialBasePrograms/OSIES/Export
MarketGuides.
Valdés, V. (2012), Recomendaciones de Política Aeronáutica para México,
USAID.
Valdés, V. (2013), ‘Fallas del mercado y del gobierno en el sector aeronáutico
Mexicano’, Revista de Economía Institucional, 15, 253–83.
Valdés, V. and J.C. Ramírez (2011), ‘Una evaluación sobre la desregulación
del mercado de aerolíneas en México’, Economía Mexicana. Nueva Época, 20,
5–35.
Vargas-Hernandez, G.J. and M.R. Noruzi (2010), ‘From entrepreneurial state to
state of entrepreneurs: ownership implications of the transformation in Mexican
governance since 1982’, Journal of Public Administration and Policy Research,
2(1), 7–19.
Vega, H. (2008), ‘Air cargo, trade and transportation costs of perishables
and exotics from South America’, Journal of Air Transport Management, 14,
324–8.
Vega, H.L. (2010), Developing Countries and their Airborne Export Flows of
Perishable and High-tech Goods, Fairfax: George Mason University.
Villena, M.J., R. Harrison and M.G. Villena (2008), ‘Impacto Económico de la
Política de Acuerdos de Cielos Abiertos en Chile’, Revista de Análisis Económico,
23, 107–49.
Wanke, P. and C.P. Barros (2016), ‘Efficiency in Latin American airlines: a

M4371-FINGER_9781786431851_t.indd 183 14/11/2017 11:14

two-stage approach combining virtual frontier dynamic DEA and simplex

regression’, Journal of Air Transport Management, 54, 93–103.
World Economic Forum (2016), A New Regulatory Model for Foreign Investment
in Airlines, Geneva.

M4371-FINGER_9781786431851_t.indd 184 14/11/2017 11:14

10. Air transport in Africa
Gianmaria Martini and Davide Scotti
INTRODUCTION
The bodies that are generally viewed as the most authoritative sources
of aviation forecasts – Boeing, Airbus and the International Civil
Aviation Organization – consider Africa as having the potential for a
rapid increase in its air transport sector.1 This sector is characterized by
generally favourable conditions for a significant expansion of its aviation
industry, partly due to the improved political and economic climates in
many African countries, particularly those in the sub-Saharan region.
The situation in North Africa is also somewhat more stable than it was a
few years ago. Adjusted estimates by the International Monetary Fund
of the national incomes of countries like Nigeria suggest faster economic
growth than previously assumed, while peaceful, democratic transitions
of power in countries like Gambia and Ghana indicate greater political
stability.
Africa is also a continent with large areas of open space and river net-
works that, unlike those of Europe, are often not conducive to trade and
are difficult terrain for constructing surface transportation. It also has a
number of large urban areas that are remote from each other and from
non-African cities, making movement between them difficult. Given these
geographical and demographical characteristics, and the significant growth
of many other continental economies seeking the resources of Africa, air
transport appears to be the ideal catalyst for the economic development of
the continent.
Partly because of its strategic importance in the colonial and immediate
post-colonial periods, and partly because the necessary private institutional
structure to finance and operate an air network did not exist, African air
transportation has traditionally been heavily regulated. However, fol-
lowing the experiences of aviation policies elsewhere, particularly in the
US and Europe, coupled with the manifest failure of many state-owned
airlines in Africa, this attitude changed and market liberalization became
seen as a way forward. This led to the beginning of Africa’s air transport
185

M4371-FINGER_9781786431851_t.indd 185 14/11/2017 11:14

deregulation process, particularly the 1988 Yamoussoukro Declaration

and the subsequent 1999 Yamoussoukro Decision.
By this time, the focus had shifted to a gradual elimination of restrictive
bilateral traffic rights that had impeded the development of international
air transport, both within the continent and with outside countries. The
new measures were seen as facilitating the expansion of the airline industry
in a largely market-oriented environment. However, some 30 years later,
political interference, regulatory constraints and poor planning continue
to slow down the growth of the industry, resulting in the African air trans-
port market remaining small in global terms despite the perceived potential
for its development. As highlighted above, however, the perception is that
things are changing and that the objectives of Yamoussoukro may now
begin to be realized.
This chapter analyses the evolution of the African airline industry by
looking at the main policies that have been adopted, trends in the various
airline markets within and outside of Africa between 1997 and 2013 and
changes in the airline industry in the near future as it confronts continuing
institutional challenges.
AFRICAN AIR TRANSPORT LIBERALIZATION
Currently, despite a landmass of 11 730 000 square miles, a population

of 1.02 billion and a population density of 87 persons per square mile,
Africa makes the lowest use of air transportation of any of the major
continents. African air transport represents somewhat less than 2 per cent
of the world’s passengers, and less than 1.5 per cent of the cargo market.
However, these data should be placed in the context of recent growth
trends in air traffic that have, in the case of sub-Saharan Africa, been faster
than most other mega-regions.
Institutionally, the African air transport industry has followed a some-
what different path to that seen in the more established and larger markets
of North America and Europe. By 1945, the African continent had, with
some exceptions, been colonized by a number of the European powers that
had emerged victorious from World War II.2 This significantly influenced
most of the African countries’ economic and political structures, not to
mention their external links; this, in turn, had implications for their trans-
portation policies (Button et al., 2015a).
It has only really been since the 1960s, when many colonies became
independent countries, that African states began to negotiate their own air
services agreements (Schlumberger, 2010) and to operate and regulate their
own airlines. From the early 1960s, many government-owned national air-

M4371-FINGER_9781786431851_t.indd 186 14/11/2017 11:14

Air transport in Africa 187
lines were created as a symbol of country independence, but also, theoreti-

cally at least, as part of national strategic economic planning initiatives.3
There was also a tendency for those countries to buy the larger jets of the
time, such as the DC-10s and Boeing 747s, for symbolic reasons, even if air
demand did not yet warrant such aircraft. In most cases, economics came
second to notions of advertising national pride and independence.
Most African flag carrier airlines adopted business models that focused
on profitable, or at least potentially profitable, international routes to
and from the territories of their former colonial masters. Such a model
enabled them to support costly domestic routes (Guttery, 1998). In this
context, the framework of bilateral agreements (that is, predetermined
market access and service capacities) that had emerged from the Chicago
Convention of 1944 and regulated all intercontinental routes and the intra-
African air transport services provided flag carriers with protection from
competition from both private airlines and the flag carriers of other states.
Consequently, the majority of African carriers had little incentive to be
efficient or to invest in the skills needed to enhance their efficiency in the
long term. Thus, they lost money and tended to be rampantly inefficient
(Schlumberger, 2010). Political interference worsened the situation, with
a tendency towards appointing senior managers as sinecures for political
services, and towards overstaffing as a way of providing employment for
skilled workers for whom there were otherwise limited job opportunities.
This led, on the one hand, to a sort of aversion by African politicians to
liberalization (considered as a threat for the inefficient African airlines and
thus their prestige) and, on the other hand, to a ‘lack of genuine intercon-
nectivity’ at the intra-African level (Button et al., 2015b). However, African
states themselves gradually recognized that new policies were needed
in light of the recent developments throughout the world, such as the
deregulation processes within the United States and (at a later stage) the
European Union (Akpoghomeh, 1999). The costs of maintaining ineffi-
cient carriers became burdensome for many slow-growing economies, and
the failure of the likes of Air Afrique undermined the image of modern,
efficient African airline business (Amankwah-Amoah and Debrah, 2014).
The generally poor safety records of African airlines added to pressures
for change. The prevailing idea became that African carriers had to grow
and merge in order to operate successfully in the markets between Africa,
Europe and the US and that it was necessary to liberalize the access to
intra-African air transport markets.
The first major step toward African aviation liberalization was the 1988
Yamoussoukro Declaration, which mainly aimed to foster airline coopera-
tion and integration, but also incorporated a gradual elimination of traffic
rights (Schlumberger, 2010). The underlying idea was to prepare for the

M4371-FINGER_9781786431851_t.indd 187 14/11/2017 11:14

international cascade effects of deregulation and the open skies policies

embraced elsewhere (Button et al., 2015a).
In 1999, the Yamoussoukro Declaration evolved into the Yamoussoukro
Decision, becoming formally effective and moving primarily toward the
liberalization of the access to intra-African air transport service markets
(Schlumberger, 2010). Indeed, the Decision established the gradual lib-
eralization of scheduled and non-scheduled intra-Africa air transport
services, opening up African skies. In more detail, as pointed out in Njoya
(2016), such an initiative intended to eliminate all non-physical barriers
and restrictions on access, capacity, frequency and fares, provide rights
from the first to fifth freedom for passengers and cargo air transport ser-
vices operated by eligible airlines, guarantee effective and fair competition
among airlines on a non-discriminatory basis, and help the African indus-
try to accomplish international safety standards.
Despite its lofty objectives, the Yamoussoukro Decision has yet to be
effectively implemented by many African civil aviation authorities. In fact,
its application has been influenced by the significant fragmentation of the
continent, with some countries having realized significant changes in their
aviation industry on a multilateral basis, while others have applied the
Decision principles on the basis of bilateral agreements as opposed to the
open skies policy embedded in it. This latter situation has mainly involved
countries that have concerns that the full implementation of the Decision
would lead to the collapse of their inefficient flag airlines as a result of
them being confronted by competitive pressure from bigger African and
non-African airlines (Button et al., 2015b). Thus, most international air
transport services in Africa are still conducted under regulatory restric-
tions on entry, capacity, fares and airline designation.
Despite such slow progress, the Yamoussoukro Decision seems to
represent a step in the right direction towards the growth of the African
civil aviation industry, as demonstrated by the overall increase in traffic
and frequency of flights registered in the last two decades. The intra-
Africa network has experienced some growth in alliances and cooperation
arrangements among established African airlines, although membership
in the global alliances remains limited (Amankwah-Amoah and Debrah,
2011).
EVOLUTION OF THE AFRICAN AIRLINE INDUSTRY
Our analysis of the African airline industry is mainly based on the Official
Airline Guide (OAG) database, from which are collected scheduled operat-
ing direct flights covering 1997 to 2013.4 Button et al. (2015b) point out

M4371-FINGER_9781786431851_t.indd 188 14/11/2017 11:14

Table 10.1 Global commercial airlines net post-tax profits by region

($ billion)
Regions 2012 2013 2014

North America 2.3 7.4 11.2
Europe 0.4 1.0 3.3
Asia-Pacific 2.7 1.9 1.2
Middle East 1.0 0.3 0.7
Latin America −0.2 0.2 0.0
Africa −0.1 −0.1 0.0
Global 6.1 10.6 16.4
Source: International Air Transport Association (2014a; 2015a).
that African airlines revenue passenger-kilometres (RPK) account for only

2 per cent of global RPK. Such a marginal weight is also confirmed in
terms of profits (Table 10.1). While global airline net post-tax profit was
$16.4 billion in 2014, African carriers exhibited a value equal to 0 – one of
the weakest performances of any region worldwide.
Despite marginal importance at the global level, the number of avail-
able seats on flights to, from and within Africa has grown significantly
between 1997 and 2013. African airlines volume increased from 57 million
seats in 1997 to more than 116 million in 2013. Such significant growth is
the result of a route consolidation trend (Njoya, 2016) of African airlines,
with a higher capacity allocated to fewer airport-to-airport routes, which
mainly link large and medium cities to hubs. Non-African airlines exhibit
relevant growth as well (from 24 to 64 million seats) as a confirmation of
the growing interest of foreign carriers, especially European and Middle
East airlines, in such a growing market.
In terms of intra-Africa and intercontinental growth of African air-
lines, available seats provided on routes within Africa registered an 86 per
cent increase from 1997 to 2013, with domestic routes representing the
most relevant component of this internal market, especially thanks to the
developments in South Africa and Nigeria. However, despite the lower
weight in terms of traffic, the routes connecting different African regions
(inter-regional links) exhibit the highest percentage growth, a pattern that
correlates with trends with market liberalization processes. Concerning
the capacity provided by African carriers on intercontinental routes, the
registered increase is from 16.5 million seats in 1997 to 39.5 million in 2013.
Turning to trends within Africa, capacity growth from 1997 to 2013 was
mainly due to the increased number of seats being provided by African

M4371-FINGER_9781786431851_t.indd 189 14/11/2017 11:14

TOTAL AFRICAN AIRLINES NON AFRICAN AIRLINES

250
200
150
100
50
0
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Figure 10.1 Number of airlines operating to, from and within Africa
(1997–2013)
registered airlines, with non-African carriers providing a constant and

relatively low number mainly associated with feeder services for their inter-
continental routes. The growth of available seats on intercontinental con-
nections, however, is mainly due to non-African carriers operating almost
60 per cent of the traffic volume in 2013 as a result of the increasingly high
attention paid by foreign carriers to the potentialities of the African air
transport market.
Figure 10.1 shows the number of airlines operating to, from and within
Africa. As seen, African airlines exhibited a growing trend in their number
during the post-Yamoussoukro Decision period until the global financial
crisis of 2008–09.5
Table 10.2 summarizes the main African airlines in terms of available
seats, flights and ASKs (Available Seat Kilometres). EgyptAir has been
the most important carrier, based on seats provided, since 2010 when it
overtook South African Airways. EgyptAir has experienced 83 per cent
growth since 1997, with most of its traffic directed toward Europe and the
Middle East. The Egyptian carrier is state-owned, but managed indepen-
dently from the government and is exploiting Egypt’s favourable geopo-
litical position as well as its attractiveness in terms of tourism and trade
(O’Connell and Warnock-Smith, 2012).
In recent years, EgyptAir has also joined the Star Alliance and the Arab
Air Carrier Cooperation Agreement, having access to the related advan-
tages. The second-largest African airline, South African Airways (SAA),
is also a Star Alliance member. Despite being second place in terms of
available seats, SAA is still the carrier that operates the highest number of
ASKs (approximately 35 billion). This carrier provides regional and inter-

M4371-FINGER_9781786431851_t.indd 190 14/11/2017 11:14

Table 10.2 Top eight African airlines by available seats (2013)
Carrier Region Seats Total Intra- To/from Change ASKs

(million) Africa Africa from 1997 (billion)
EgyptAir Northern Africa 14.7 4.8 9.8 183% 31.6
South African Southern Africa 14.0 11.7 2.3 131% 35.0
Airways
Royal Air Maroc Northern Africa 9.4 3.6 5.8 1161% 18.9
Ethiopian Airlines Eastern Africa 8.8 5.7 3.1 1319% 26.2
Air Algerie Northern Africa 6.9 2.8 4.2 146% 9.4
Tunis Air Northern Africa 6.3 1.5 4.7 1120% 9.1
Kenya Airways Eastern Africa 6.1 4.8 1.2 1338% 13.7
Arik Air Western/Central 4.8 4.6 0.2 N.A. 4.4
Africa
continental connections from its major African hub, Johannesburg’s OR

Tambo International Airport. The limited growth of the carrier between
1997 and 2013 is largely related to the stronger competition in the domes-
tic market and the unfavourable geographical position of South Africa in
the global air transport network, which limits its intercontinental traffic
growth. Among the Northern African carriers, Royal Air Maroc real-
ized the largest growth in seats. The Moroccan carrier benefited from the
Morocco–EU Open Skies agreement signed in 2006. This carrier focuses
on the more profitable intercontinental connections, facing the competi-
tion of established full-service and low-cost European carriers.6 Eastern
Africa carriers – namely Ethiopian Airlines and Kenya Airways – have
imposed themselves as leading carriers in the African network. Their
growth has been stimulated by their admission into the Star Alliance and
the Sky Team, respectively. These carriers allocate most of their capacity
on intra-African routes departing from/arriving at their hubs of Addis
Ababa and Nairobi. Connections for the Middle East and Asia exhibit a
lower weight.
A significant part of these East African airlines’ passengers from West
Africa connect onwards to intercontinental endpoints located in Asia, also
creating an attractive mass of travellers for non-African airlines such as
the Gulf carriers (O’Connell, 2011). Finally, the first Central/West African
carrier is Nigerian Arik Air, which took over the bankrupted Nigeria
Airways operations and successfully exploited the country’s economic
growth. However, its network exhibits a lower size than those of the top
carriers of the other African regions and is mostly limited to intra-Africa
connections (especially within the Nigerian domestic market). In this
sense, the data support the idea of an unbalanced regional distribution

M4371-FINGER_9781786431851_t.indd 191 14/11/2017 11:14

Table 10.3 Top eight non-African airlines by available seats (2013)
Carrier Region Seats Intra- To/from Change ASKs

(million) Africa Africa from 1997 (billion)
Emirates Middle East 6.3 0.5 5.8 11561% 29.5
Air France Western Europe 6.2 0.7 5.5 186% 24.7
British Airways Western Europe 5.7 3.1 2.6 133% 20.2
Turkish Airlines Eastern Europe 4.0 0.6 3.4 11935% 11.9
Saudi Arabian Middle East 3.1 0 3.1 188% 6.0
Airlines
Ryanair Western Europe 2.9 0 2.9 N.A. 5.1
Lufthansa Western Europe 2.4 0.5 1.9 153% 9.1
Aigle Azur Western Europe 2.3 0 2.3 N.A. 2.8
of African air traffic, as pointed out in Njoya (2016). While Northern,

Eastern and Southern African countries have successfully developed a
robust air transport network, Central and Western Africa have been less
effective.
The large increase in volume registered by non-African carriers is mainly
due to the consistent traffic growth from Europe and the Middle East.
Many airlines created new services directed toward the African continent.
The number of non-African airlines operating in the African network
increased from about 60 in 1997 to 105 in 2013. Such growth may be seen
as showing the increasing relevance of the African market in the global
network. These airlines provide increasing flight frequency on the connec-
tions between Africa and their base countries, although their operation
within Africa still seems to be negligible (with the exception of British
Airways).
Table 10.3 shows the top eight non-African airlines, in terms of seats
provided. As can be seen, the capacities of the European and Middle
East carriers are comparable to those of the major African airlines.
European airline connections are mainly directed toward their former
colonies (for example, Air France’s primary destinations are the Northern
Africa countries). British Airways is the only carrier that competes
strongly with African carriers on intra-African routes, with 3.1 million
seats provided mainly in the domestic South African market as feeding
services for its intercontinental flights.7 Emirates exhibited significant
growth between 1997 and 2013, second only to Turkish Airlines. This
was a period when, due to safety concerns, many African airlines were
prohibited from flying to the European Union, leaving carriers like
Emirates, which serves several African destinations, with the opportuni-
ties to grow their patronage (O’Connell, 2011).8 Added to this was the

M4371-FINGER_9781786431851_t.indd 192 14/11/2017 11:14

Table 10.4 Percentages of routes and seats by market structure in the

intra-African network
1997 2013
Routes Monopoly 76% 68%
Duopoly 13% 16%
More than 2 airlines 11% 16%
Seats Monopoly 37% 28%
Duopoly 17% 18%
More than 2 airlines 48% 54%
role of low-cost carriers in connecting Europe and Northern Africa, with

Ryanair in particular operating 2.9 million seats in 2013 and also EasyJet
and Jet4You providing significant capacity on these cross-Mediterranean
routes.
To measure the effect of the liberalization process it may be of interest
not only to focus on increased volumes and operators, but also to have an
idea of the level of competition that African airlines face on their con-
nections, especially on intra-African routes, which are the focus of the
Yamoussoukro Decision. The higher the number of operating carriers on
a given route, the higher the level of competition is supposed to be. For
this purpose, Table 10.4 compares 1997 to 2013 in terms of routes’ market
structures. It can be seen that the percentage of intra-African routes oper-
ated under monopoly is reduced from 76 per cent to 68 per cent. This trend
is confirmed by the fact that available seats provided on monopoly routes
fell from 37 per cent to 28 per cent.
FORECASTED GROWTH AND PERSISTENT

PROBLEMS FOR AFRICAN AIRLINES
The African air transport industry is expected to be an important input to

the forecasted economic growth of Africa in the near future. In the next 20
years, there are forecasts of annual GDP increases of 3.7 per cent for the
continent (that is, higher than the world average of 2.9 per cent), making
it one of the fastest-growing regions in the world (Boeing Commercial
Aeroplanes, 2016). Such growth is driven by several factors, such as the
rising demand for natural resources available in Africa, and the recent
development of service industries such as banking, retail and telecom-
munication. Furthermore, the beginning of the process leading to the

M4371-FINGER_9781786431851_t.indd 193 14/11/2017 11:14

2035
Africa–Southeast Asia 2015
2008
Africa–North America
Africa–Middle East
Africa–Europe
Africa–Africa
0 50 100 150 200 250 300 350 400
Source: Boeing Commercial Aeroplanes (2016).
Figure 10.2 Air passenger traffic growth by regional flow in billions of

RPK
emergence of a middle class is seen as a catalyst for the growth of African

air transport industry more generally that, for its part, provides the oppor-
tunity for aviation to become a facilitator of international market intercon-
nectivity, of intra-Africa trade and of firms’ linking into the global supply
chain.
Other favourable factors pointing to growth in aviation are low popula-
tion density combined with high land availability, which makes the envi-
ronmental costs associated with an expansion of the aviation industry, and
particularly its infrastructure, much less restrictive than in other regions.
As a result, the traffic to, from and within Africa is projected to grow by
about 6 per cent per year up to 2035 (Figure 10.2), which is higher than
the world average of about 5 per cent and considerably above the expected
growth in mature markets such as North America and Europe (Boeing
Commercial Aeroplanes, 2016).
Thus, the potential for growth is also substantial for African airlines,
although a plethora of challenges remain to be overcome. These range
from the scarcity of demand across the continent to the higher costs
(compared to airlines in other regions of the world) borne by African car-
riers, as pointed out by Ssamula (2008), Heinz and O’Connell (2013) and
Button et al. (2105b). Other challenges include the legacy issues of limited
infrastructure, particularly away from the main urban areas, major ports
and mining areas.

M4371-FINGER_9781786431851_t.indd 194 14/11/2017 11:14

There are several reasons for such cost inefficiencies, mainly related to
disadvantages in fuel purchasing, mismanagement, fleet ageing, low load
factors, and expensive ticket distribution and complementary services.
Fuel transportation (across countries that are landlocked and/or have poor
infrastructure) and fuel purchasing (due to insufficient scale to hold bar-
gaining power with suppliers), along with the fact that fuel hedging is still
not a very common practice, put many African airlines at a disadvantage.
A second category of costs is the heritage of a long-standing policy of
government control of airlines that have traditionally been operated more
for political objectives rather than for economic efficiency. In particular,
many countries have consistently subsidized the operations of their inef-
ficient flag carriers, often creating costs associated with mismanagement.
Also, the ageing of fleets implies not only higher costs of maintenance but
also increased fuel consumption, poor reliability and increased downtime.
The inability to exploit the full economies of optimal hardware use is com-
pounded by poor scheduling on the part of the carriers and night flying
restrictions by the authorities which leads to one of the lowest aircraft
utilization rates in the world.
Also, African nations have the lowest average load factors in the world
(Figure 10.3). This not only reflects high fares but also lack of reliability
in services and the late emergence of a middle-income class, the group that
tends to make the most use of aviation.
Furthermore, the low penetration of both the internet and credit cards
in Africa makes ticket distribution more complicated and expensive.9
Moreover, because they were state-owned monopolies, airports and air
90% 2013 2014 2015

80%
70%
60%
50%
40%
30%
20%
10%
0%
AFRICA MIDDLE EUROPE NORTH LATIN ASIA
EAST AMERICA AMERICA AND
AND PACIFIC
CARIBBEAN
Source: International Air Transport Association (2013; 2014b; 2015b).
Figure 10.3 Regions’ intra- and inter-continental passenger load factors

M4371-FINGER_9781786431851_t.indd 195 14/11/2017 11:14

14%
12%
% of Total Seats
10%
8%
6%
4%
2%
0%
01
02
03
04
05
06
07
08
09
10
11
12
13
)
ay
20
20
20
20
20
20
20
20
20
20
20
20
20
–M
an
(J
14
20
Source: CAPA (2014).
Figure 10.4 LCCs’ share of total available seats within Africa
navigation service providers used to charge higher fares than their counter-
parts in the rest of the world.
These problems faced by African airlines have significantly hampered
the development of low-cost carriers (LCCs) in Africa. Unlike most
regions in the world, where LCCs have boomed followed the liberaliza-
tion of the aviation industry, only slightly more than 10 per cent of avail-
able seats in the intra-African market were offered by low-cost carriers
(Figure 10.4), compared to one-third of the global average of 26 per cent.
These numbers are even more notable when compared to those of other
developing regions such as South Asia and South-East Asia, which has
36 per cent and 53 per cent respectively of LCC market shares (Boeing
Commercial Aeroplanes, 2014). LCC services were provided mainly in
large domestic markets, such as South Africa and Kenya.
Furthermore, despite the improvements discussed in the previous
section, it seems that Africa’s airline network is not yet sufficiently devel-
oped. Ssamula (2008) suggested that Africa’s unique situation means it
must move toward a hub-and-spoke network structure based on four
main hubs, one for each African region: Northern, Western, Eastern and
Southern Africa. Such a structure should help airlines minimize their oper-
ating and network costs. However, many airports are still competing for
this position, especially in the Northern and Eastern parts of Africa, and
this struggle may slow down the development of the African air transport
network.
A further persistent problem with the African airlines has been their
safety record; in 2005, close to 25 per cent of the world’s aircraft crashes
occurred in Africa, while African flights only make up 5 per cent of the

M4371-FINGER_9781786431851_t.indd 196 14/11/2017 11:14

global traffic. However, there has been some improvement in recent years,
partly due to the pressure of the International Civil Aviation Organization
but also because of a general improvement in aviation safety. Despite the
improvements achieved in recent years, poor maintenance and a lack of
monitoring and control led to an increasing trend of blacklisting airlines
(Ssamula, 2008). As a result, several airlines have been prevented from
flying to the European Union and the United States, with a detrimental
effect on the growth of the African air transport sector. Moreover, the
infrastructures of most African airports are not good enough to meet
the new technological standards and the security demands imposed by the
industry.
INFRASTRUCTURE DEVELOPMENTS
Each country controls its own air space and, as such, is responsible for its
air navigation system (which includes air traffic control). However, there
has been some coordination in their operation, with the 19 members of the
Agency for Aerial Navigation Safety in Africa and Madagascar manag-
ing 620 000 square miles of airspace covering Antananarivo, Brazzaville,
Dakar Oceanic and Terrestrial, Niamey and N’Djamena from its control
centres located at the international airports in each region. There have
also been changes in the ways countries provide their own systems. Most
remain state-owned, but some are now institutionally separate from gov-
ernment; for example, South Africa’s Air Traffic and Navigation Services
Ltd is a not-for-profit joint-stock corporation, although a ministerial com-
mittee regulates its rates.
The quality of the air traffic control systems varies widely between and
even within countries. In general, North Africa, Morocco and Egypt have
reasonable infrastructures; west sub-Saharan Africa, Senegal, Guinea,
Ivory Coast and Ghana are adequately served, as are Kenya, Rwanda
and Tanzania on the east coast, and South Africa in the south. Indeed,
South Africa’s air traffic control is ranked 15th in the world by the World
Economic Forum in terms of its air transport infrastructure, while Tunisia,
Mauritius, Ethiopia, Morocco, Seychelles and Namibia are all in the top
60.10 The African interior, on the other hand, especially the vast expanses
of the Central African Republic and the Republic of the Congo, is largely
devoid of radar and good radio communications. In such areas, inter-
aircraft communication is the main way of maintaining separation.
One of the major problems in operating all infrastructures in most of
Africa, including air traffic control, is labour supply. The implications of a
shortage of suitable equipment in many parts of Africa are worsened by a

M4371-FINGER_9781786431851_t.indd 197 14/11/2017 11:14

shortage of skilled labour, and this is a problem that extends to other ele-
ments of the aviation supply chain. For example, if African aviation does
grow as predicted by Boeing and others, there will be a need for over 20 000
additional pilots over the next two decades. Problems of poor pay hinder
retention and the ‘poaching’ of qualified personnel by non-African airlines
adds to the issue of limited training capacity.
Most airports in Africa are relatively small in international terms.
No African airport even approaches the top 50 airports in the world in
terms of passenger numbers. In 2014, the continent’s three largest air-
ports in terms of passengers handled were, OR Tambo International
Airport Johannesburg (19 164 000 passengers), Cairo International Airport
(14 684 892), and Cape Town International Airport (8 636 294). In terms
of airport quality, there are also vast differences between countries and
between facilities within countries, with many airports being outdated and
poorly equipped, particularly in sub-Saharan Africa. Some of these facili-
ties have long since become incapable of efficiently handling the thousands
of passengers flowing through them. Added to this, there is considerable
variation in efficiency, even among the older airports. In many cases they
are also poorly used, and in some cases there is demonstrable underutiliza-
tion (Stephens and Ukpere, 2011).
Recently, there have been significant efforts to foster public−private
partnerships (PPPs) of various kinds, and joint ventures to both finance
investments in airports and to enhance their subsequent operational effi-
ciency. Despite the challenges of developing PPP contracts, there have been
some successes that, in conjunction with more conventional government-
based initiatives, have led to increases in actual and projected airport
expenditures. International agencies have also become more engaged in
airport development, both in terms of developing new facilities and signifi-
cantly upgrading the existing capacity. Overall, the Center for Asia Pacific
Aviation (CAPA) has found that at least US$33.8 billion has been invested,
or is earmarked to be invested, in construction and associated projects at
existing airports in 77 projects in more than nine African countries, with
an average price tag per project of almost $440 million. Examples include
Angola (30 projects costing $2.16 billion), Nigeria ($870 million on multi-
ple passenger and cargo terminal projects), Senegal (a new airport at Dakar
costing $700 million), Rwanda (a new airport at Kigali, $650 million), and
Kenya (new terminals and a second runway at Nairobi, $1 billion).
China has also become a major player in developing airports in Africa,
and although the exact scale of the commitment is unclear, the Chinese
Ministry of Commerce in 2014 suggested the net annual flow of Chinese
direct investment into Africa increased eight-fold between 2005 and 2014,
to $3.2 billion, with a significant part going to infrastructure of one kind or

M4371-FINGER_9781786431851_t.indd 198 14/11/2017 11:14

another, including aviation. As an example, Ethiopian Airports Enterprise

commenced the third and final phase of a $550 million expansion of Bole
International Addis Ababa Airport in 2014 with the work being under-
taken by China Communications Construction Company, supported by
funding provided by the Export–Import Bank of China.
CONCLUSION
The African air transport industry has experienced a process of partial lib-
eralization in recent years. The principles that inspired the Yamoussoukro
Decision have been implemented heterogeneously across the continent.
As a result, despite the growth in traffic flows – registered on both intra-
African and intercontinental connections – and the increases in both the
number of airlines and the level of competition, air transportation in
Africa is still suffering from some long-standing weaknesses, including a
lack of interconnectivity, an unbalanced traffic distribution, higher (com-
pared to the rest of the world) fares and airline costs, and sparse demand.
In sum, the market for air services in Africa is, overall, clearly healthier
than in the past but, with a few notable exceptions, still lacks the capacity
and services levels found in most other parts of the world.
NOTES
1. The forecasts are somewhat optimistic. Boeing Commercial Aeroplanes (2016), for
example, predicts that intra-Africa revenue passenger miles flown will grow by an
annual average of 6.7 per cent between 2015 and 2034, and those between Africa and the
Middle East and Asia by 7.3 per cent and 7.1 per cent, respectively. Similarly, air cargo
is projected to grow by 6.6 per cent annually compared to a global growth rate of 5 per
cent. The UN’s International Civil Aviation Organization (2014) indicates steady growth
with passenger traffic to, from and within Africa, increasing at an annual average rate of
5.3 per cent until 2030, while cargo traffic is projected to grow at 5.6 per cent. Africa–
Middle East traffic is forecast to expand at 9.9 per cent, the most rapid growth in the
world. Forecasts produced by Airbus offer a very similar picture.
2. The pre-war civil aviation market was extremely limited. For example, the total number
of passengers flown by British Imperial Airways throughout the 1930s was only about
50 000, and Africa involved only part of this network.
3. Many of them failed in the following years, mainly due to low demand and
mismanagement.
4. Where not otherwise specified, tables and figures are the result of authors’ elaboration
of OAG data.
5. The main problems at that time were about low security standards and unstable political
situations in many countries, which had led to the bankruptcy of many African airlines
in recent years, including relevant national carriers such as Air Nigeria and Air Gabon.
6. Ryanair and EasyJet, namely the two most important European LCCS, provide direct
flights toward Morocco and Egypt.

M4371-FINGER_9781786431851_t.indd 199 14/11/2017 11:14

7. British Airways also operated indirectly in the domestic South African air transport
market through its Comair and kulula.com brands.
8. O’Connell (2006) points out that almost 11 per cent of the Emirates traffic coming from
Europe and stopping in Dubai has a final destination located in Africa.
9. Commission payable to the travel agencies is typically about 7 per cent of the ticket price
(Heinz and O’Connell, 2013).
10. Button and McDougall (2006) also offered a detailed comparison of 11 major air traffic
control systems, including South Africa’s.
REFERENCES
Akpoghomeh, O.S. (1999), ‘The development of air transportation in Nigeria’,

Journal of Transport Geography, 7, 135–46.
Amankwah-Amoah, J. and Y.A. Debrah (2011), ‘The evolution of alliances in
the global airline industry: a review of the Africa experience’, Thunderbird
International Business Review, 53, 37–50.
Amankwah-Amoah, J. and Y.A. Debrah (2014), ‘Air Afrique: The demise of a
continental icon’, Business History, 56, 517–46.
Boeing Commercial Aeroplanes (2014), Current Market Outlook 2014–2033, avail-
able at www.boeing.com.
Boeing Commercial Aeroplanes (2016), Current Market Outlook 2016–2035, avail-
able at www.boeing.com.
Button, K.J. and G. McDougall (2006), ‘Institutional and structural changes
in air navigation service-providing organizations’, Journal of Air Transport
Button, K.J., A. Brugnoli, G. Martini and D. Scotti (2015b), ‘Connecting African
urban areas: airline networks and intra-Sub-Saharan trade’, Journal of Transport
Button, K.J., G. Martini and D. Scotti (2015a), ‘African decolonisation and air
transportation’, Journal of Transport Economics and Policy, 49, 626–39.
CAPA (2014), World Aviation Yearbook 2014, available at http://centreforaviation.
com/reports/files/39/yearbook_2014_final.pdf.
Guttery, B.R. (1998), Encyclopedia of African Airlines, Jefferson, NC: McFarland
and Company.
Heinz, S. and J.F. O’Connell (2013), ‘Air transport in Africa: toward sustainable busi-
ness models for African airlines’, Journal of Transport Geography, 31, 72–83.
International Air Transport Association (2013), Air Passenger Market Analysis,
May 2013, available at www.iata.org/economics.
International Air Transport Association (2014a), Economic Performance of the
Industry Mid-Year 2014 Report, available at www.iata.org/economics.
International Air Transport Association (2014b), Air Passenger Market Analysis,
July 2014, available at www.iata.org/economics.
International Air Transport Association (2015a), Economic Performance of the
Industry Mid-Year 2015 Report, available at www.iata.org/economics.
International Air Transport Association (2015b), Air Passenger Market Analysis,
July 2015, available at www.iata.org/economics.
International Civil Aviation Organization (2014), 2013: State of Air Transport,
available at http://www.icao.int.

M4371-FINGER_9781786431851_t.indd 200 14/11/2017 11:14

Njoya, E.T. (2016), ‘Africa’s single aviation market: the progress so far’, Journal of
Transport Geography, 50, 4–11.
O’Connell, J.F. (2006), ‘The changing dynamics of the Arab Gulf based airlines
and an investigation into the strategies that are making Emirates into a global
player’, World Review of Intermodal Transportation Research, 1, 94–114.
O’Connell, J.F. (2011), ‘The rise of the Arabian Gulf carriers: an insight into the
business model of Emirates Airline’, Journal of Air Transport Management, 17,
339–46.
O’Connell, J.F. and D. Warnock-Smith (2012), ‘Liberalization and strategic change
in air transport: an examination of current and future variations in tourist
traffic to and from Egypt resulting from policy changes at EgyptAir’, Tourism
Economics, 18, 845–70.
Schlumberger, C.E. (2010), Open Skies for Africa: Implementing the Yamoussoukro
Decision, Washington, DC: World Bank.
Ssamula, B. (2008), Strategies to Design a Cost-effective Hub Network for Sparse
Air Travel Demand in Africa, PhD. dissertation, University of Johannesburg.
Stephens, M.S. and W.J. Ukpere (2011), ‘Airport capacity utilization in Nigeria: a
performance and efficiency analysis’, African Journal of Business Management,
5, 11104–15.

M4371-FINGER_9781786431851_t.indd 201 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 202 14/11/2017 11:14

PART II
Topical issues

M4371-FINGER_9781786431851_t.indd 203 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 204 14/11/2017 11:14

11. Aviation safety in the age of
liberalization
Clinton V. Oster Jr, John S. Strong and
C. Kurt Zorn
This chapter analyzes the safety record of worldwide commercial aviation

in an era of liberalization. The first section describes the evolution of liber-
alization and the accompanying concerns about safety, the second section
analyzes the worldwide safety record during the liberalization era, the
next section examines the safety record in countries and regions pre- and
post-liberalization and the final section offers concluding observations and
lessons for improving safety in the future.
LIBERALIZATION AND AVIATION SAFETY

CONCERNS
By 2016, most of the world’s commercial passengers were flying in a

market-driven, liberalized aviation environment. But from the beginning
of the liberalization era, there was major concern that aviation safety
would degrade, as new airlines quickly began service, as airlines expanded
to new routes, and as lower fares spurred unprecedented industry growth
(see Moses and Savage, 1990). The deregulation of airlines in the United
States in the late 1970s was followed by a wave of aviation privatization
and liberalization around the world. As liberalization spread from North
America and Australia/New Zealand to the European Union and then to
much of the rest of the world, concerns about the possibility of reduced
safety continued.
The economic analyses of airline safety in the 1980s and early 1990s
focused on the potential safety effects of deregulation in the United States
(see Moses and Savage, 1989 and Bier et al., 2003). Analysts focused on
the comparative safety performance of industry segments, especially new
entrant carriers and the growth of smaller regional and commuter air-
lines that initially operated under less stringent safety regulations. Early
205

M4371-FINGER_9781786431851_t.indd 205 14/11/2017 11:14

studies found that airline safety records had continued to improve during
the deregulated era (Oster and Zorn, 1987). However, safety records for
new entrant airlines and for regional/commuter carriers were sometimes
significantly worse than for established carriers (Savage, 1999). Even with
this segment’s poorer safety performance, Savage (1999) noted that overall
intercity travel safety improved as passengers shifted modes from auto
travel to airlines.
Liberalization in Canada and Australia differed from the United States
because of the prior dominance of two airlines (one state-owned) in each
country. As such, privatization went hand-in-hand with liberalization and
proceeded on much more of an evolutionary track. Studies found no sig-
nificant effect on airline safety following liberalization in these countries,
perhaps in part because changes in industry structure and the regulatory
environment were quite different from those in the United States (Forsyth,
1991; Oster et al., 1992).
Liberalization has also had different characteristics across countries and
regions. Some countries adopted evolutionary approaches to deregulation,
as in the three liberalization packages in Western Europe between 1987 and
1993. Others adopted a more complete change all at once, such as occurred
in India and South Africa. In some settings, larger changes in the macro
environment encompassed the aviation sector, as occurred in Russia, the
countries of the former Soviet Union, China, India and Brazil.
Many of the concerns that airline deregulation might adversely impact
safety stem from four factors: industry growth, new entrants, financial
stress, and changes in aviation markets (Moses and Savage, 1989; Oster
et al., 1992). First, industry growth following liberalization has been a
common experience worldwide. Pricing deregulation and new entrants
(especially low-cost airlines) have resulted in lower fares and spurred
demand, with passenger volumes sometimes growing quite dramatically.
This growth creates challenges for safety. Rapid growth and the emergence
of new mainline and regional carriers have also dramatically increased the
workload on aviation regulatory authorities, who have not always been
able to respond quickly in terms of budgets and other resources (Button,
2004; Finger and Piers, 2005; General Accounting Office, 1996; Yadav and
Nikraz, 2014). Airport and aviation infrastructure may face capacity con-
straints that may induce congestion and operational challenges, and affect
workloads of air traffic control organizations.
Second, new entrant carriers may pose challenges to aviation safety, as
they have often operated in new markets, with older equipment and less
experienced personnel. New entrants frequently begin service with older,
leased aircraft, which can present maintenance challenges (Kanafani and
Keeler, 1990). Industry growth also increases demand for skilled labor,

M4371-FINGER_9781786431851_t.indd 206 14/11/2017 11:14

Aviation safety in the age of liberalization 207
especially pilots and mechanics, which can create a situation where the
least experienced personnel are operating in more challenging environ-
ments. This growth has consequences for flight crew and maintenance
training, which also has the potential to affect safety.
Third, lower fares and legacy costs often produce financial stress in the
older airlines as well. There is always a concern that a more competitive,
market-driven environment will create incentives to cut costs and take risks
that could result in decreases in safety (Dionne et al., 1997; Golbe, 1986;
Madsen, 2013; Rose, 1990).
Fourth, the consequences of liberalization have varied as well. While
lower fares and rapid passenger growth have generally resulted, other
industry structure effects have been more varied. For example, in some
countries liberalization has resulted in a boom of new entrant carriers,
some of which have succeeded, others of which have not. As a result of
increased competition, liberalization has also brought both financial and
operating pressures to legacy, state-owned carriers. More recently, some
regions have seen an increase in industry consolidation. The merger of two
carriers involves combining different maintenance, inspection, training
and safety programs. Even if the original programs were well designed and
well executed, the consolidation to a single program has the potential to
impact safety negatively.
However, liberalization may also have improved safety by changing
the nature of aviation markets. In general, scheduled services have grown
faster than, and partially substituted for, non-scheduled charter operations.
Scheduled services have long had a better safety record. Industry growth
has spurred the development and adoption of a new generation of smaller
jets, replacing older and less safe piston and turboprop aircraft (Barnett,
1990). There also is the potential for learning: countries that liberalized in
more recent periods had the advantage of the lessons from earlier deregu-
lation experiences. In addition, aviation has been able to learn from safety
analysis and management in other industries that have similar risk char-
acteristics in terms of key personnel, technology and systems (Biermann,
2016; Drax et al., 2014; Stolzer and Goglia, 2016).
THE WORLDWIDE SAFETY RECORD IN THE ERA

OF LIBERALIZATION
In the aggregate, such safety concerns may at first seem unwarranted.

While there has been year-to-year variation, the worldwide airline accident
rate (in terms of either accidents per million flights or passenger fatalities
per million passengers carried) has generally declined over the past half

M4371-FINGER_9781786431851_t.indd 207 14/11/2017 11:14

century. These accident rates have continued to decline in the past three
decades of liberalization. But in terms of safety, the airline industry is far
from homogeneous and far from static. Both accident rates and underlying
causes have changed over time, and have varied across regions of the world
and across different types of carriers.
Assessing safety performance pre- and post-liberalization is difficult for
a number of reasons. First, determining the pre- and post- periods is chal-
lenging. Many countries experienced liberalization as a series of stages that
moved in an evolutionary way from a heavily regulated to a largely deregu-
lated industry. The best example of this is Western Europe, in which three
liberalization ‘packages’ occurred from 1987–93, with phase-in periods
to each. In addition, liberalization can take place in both domestic and
international settings, which may have different effects on safety perfor-
mance. Second, liberalization has had different characteristics in different
settings. Third, data limitations have become more prevalent as liberaliza-
tion has spread to emerging markets and regions. Many safety records are
incomplete. Operational data is often inconsistently reported, making it
more challenging to construct and interpret accident rates. Moreover, lack
of data consistency and availability makes it very hard to analyze specific
industry segments, let alone individual airlines.
In this section, to classify causal factors and the type of service, we
analyze the comprehensive set of 771 worldwide commercial accidents in
which passengers were killed between 1990 and 2015. Several data issues
have shaped our approach in this chapter. First, our focus is on commercial
accidents in which passengers were killed, and the fatality rates for those
accidents. Data reporting requirements and effectiveness vary across coun-
tries, regions and types of service, and non-fatal accidents are often unre-
ported or lack key data about what occurred. Passengers are also killed in
private or general aviation aircraft, but here again, reporting of those acci-
dents is inconsistent and operations data are often unavailable, particularly
in developing countries. In contrast, fatal commercial accidents generally
involve more extensive reporting and investigation. Second, accidents are
classified in terms of carrier and home country, rather than location. (For
most cases, these are within the same country or region.) Third, we define
pre- and post-liberalization periods in broader, multi-year terms. Fourth,
where data is missing or inconsistent, we discuss what effect this might
have on our analysis and conclusions, as well as the implications for future
research.
Our primary source for fatal accident data is the World Air Accident
Summary (WAAS) (Ascend Flight Global, 2016). Researched and pub-
lished on behalf of the UK Civil Aviation Authority, WAAS includes
detailed descriptions for 8000 accidents involving jet and turboprop and

M4371-FINGER_9781786431851_t.indd 208 14/11/2017 11:14

Indices of Passengers, Fatalities and Fatality Rates

300
250
Index 1990–1994 = 100
200
Passengers Index
150
Fatalities Index
Fatality Rates Index
100
50
0
1990
1995
2000
2005
2010
2015
Source: Based on ICAO and WAAS data.
Figure 11.1 Indices of passengers carried, passenger fatalities, and

passenger fatality rates, 1990–2015, 5-year moving average
piston-powered aircraft as well as helicopter accidents. Our primary

resource for operational data is ICAO Data Plus from ICAO, the
International Civil Aviation Organization (International Civil Aviation
Organization, 2016). ICAO Data comprises detailed financial, operational,
traffic, personnel and fleet information about commercial air carriers from
191 member countries for the 1990–2015 period. ICAO Data prior to 1990
is often incomplete and in many cases missing. Our analysis uses opera-
tions and safety data from 1990–2015.
Figure 11.1 shows trends in passengers carried, passenger fatalities and
fatal accident rates, calculated as an index, with 1990–1994 5 100 as a base.
We used a five-year moving average in this figure to identify trends without
distortions from a given year. Fatal accidents are discrete events and, fortu-
nately, don’t happen very often. Plotting fatality rates or fatalities year by
year for a country or region or type of service can result in a very jumpy
plot where the underlying trends are more difficult to see and compare.
Figure 11.1 suggests that passenger fatality rates have fallen stead-
ily throughout most of the 1990 to 2015 period. Worldwide, passenger

M4371-FINGER_9781786431851_t.indd 209 14/11/2017 11:14

fatalities per 1 million passengers carried fell over 80 percent from 1.01 in
the 1990–94 period to 0.18 in the 2011–15 period. At the global level, it
appears that the airline safety has continued to improve during the exten-
sion of liberalization across the world since 1990.
For our safety index, we use passenger fatalities per million passengers
carried. It would also be possible to use fatal accidents per million aircraft
departures; patterns seen using that measure would be similar to Figure
11.1. It is important to realize, however, that in spite of the impression
one might get from looking at media coverage of aircraft accidents, many
passengers survive crashes in which one or more passengers is killed.
Historically, in about 51 percent of fatal crashes, all passengers on board
are killed. In 49 percent of fatal crashes, there are survivors. Overall, about
68 percent of passengers on board in fatal crashes are killed, and about 32
percent survive.
Table 11.1 shows the distribution of commercial accidents with passenger
fatalities by type of service. During the entire 1990–2015 period, with more
than 48 billion passengers carried, there were 771 fatal accidents involving
commercial passenger airlines, involving 21 727 fatalities. The global fatal
accident rate for the entire period was 0.45 passenger fatalities per million
passengers. As can be seen in the table, scheduled service dominates pas-
senger service worldwide, accounting for 96 percent of passengers carried,
but accounting for only 80 percent of passenger fatalities. Non-scheduled
service, especially non-scheduled domestic service, accounts for a dispro-
Table 11.1 Passengers carried, passenger fatalities and passenger fatality

rates by type of service
Type of Service Passengers Share of Passenger Share of Passenger

1990–2015 Carried Passengers Fatalities Passenger Fatalities
Carried Fatalities per Million
Passengers
Scheduled 30 168 410 353 62% 10 127 47% 0.34
Domestic
Scheduled 16 314 265 654 34% 7094 33% 0.43
International
Nonscheduled 1 627 015 478 3% 2007 9% 1.23
International
Nonscheduled 258 512 328 1% 2499 12% 9.67
Domestic
All Services 48 368 203 813 100% 21 727 100% 0.45
Source: Authors’ calculations from ICAO and WAAS data.

M4371-FINGER_9781786431851_t.indd 210 14/11/2017 11:14

portionate share of passenger fatalities. In many countries, this segment

operates under less stringent safety regulation than scheduled airlines, in
terms of surveillance of flight operations, pilot regulations, regulatory
standardization, maintenance quality assurance and regulatory standardi-
zation. Ballard (2014) provides an analysis of these factors in United States
non-scheduled service. Taken together, non-scheduled domestic and inter-
national services account for only 4 percent of passengers carried but about
20 percent of passenger fatalities. Passenger fatality rates for non-scheduled
domestic services worldwide were more than twenty times higher than those
for scheduled services during the 1990–2015 period.
These numbers need to be interpreted with some caution. While we
believe that the passenger fatality data is reasonably complete because of
the worldwide attention given to fatal commercial accidents, we also believe
that the passengers carried data is not as accurate or consistent. We believe
that the scheduled data is reasonably accurate in the aggregate because it
is dominated by very large jet carrier airlines who keep accurate track of
their data and who report it regularly to ICAO. Even here, however, we
find what appears to be missing data when a carrier will simply not report
data for a given year when they report similar data for the years prior to
and following the missing data. These omissions suggest that the actual
numbers of passengers carried might be higher. On the other hand, we also
found cases where a carrier double-reported passenger data by reporting it
once as service to a particular destination and then again as an aggregate
of service to multiple destinations. Similarly, there were also cases where a
fairly steady stream of data for a given carrier had a single year that was
many times higher than the years surrounding it. While we found and cor-
rected for some of these issues, it is likely that we did not identify all situa-
tions. We believe, though, that data underreporting would not change the
major conclusions. For the most part, however, these kinds of irregularities
were found in small carriers and had little effect on the aggregate numbers.
With non-scheduled service, we found numerous instances where an
accident was reported for a carrier but no operations data was reported
for that carrier at all or none was reported for the year of the accident.
These carriers who underreport appear to be very small carriers, but there
also appear to be more of them than is the case for scheduled service carri-
ers. With non-scheduled carriers, we believe that the aggregate passengers
carried numbers are underreported, but we don’t know by how much.
That said, given that the non-scheduled domestic passenger fatality rate is
well over twenty times higher than the scheduled domestic fatality rate, we
believe that even if it were possible to correct for underreporting of pas-
senger data, the rate would still be dramatically higher for non-scheduled
service than for scheduled service.

M4371-FINGER_9781786431851_t.indd 211 14/11/2017 11:14

Table 11.2 Passengers carried and passenger fatalities by region,

1990–2015
Region Passengers Share of Passenger Share of Passenger

Carried Passengers Fatalities Passenger Fatality
Carried Fatalities Rate
Africa 403 176 030 1% 2892 13% 7.17
Australia and Oceania 933 206 315 2% 321 1% 0.34
Central America and 761 622 722 2% 1026 5% 1.35
Caribbean
China 4 450 113 122 9% 1764 8% 0.40
Eastern Europe 469 963 724 1% 228 1% 0.49
Former Soviet Union 1 196 007 046 2% 2779 13% 2.32
Middle East and 2 830 901 425 6% 2311 11% 0.82
North Africa
North America 17 267 574 023 36% 2062 9% 0.12
North Asia 3 205 865 855 7% 384 2% 0.12
South America 2 166 907 509 4% 2121 10% 0.98
South Asia Pax 1 132 309 519 2% 1322 6% 1.17
Southeast Asia 2 467 032 439 5% 2678 12% 1.09
Western Europe 11 083 524 084 23% 1839 8% 0.17
Worldwide 48 368 203 813 100% 21 727 100% 0.45
Table 11.2 shows the distribution of fatal accidents by region. There are
marked variations. The fatal accident rate for sub-Saharan Africa
is by far the highest, at 7.17 fatalities per million passengers. Other
regions with rates much higher than the worldwide average include the coun-
tries from the former Soviet Union, Central America and the Caribbean,
South Asia, and Southeast Asia. These regions all have fatal accident rates of
more than about 1 per million, well over twice the worldwide average. There
are also individual countries which have statistically significantly higher rates,
including Russia (1.77) and Indonesia (1.60). It should be noted, though,
that even these higher rates represent improvements in aviation safety in these
countries and regions relative to their performance in the 1970–89 period
(Oster et al., 1992).
Why did these airplanes crash? The approach for assigning causes to
accidents which is taken in this chapter and in the authors’ prior work, is
to select as the cause the factor that initiated the sequence of events that
culminated in the accident (Oster et al., 1992; see also Ancel et al., 2015).
The assumption behind this approach is that, in the absence of the factor

M4371-FINGER_9781786431851_t.indd 212 14/11/2017 11:14

Table 11.3 Share of passenger fatalities by cause by service, 1990–2015
Cause Category Share of Passenger Fatalities
All Domestic International International Domestic Non-

Service Scheduled Scheduled Non- scheduled
scheduled
Pilot Error 40% 40% 44% 31% 39%

Equipment Failure 25% 23% 27% 33% 18%
Terrorism/Criminal 9% 8% 12% 12% 6%
Activity
Weather/Environment 9% 15% 3% 6% 8%
Other Aircraft 4% 3% 6% 3% 5%
Non-pilot Personnel Error 4% 2% 4% 13% 0%
Air Traffic Control Error 2% 4% 0% 1% 0%
Aircraft Not Recovered 1% 0% 3% 0% 1%
Cause Ambiguous 6% 6% 1% 2% 23%
Total Passenger Fatalities 21 727 10 127 7094 2007 2499
Passenger Fatalities by 100% 47% 33% 9% 12%
Service
Passengers Carried by 100% 62% 34% 3% 1%
Service
Fatalities per Million 0.45 0.34 0.43 1.23 9.67
Passengers
that initiated the chain of events resulting in an accident, the accident

could have been avoided. A benefit of focusing on the sequence initiating
cause means that when, for example, pilot error is identified as the cause, it
refers to what may be characterized as an ‘unforced’ pilot error rather than
a failure to respond properly to an emergency when there may be a conflu-
ence of events that are difficult to respond to regardless of how talented
the pilot is or how good their training was. The authors have developed,
and refined over many years and after reviewing thousands of accidents, a
set of rules and definitions to guide how causes are assigned to accidents.
The goal in developing these rules is to be consistent in assigning causes so
that it is possible to make meaningful comparisons of how the distribution
of causes varies over time, across different segments of the industry, and
across countries or regions (Oster et al., 1992).
Table 11.3 shows the share of accidents overall and in each type of
service by cause. Pilot error was found to be the most frequent cause,
accounting for around 40 percent of all fatal accidents, with equipment
failure the second most frequent cause, accounting for around 25 percent.

M4371-FINGER_9781786431851_t.indd 213 14/11/2017 11:14

These two primary causes are relatively unchanged from the 1970–89
period. Terrorism and criminal activity were a more common cause in the
1990–2015 period than earlier.
Also of concern is the large share of accidents in non-scheduled service
where the cause was ambiguous. These were cases where there simply was
not enough information about the crash to determine what happened.
Many non-scheduled accidents, particularly in developing countries, simply
do not receive the investigative attention that the scheduled service crashes
do. Also, many of the aircraft used in non-scheduled service are smaller,
turboprop or piston driven aircraft that are not equipped with flight data
recorders or cockpit voice recorders. In domestic scheduled service, for
example, jet aircraft account for 37 percent of accidents and piston powered
aircraft account for 7 percent of accidents. In contrast, for domestic non-
scheduled service, jet aircraft account for only 12 percent of accidents while
piston powered aircraft account for 47 percent. Many non-scheduled air-
craft also crash in remote locations where it can be days (or longer) before
safety officials can examine the site of the crash. This represents an impor-
tant challenge in how to devote resources to improve safety in the future.
AVIATION SAFETY PRE- AND

POST-LIBERALIZATION
As noted above, attempting to define precise starting points for aviation

liberalization is difficult because most changes occurred either in stages or
as an evolution. We attempted to establish liberalization dates in terms of
three criteria: the removal of restrictions on domestic or regional carrier
entry; the removal of restrictions on route entry; and fare deregulation.
These corresponding dates for countries and regions in our analysis are
shown in Table 11.4. The table shows a worldwide rollout of liberalization
in the 1990s and early 2000s, following deregulation in the United States
in 1978 and liberalization in Canada, the UK/Ireland and Europe in the
1980s. By the early 2000s, liberalized aviation had become widespread.
Safety performance can be analyzed by studying accident rates in the
multiyear period before and after liberalization. Table 11.5 shows pas-
senger fatality rates for Southeast Asia, Turkey, Mexico, Indonesia, Brazil,
China, and India before and after their respective liberalizations. The
table also shows worldwide rates for the same periods as a baseline for
comparison and the percentage change in fatal accident rates following
liberalization. In five of the seven cases (Southeast Asia, Mexico, Brazil,
China, and India), the passenger fatality rate improved much more follow-
ing liberalization than it had improved worldwide during the same periods.

M4371-FINGER_9781786431851_t.indd 214 14/11/2017 11:14

Table 11.4 Airline liberalization by country and region
Country Year Information

United States 1978 Airline Deregulation Act; limited prior
pricing liberalization 1975–77
Canada 1984 Entry, exit, fares, routes deregulated
UK, Ireland 1986 Liberalized air services agreements: new
carriers and route and pricing freedoms
Europe 1987 First Package: ability to lower fares; route
entry to new carriers
1990 Second Package: expanded route and market
access; more fare deregulation
1993 Third Package: single licensing in region;
expanded and made permanent entry,
capacity, and pricing freedoms
Australia 1990 End of two airline policy; route fare freedoms
in domestic market
New Zealand 1990 Progressive liberalization late 1980s; domestic
and regional deregulation with Australia
Brazil 1991 Privatizations and bilateral liberalization
1998 Route and fare liberalization
Colombia, Peru, Chile 1991 Andean pact liberalized entry and pricing in
domestic and regional routes
Mexico 1991 New carrier entry relaxed
1995 New civil aviation law
2002 Route and fare liberalization
2005 New wave of low cost carriers; privatization
of Mexicana
2007 Privatization of Aeroméxico
Nigeria 1991 Deregulated routes, fares, services and new
entry
Russia 1991 Break-up of Aeroflot; new private carrier
entry; route and fare deregulation
South Africa 1991 New domestic air policy; free route, pricing
and entry restrictions
Scandinavia 1992 Opened regional market following
privatization and domestic deregulation
India 1994 Permitted entry of private airlines; removed
restrictions of fares and en-route entry,
capacity, frequency and services
China 1997 Liberalization of routes and fares;
subsequent establishment of Big 3 policy
2002
Indonesia 2000 New Aviation Policy 2000 allowed new
carriers; deregulated fares and route entry
Turkey 2003 Domestic new carrier entry permitted; routes,
fares deregulated
Southeast Asia 2008 Beginning of 5-year phased Single Aviation
Market in ASEAN countries

M4371-FINGER_9781786431851_t.indd 215 14/11/2017 11:14

Table 11.5 Safety before and after liberalization
Country/ Year of Fatalities per Million Passengers

Region Liberali-
Liberalized Country/Region Worldwide
zation
Rate Five Rate Five Percent Rate Five Rate Five Percent
Years Years Change Years Years Change
Before After in Rate Before After in Rate
Southeast 2008 0.94 0.08 –91% 0.37 0.20 –47%
Asia
Turkey 2003 0.01 0.42 3774% 0.65 0.34 –48%
Mexico 2002 0.71 0.00 –100% 0.69 0.37 –46%
Indonesia 2000 4.60 2.15 –53% 1.02 0.49 –52%
Brazil 1998 1.06 0.16 –85% 1.03 0.57 –45%
China* 1997 3.01 1.05 –65% 1.08 0.65 –40%
India* 1994 4.91 0.16 –97% 0.99 0.95 –3%
Note: *Because of data limitations the rates for China and India were calculated for four
years before and after rather than five years.
In one case, Indonesia, it improved only slightly more than the worldwide
rate. In only one case, Turkey, did the fatality rate rise. In Turkey, however,
the five-year period prior to liberalization had one accident in which only
one passenger was killed. In the five-year period following liberalization,
Turkey also had only one accident, this time in which 50 passengers were
killed. While Turkey’s fatal accident rate per million passengers increased
following liberalization, the number of fatal accidents was unchanged from
the prior period.
Excluding Turkey, all of the countries experiencing liberalization had a
lower passenger fatality rate in the period following liberalization than in
the similar period prior to liberalization. Also, again with the exception of
Turkey, all of these countries saw their passenger fatality rate drop by more
than the worldwide passenger fatality rate over the same periods.
Recall from Table 11.4 that many of the liberalizations happened before
1990 (when our data begins) or so soon after 1990 that it wasn’t possible to
construct multi-year before and after comparisons. Figure 11.2 compares
what has happened to passenger fatality rates for four of those liberaliza-
tions, Canada, Western Europe, the United States and Australia, during
the 1990 to 2015 period. In all four cases, fatality rates were lower than the
worldwide rate at the beginning of the period and continued to decline,
remaining below the worldwide rate.

M4371-FINGER_9781786431851_t.indd 216 14/11/2017 11:14

Fatality Rate 5-Year Moving Average

1.2
1.0
Fatalities per Million Passengers
0.8
Worldwide
Canada
0.6
Western Europe
United States
0.4
Australia
0.2
0
1990
1995
2000
2005
2010
2015
Figure 11.2 Fatality rates for Canada, Western Europe, the United States
and Australia
CONCLUDING OBSERVATIONS
Starting with airline deregulation in the United States in the late 1970s and
continuing through the recent liberalizations, there have been continuing
concerns that liberalizations might cause safety to degrade through some
combination of the pressur es on personnel and infrastructure of rapid
industry growth and the emergence of new entrant airlines. But these liber-
alizations have occurred against a backdrop of steadily improving aviation
safety because of lessons learned from investigating accidents, improved
equipment, and improved approaches for training pilots, mechanics and
others involved in aviation (Cui and Li, 2015; Evans, 2016). Also, each
liberalization brought forth lessons for future reforms, so that countries
could shape and be better prepared as their own version of deregulation
unfolded.
In examining the safety records of countries and regions both before
and after liberalization, our analysis finds that safety has generally

M4371-FINGER_9781786431851_t.indd 217 14/11/2017 11:14

improved following liberalization and often at rates that were greater than
the worldwide improvement in safety rates. These results suggest that lib-
eralization has not prevented or even hampered improvements in aviation
safety. Indeed, to the extent that liberalization has sped up the substitution
of scheduled service for the generally less safe unscheduled service, liberali-
zation may even have helped improve aviation safety.
While overall aviation safety has continued its improvement through
the age of liberalization, it remains true that variations in safety perfor-
mance persist across regions, carriers and segments of the industry. Rapid
growth has been especially challenging in developing countries and in set-
tings where new entrant carriers are operating older piston and turboprop
aircraft. Looking ahead, a major challenge to improving safety is that a
significant proportion of airline accidents now involve situations where
information about what happened is limited or non-existent, especially in
non-scheduled services in developing aviation markets.
REFERENCES
Ancel, E., A.T. Shih, S.M. Jones, M.S. Reveley, J.T. Luxhoj and J.K. Evans (2015),
‘Predictive safety analytics: inferring aviation accident shaping factors and cau-
sation’, Journal of Risk Research, 18(April), 428–51.
Ascend Flight Global (2016), ‘World air accident summary’, database available at
http://www.ascendworldwide.com/what-we-do/ascend-data/accident-and-loss-
data/.
Ballard, S. (2014), ‘The U.S. commercial air tour industry: a review of aviation
safety concerns’, Aviation, Space, and Environmental Medicine, 85(2), 160–66.
Barnett, A. (1990), ‘Air safety: end of the golden age’, Chance: New Directions for
Statistics and Computing, 3, 8–12.
Bier, V., J. Joosten, D. Glyer, J. Tracey and M. Welsh (2003), Effects of Deregulation
on Safety: Implications Drawn from the Aviation, Rail, and United Kingdom
Nuclear Power Industries, Norwell, MA: Kluwer Academic.
Biermann, T. (2016), Safety Management in Aviation – and Beyond, Wildau: Wildau
Verlag.
Button, K. (2004), ‘Conforming with ICAO Safety Oversight Standards’, Journal
of Air Transport Management, 10(4), 249–55.
Cui, Q. and Y. Li (2015), ‘The change trend and influencing factors of civil avia-
tion safety efficiency: the case of Chinese airline companies’, Safety Science, 75
(June), 56–63.
Dionne, G., R. Gagné, F. Gagnon and C. Vanasse (1997), ‘Debt, moral hazard, and
airline safety’, Journal of Econometrics, 79, 379–402.
Drax, C., A. Wittmer and R. Muller (2014), Aviation Risk and Safety Management:
Methods and Applications in Aviation Organizations, Cham: Springer.
Evans, J.K. (2016), Differences in Characteristics of Aviation Accidents During
1993–2012 Based on Flight Purpose, Hampton, VA: NASA Center for Aerospace
Information.

M4371-FINGER_9781786431851_t.indd 218 14/11/2017 11:14

Finger, M. and M. Piers (2005), ‘Air transport regulation under transformation: the
case of Switzerland’, Journal of Air Transportation, 10(2), 20–32.
Forsyth, P. (1991), ‘The regulation and deregulation of Australia’s domestic airline
industry’, in K. Button (ed.), Airline Deregulation: International Experiences,
New York: New York University Press, pp. 48–84.
General Accounting Office (1996), Airline Safety: New Airlines Illustrate Long-
Standing Problems in FAA’s Inspection Program, Report GAO/RCED-9 7–2,
Washington, DC: US Government Printing Office.
Golbe, D.L. (1986), ‘Safety and profits in the airline industry’, Journal of Industrial
Economics, 34(3), 305–18.
International Civil Aviation Organization (2016), ICAOData1 database, available
at https://www4.icao.int/NewDataPlus/.
Kanafani, A. and T. Keeler (1990), ‘Air deregulation and safety: some econometric
evidence from time series’, Logistics and Transportation Review, 26(3), 203–209.
Madsen, P.M. (2013), ‘Perils and profits: a reexamination of the link between
profitability and safety in U.S. aviation’, Journal of Management, 39(3), 763–91.
Moses, L. and I. Savage (1989), Transportation Safety in an Age of Deregulation,
New York: Oxford University Press.
Moses, L.N. and I. Savage (1990), ‘Aviation deregulation and safety: theory and
evidence’, Journal of Transport Economics and Policy, 24(2), 171–88.
Oster, C.V. and C.K. Zorn (1987), ‘Deregulation’s impact on airline safety’, Journal
of the Transportation Research Forum, XXVIII(1), 3–12.
Oster, C., J. Strong and C.K. Zorn (1992), Why Airplanes Crash, New York: Oxford
University Press.
Rose, N. (1990), ‘Profitability and product quality: economic determinants of
airline safety performance’, Journal of Political Economy, 98(5), 944–64.
Savage, I. (1999), ‘The economics of commercial transportation safety’, in J.
Gomez-Ibanez et al. (eds), Essays in Transportation Economics and Policy: A
Handbook in Honor of John R. Meyer, Washington, DC: Brookings Institution
Press.
Stolzer, A.J. and J.J. Goglia (2016), Safety Management Systems in Aviation,
London: Routledge.
Yadav, D.K. and H. Nikraz (2014), ‘Implications of evolving civil aviation safety
regulations on the safety outcomes of air transport industry and airports’,
Aviation, 18(April), 94–103.

M4371-FINGER_9781786431851_t.indd 219 14/11/2017 11:14

12. Small community impacts of
liberalization and the provision
of social air services
Aisling Reynolds-Feighan
INTRODUCTION
Air transportation is an important ingredient in the development and

growth of economies, and plays an important strategic role in these pro-
cesses at varying spatial scales. Air transport provides a relatively low-cost
transport solution for facilitating rapid access to isolated communities or
regions (islands, peripheral locations, difficult terrain), and in large coun-
tries can provide political and economic cohesion by facilitating higher
levels of interaction than other transport modes. Historically, govern-
ments have funded air transport infrastructure provision, and while this is
still the case in most global regions, there is greater use of public–private
partnerships as well as solely private sector provision. A report for the EU
in 2016 suggested that 15 per cent of airports around the world were fully
privatized, 18 per cent were in public–private partnerships and the remain-
ing 67 per cent were in public ownership. The privatized or commercialized
airports account for 50 per cent of airport passenger traffic.1 As well as
providing facilities, governments have traditionally been heavily involved
in the operational and regulatory aspects of the industry directly and
through all sorts of agencies promoting various agendas such as tourism
and local business development.
Air transport service provision has undergone significant changes in
many countries and regions globally since the 1970s. The deregulation
of the US domestic cargo and passenger air transport markets in 1977
and 1978 led to significant industry structure and performance changes
in the following decade. The US industry has been extensively studied
and has provided strong evidence for social benefits associated with a
market-oriented approach to the determination of air service provision
and pricing. Deregulation has permitted airlines to make determinations
on which communities and routes to serve and on capacity and pricing;
220

M4371-FINGER_9781786431851_t.indd 220 14/11/2017 11:14

Small community impacts of liberalization 221
it has removed restrictions mandating airlines to serve particular com-

munities (often small, remote communities) and reduced the need for
cross-subsidization within the carrier’s operations. In an international
context, liberalization of the traditionally restrictive bilateral air service
agreements between pairs of countries gathered momentum during the
1980s and 1990s, giving airlines greater freedom to choose where and how
they operate and price their services. Liberalization relates to the trade
rules determining market access, national treatment and levels of foreign
ownership and other non-tariff barriers (Decurtins, 2007). The formation
of multi-country trade blocs – such as the European Economic Area, Latin
American MERCOSUR (Southern Common Market) and Association
of Southeast Asian Nations (ASEAN) – which have common markets,
and multilateral free-trade areas – such as the Common Market for
Eastern and Southern Africa (COMESA), Greater Arab Free Trade Area
(GAFTA), North American Free Trade Agreement (NAFTA) and Pacific
Alliance (PAFTA) – have included air transport liberalization agreements
that facilitate and develop air transport services between countries. These
types of agreements facilitate economic growth and development and are
linked with globalization; connectivity and accessibility of the world’s
economies have been greatly facilitated by air transportation.
Under regulation, service provision to regional airports and small com-
munities was required as part of carrier licences. In some cases, regulation
prevented entry and exit from an agreed list of routes, so some degree of
cross-subsidization was assumed between profitable, heavily trafficked
routes and thin routes to remote or small communities.
In liberalized markets, governments have developed frameworks to
provide social air services so that small communities can have, or continue
to have, access to regional or national air transport networks. Competitive
tendering to provide social air services has emerged as the international
best practice for filling air service needs or requirements when the market
will not produce the service. In addition, governments continue to provide
air transport infrastructure to small communities or remote regions. There
is an implicit understanding that the infrastructure will benefit the com-
munity if services are provided – so service provision processes are increas-
ingly being put in place.
For small and medium-sized communities, access to national and interna-
tional markets is important for economic and political cohesion and develop-
ment. The processes of air transport deregulation and liberalization have had
significant and sustained impacts on the growth of the air transport industry.
This chapter will focus on the distributional effects of these processes, and
particularly examine how air transport services have evolved at small and
medium-sized communities worldwide. In the next section, major global

M4371-FINGER_9781786431851_t.indd 221 14/11/2017 11:14

trends are reviewed in a comparative regional analysis of key drivers and

indicators of air transport activity in the last 20 years. Following from this, a
classification of air transport locations is presented, focusing on the perfor-
mance of medium and small communities as well as exploring the patterns
of connectivity to the larger centres worldwide. The major global regions are
compared and contrasted, and particular features are highlighted that have
shaped the development of air transport in different settings. Finally the
provision of essential air services by governments is reviewed before some
general conclusions are set out in the concluding section.
MAJOR TRENDS IN AIR TRANSPORT, 1996–2015
Detailed reliable air traffic data covering all of the world’s commercial air
transport activity was sourced from the Official Airline Guide. The OAG
Max Historical Plus databases give ex post daily schedules of all of the
commercial air transport services offered for sale. Using these databases,
annual traffic capacity data series were compiled and matched with popu-
lation data and basic geographic information on cities and countries. The
analysis presented in the chapter draws from these data series and presents
a supply-side view of air transport activity, since it is capacity rather than
actual passenger volumes that are analysed. The analysis is restricted to jet
services (including regional jets).
Passenger air transport activity, as measured by non-stop departure
movements, has grown at an annual average rate of 3.2 per cent through-
out the 1996 to 2015 period, and a 3.4 per cent average annual growth in
available seats, despite two major global recessions in 2001 to 2002 and
2008 to 2010 being observed. The deployment of significant fleets of small
regional jets during the 2000s, particularly in the North American market,
reduced the average aircraft size during this decade. In the more recent
period, average aircraft size has increased and the trend is particularly
influenced by the strong growth in the Asian market since 2008, where
typically larger aircraft are used.
To compare trends in the major global markets, a regional nomenclature
is utilized: the regional classification is IATA-based and does not align
with economic trading blocs. For example, Europe (EU) is defined as all
areas west of the Ural Mountains, so that Russia is split between Europe
and Asia (AS); North America consists of Canada and the USA but not
Mexico, which is included in Latin America (LA). The major regions are
further subdivided into smaller country groupings (for example, there are
four Latin American and four Asian sub-regions), but these will not be
presented in this chapter.

M4371-FINGER_9781786431851_t.indd 222 14/11/2017 11:14

700
600
NUMBER OF AIRPORTS
500
400
300
200
100
0
1996 1998 2000 2002 2004 2006 2008 2010 2012 2015
AF AS EU LA ME NA SW
Figure 12.1 Number of airports by major region receiving jet air traffic,

1996–2015
Considering traffic shares for the major continental regions over the period
from 1996 to 2015, the North American (NA) region share fell from 46 per
cent of global departure movements in 1996 to 29 per cent in 2015. The
Asian region share rose from 15 per cent in 1996 to 29 per cent in 2015,
while Europe slightly increased its share (22 per cent in 1996 to 24 per cent
in 2015), reflecting slightly higher growth than the average annual rates
cited earlier. The Middle East (ME) region increased its share from 2 per
cent in 1996 to 2.9 per cent in 2015, while Latin America (LA) and the
Southwest (SW) have maintained their shares.
Figure 12.1 shows the number of airports receiving passenger jet ser-
vices in each region over the analysis period. There are dramatic increases
in the number of Asian airports (from 447 to 655) and in the number
of European airports (from 419 to 553). For North America, there was
a significant drop in the number of airports receiving jet air transport
services after deregulation in the US in 1978, and this was sustained until
after 2010; since then there has been a significant increase in the number
of airports receiving scheduled jet services, from 361 in 2010 to 444 in
2015. Smaller increases are observed in the numbers of airports in Latin
America, Africa (AF), the Middle East and Southwest regions.
To examine the distribution of traffic across these continental systems
of airports over the analysis period, two approaches were used. The first

M4371-FINGER_9781786431851_t.indd 223 14/11/2017 11:14

0.88
0.86
GINI INDEX SCORE [0,1]
0.84
0.82
0.8
0.78
0.76
0.74
0.72
1996 1998 2000 2002 2004 2006 2008 2010 2012 2015
AF AS EU LA ME NA SW
Figure 12.2 Gini Index scores for air traffic movements for each major
region, 1996–2015
approach uses the Gini Index, which gives a measure of the extent of con-
centration in a traffic distribution and summarizes the deviations from an
equal share traffic distribution. A score of 0 indicates an equal share across
all airport communities; a score close to 1 indicates a highly-concentrated
distribution focused on a relatively small number of communities. Using
traffic movement and seating capacity shares at each airport across the
system of airports in each year and for each region, the Gini Index was
computed and adjusted to take account of changes in the number of air-
ports receiving services (for a discussion on these adjustment factors see
Reynolds-Feighan, 2007). Figure 12.2 shows the Gini Index scores for the
regional movements traffic distributions over the analysis period.
The Gini scores range between 0.74 and 0.87, which reflects a high degree
of concentration in the traffic flows in all regions. The Gini index scores for
seating capacity are consistently about 2 per cent higher but show a similar
trend to the departure movement traffic trends shown in Figure 12.2,
reflecting the fact that larger aircraft are deployed at the largest airports
and therefore account for higher shares of seating capacity compared
with movements. Traffic concentration declined in the European, North
American and Asian regional markets, while concentration increased in
the Middle East, African, Southwest, and Latin American markets.

M4371-FINGER_9781786431851_t.indd 224 14/11/2017 11:14

To explore these trends further, the second approach involves adapt-

ing the US Federal Aviation Administration (FAA) Hub Classification
scheme2 and categorizing air traffic communities based on the share of
annual air traffic activity. The FAA developed its hub structure in the
1950s as a reporting and funding evaluation mechanism. This approach
groups airports into communities based on the cities and metropolitan
areas that they serve. Most communities are served by a single airport,
but in the case of many large metropolitan areas, they may be served by
multiple airports. To map airports to the cities that they serve, the IATA
location identifier was utilized for each airport. IATA publishes location
identifiers consisting of a three-letter code for a location or airport in
its Airline Coding Directory.3 The city code can be used to map airports
to cities for communities served by multiple airports; only airports and
the carriers utilizing the airport can apply to have the location identifier
changed. Table 12.1 shows the number of cities and number of airports
serving the cities by continental region for 1996 and 2015.
The FAA hub classification examines the traffic in a one-year period.
In the adaptation here, large hubs are identified as those communities
receiving 1 per cent or more of the annual traffic. Medium hubs are those
communities receiving between 0.25 per cent and 1 per cent of annual
traffic; small hubs receive 0.05 per cent to 0.25 per cent of annual traffic.
Non-hubs are those communities receiving less than 0.05 per cent of
annual traffic. The non-hubs in this adaptation are further subdivided into
three categories based on threshold numbers of departure movements,
so that the smallest air transport communities may be distinguished. The
scheme used for annual air traffic movements is set out in Table 12.2. The
classification system is applied to the global air traffic distribution as well
as the major regional markets and relates to the non-stop departure move-
ments data series. Table 12.3 shows the numbers of each hub type for each
region in 2010 and 2015. The number of large hubs in the global scheme as
well as in each region has reduced over time, reflecting a less concentrated
distribution of traffic. The number of small hubs and non-hubs of type ‘A’
have seen the biggest increases, and their traffic shares have also increased
slightly from 15 per cent collectively in 2010 to 17 per cent in 2015.
The number of communities being served by multiple airports has been
growing over time. In 2015, 70 of the 2378 global air transport communi-
ties had two airports, 11 had three airports, and five cities had four or
more airports (these were London (6), Paris (5), Milan (4), Stockholm
(4) and New York (4)).4 The large hubs serve the world’s largest cities
and metropolitan areas and are key connection nodes, both globally and
within. There are consistently about 20 to 25 large hubs in each regional
scheme handling well over half of the traffic for their region. There has

M4371-FINGER_9781786431851_t.indd 225 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 226
Table 12.1 Number of cities and airports serving the cities by continental region
No. 1996 – Region 2015 – Region

Airports
AF AS EU LA ME NA SW 1996 AF AS EU LA ME NA SW 2015
per city
Total Total
1 238 439 361 313 73 316 94 1834 258 632 489 339 101 371 102 2292
226
2 4 16 5 2 6 33 3 10 18 6 5 27 1 70
3 1 2 3 1 3 1 5 1 11
4 1 1 2 1 3
5 1 1 1 1
6 1 1
Total 238 443 380 318 75 324 94 1872 261 643 514 346 106 404 104 2378

14/11/2017 11:14
Table 12.2 Hub classification scheme used to categorize air transport

communities
Traffic Criterion Hub Type

1% or more of annual departure movements Large
At least 0.25%, but less than 1% of annual departure Medium
movements
At least 0.05%, but less than 0.25% of annual departure Small
movements
More than 500 but less than 0.05% of annual departure Non-Hub: Type A
movements
Between 10 and 499 annual departure movements Non-Hub: Type B
Less than 10 annual departure movements Non-Hub: Type C
Table 12.3 Number of each type of hub by region, 2010 and 2015
Region Year Large Medium Small Non- Non- Non- Total

hub ‘A’ hub ‘B’ hub ‘C’
AF 2010 22 44 61 0 87 7 221
AS 2010 23 54 142 206 131 5 561
EU 2010 25 54 117 154 100 17 467
LA 2010 23 51 118 41 77 4 314
ME 2010 18 18 36 0 34 0 106
NA 2010 24 49 95 117 46 14 345
SW 2010 15 18 22 0 32 1 88
All 2010 150 288 591 518 507 48 2102
AF 2015 20 51 77 0 97 16 261
AS 2015 22 58 156 270 124 13 643
EU 2015 23 48 132 172 116 23 514
LA 2015 21 50 131 52 82 10 346
ME 2015 18 17 28 3 36 4 106
NA 2015 23 42 90 169 63 17 404
SW 2015 15 20 27 0 36 6 104
All 2015 142 286 641 666 554 89 2378
been an increasing average aircraft size at all classes of hubs and non-hubs
in all regions. Because of congestion at several of the large hubs, some
have reduced the numbers of short-haul or smaller aircraft services and
focused on the development of longer-haul traffic and facilitating larger
equipment. The medium and small hubs play an important role in provid-

M4371-FINGER_9781786431851_t.indd 227 14/11/2017 11:14

ing connectivity for smaller communities and in connecting urban centres

within the continents. Also as the hub size gets smaller, so does the average
aircraft size. The significantly smaller average aircraft size in the North
American market at all hub classes may also be noted, though this has
increased since 2008.
It would be misleading to think of the large hubs as serving predomi-
nantly long-haul and international traffic. Table 12.4 shows the three
busiest city pairs in each of the major regions in 1996 and 2015. Many
of the routes are short haul and have route lengths of less than 500 km.
When the busiest airport pairs were identified, the average distance for the
top three routes in each region fell, particularly in the 1996 period. For the
Southwest region, the three busiest city pairs accounted for 19 per cent of
the region’s traffic in 2015, just slightly lower than in 1996.
Population Characteristics
The air traffic communities were linked to urban population data gathered
from the UN Population Division reports, the World Bank and from the
Tableau databanks.5 For smaller communities, internet searches were con-
ducted to find up-to-date urban population information.6 The average pop-
ulation of each category of hub is presented by region in Table 12.5 for 2015.
The large hubs generally serve large cities with populations exceeding 1.5
million. For the Asian and Latin American markets, the average large hub
city size is significantly greater than in other regions. In Asia, the medium
hubs are also very large urban centres, with an average population of 2.5
million. The relatively low propensity to travel by air in Asian markets is
associated with lower incomes, low rates of private consumption and a small
middle-class population. The leading industry forecasts produced by Airbus
and Boeing point to changes in these factors as the key drivers behind their
forecast annual growth rates of between 4.5 per cent and 4.8 per cent until
2035 (Airbus Industries, 2016; Boeing, 2016), with strong growth in private
consumption in Asia, particularly in the Chinese domestic market expected
over the forecast period. For Africa, it is anticipated that there will be 22
cities with populations of at least 4 million by 2025 (Airbus Industries, 2016).
While intraregional traffic has increased significantly in the last five years,
the propensity to travel by air is well below rates in other global regions. For
the North American market, the smaller average size of the hub populations
is an indicator of the high propensity to travel and higher incomes in this
region. The small hubs and non-hub communities have substantial popula-
tions, particularly in Asia, Latin America, the Middle East and Africa. By
contrast, the North American and Southwest non-hubs have populations of
under 100 000, and less than 30 000 in the case of the Southwest.

M4371-FINGER_9781786431851_t.indd 228 14/11/2017 11:14

Table 12.4 Busiest city pairs in each region in 1996 and 2015
1996
Region City Pair No. City Pair Distance Percentage of Regional Number of Carriers
Airports (km) Traffic
M4371-FINGER_9781786431851_t.indd 229
AF Johannesburg Cape Town 2 1271 6.3 14
Durban Johannesburg 2 501 4.9 12
Benghazi Tripoli 2 669 1.6 1
AS Kaohsiung Taipei 3 301 2.7 8
Busan Seoul 2 336 1.2 4
Seoul Jeju 2 451 1.1 2
EU Dublin London 6 464 1.1 9
229
Rome Milan 5 498 1.0 13
Madrid Barcelona 2 483 1.0 11
LA Rio de Janeiro Sao Paulo 4 366 3.4 27
Guadalajara Mexico City 2 459 1.8 5
Mexico City Monterrey 2 713 1.3 6
ME Riyadh Jeddah 2 853 3.3 3
Dharan Riyadh 2 373 3.2 4
Bahrain Doha 2 145 1.8 10
NA New York Chicago 6 1166 0.8 17
Kahului Honolulu 2 163 0.7 6
Houston Dallas/Fort Worth 4 368 0.7 7
SW Sydney Melbourne 2 703 9.4 21
Sydney Brisbane 2 740 6.5 10
Wellington Auckland 2 480 4.4 2

14/11/2017 11:14
Table 12.4 (continued)
2015
Region City Pair No. City Pair Distance Percentage of Regional Number of Carriers
M4371-FINGER_9781786431851_t.indd 230
Airports (km) Traffic
AF Johannesburg Cape Town 3 1271 5.1 12
Durban Johannesburg 3 501 3.0 9
Cairo Jeddah 2 1217 3.0 9
AS Jeju Seoul 3 450 0.9 8
Sapporo Tokyo 4 801 0.6 9
Tokyo Fukuoka 4 909 0.6 8
EU Istanbul Izmir 3 336 0.5 9
230
Dublin London 7 463 0.5 9
London Amsterdam 7 341 0.5 11
LA Sao Paulo Rio de Janeiro 5 366 3.7 19
Brasilia Sao Paulo 4 841 1.5 10
Belo Horizonte Sao Paulo 5 505 1.4 7
ME Riyadh Jeddah 2 850 2.4 6
Dubai Doha 3 370 2.3 5
Kuwait Dubai 3 853 1.6 7
NA New York Chicago 12 1175 0.7 16
Boston New York 11 302 0.5 10
San Francisco Los Angeles 2 544 0.5 5
SW Sydney Melbourne 4 712 9.4 7
Brisbane Sydney 4 750 5.5 4

Brisbane Melbourne 2 1379 4.2 4
14/11/2017 11:14
M4371-FINGER_9781786431851_t.indd 231
Table 12.5 Average population of air transport communities for major regions and hub types in 2015
Region Large Medium Small Non-hub ‘A’ Non-hub ‘B’ Non-hub ‘C’ Average for
Region
AF 2 576 220 1 011 142 372 920 209 307 70 433 587 116
AS 7 749 732 2 681 584 85 734 439 460 314 056 124 408 962 677
231
EU 2 358 831 813 727 382 665 219 958 136 902 105 486 389 538
LA 3 921 544 849 151 427 915 183 557 194 455 64 580 598 274
ME 1 904 106 828 383 325 665 161 303 234 413 173 228 632 931
NA 1 333 028 461 868 158 670 68 781 48 144 90 624 199 773
SW 1 150 177 75 945 24 220 29 864 23 335 198 467
Average per hub type 3 104 298 1 131 526 456 197 267 258 187 128 92 019 566 948

14/11/2017 11:14
Airlines and Market Structures
Over the period 1996 to 2015, the North American, Latin American and
Southwest regions saw very little change in the number of carriers, includ-
ing those externally registered, serving the markets. There was an increase
in North America during the 2000s, but numbers have fallen again since
2008. The European market has seen the most dramatic change over the
period, with a rapid and substantial increase in the number of carriers
between 1996 and 2004/05. Since then, numbers have fallen equally dramat-
ically due to consolidation, code-sharing arrangements through alliance
partners and withdrawal by some international carriers. The Asian, Middle
East and African markets have seen increases in the number of carriers
since the mid-2000s, though numbers stabilized in the most recent period.
The numbers of city pairs departing from each region are shown in
Figure 12.3. Europe has by far the most extensive network of city-pair
routes and this has continued to increase over time. Many of the routes
have a low frequency of service, but facilitate very high levels of connectiv-
ity and accessibility compared to other regions. It can be noted that the
Asian route network is expanding significantly in the most recent period.
The North American market has a relatively small number of city pairs
that typically have high frequency of service. The route networks in the
African and Southwest markets are relatively small by contrast.
In terms of the average number of carriers operating on city-pair routes
for the different hub classes by region in 2010 and 2015, the large hubs in
18 000 50
Thousands
45
Number of city pairs per region
16 000
Total global city pair markets
14 000 40
35
12 000
30
10 000
25
8000
20
6000
15
4000 10
2000 5
0 0
1996 2000 2006 2010 2015
SW ME AF LA
NA AS EU Worldwide
Figure 12.3 City pairs served by region, and world city pair routes,
1996–2015

M4371-FINGER_9781786431851_t.indd 232 14/11/2017 11:14

Table 12.6 Percentage of single-carrier city pairs by region for selected

years
Region Year
1996 2000 2006 2010 2015
AF 71% 67% 63% 63% 67%
AS 58% 59% 55% 54% 56%
EU 61% 60% 62% 63% 61%
LA 58% 54% 59% 57% 55%
ME 64% 66% 64% 62% 62%
NA 62% 65% 66% 60% 60%
SW 55% 59% 63% 60% 62%
Worldwide 61% 61% 62% 60% 59%
all regions typically have two carriers operating on routes departing from
these communities. Asian markets have more carriers per route compared
with other regions, while the North American, Latin American, Middle
East and African markets have between 1.8 and 1.9 carriers per route from
their large hubs. As the community size gets smaller, the number of carriers
per route also declines, with most non-hub routes having just one carrier.
This trend is observed in all regions. The average number of carriers per
route has increased marginally over the 2009–15 period for medium and
small hubs and for non-hub types A and B communities.
The extent of competition on city-pair routes is further explored in
Table 12.6, where the percentage of all routes with just a single carrier pro-
viding service is presented, while Table 12.7 shows the percentage of depar-
ture movements on single carrier and two-carrier routes for 1996, 2000,
2006, 2010 and 2015. While most routes in every region are single-carrier
routes, these routes account for just below 20 per cent of all traffic world-
wide in 2015. The African market has the largest share of single-carrier
routes at 67 per cent in 2015, and this accounted for 28 per cent of move-
ments. The number of single-carrier routes has declined over time, as the
network of city pairs has expanded in every region. New routes tend to be
single-carrier routes initially, but as traffic expands, the number of carriers
increases. Air transport markets are generally becoming more competitive,
particularly those routes operating from large hubs.
The simple average stage length for city-pair routes from each type of
hub and non-hub was computed for each period, and gives an indication of
the range of services operated from each hub type and is not weighted by
the share of movements. Ignoring the small number of type-‘C’ non-hubs
in each region, as the hub size increases, the average route stage length

M4371-FINGER_9781786431851_t.indd 233 14/11/2017 11:14

Table 12.7 Percentage of departure movements on single- and two-carrier

routes for selected years, 1996–2015
Region 1996 2000 2006 2010 2015

Number of Carriers per Route
1 2 1 2 1 2 1 2 1 2
AF 38% 21% 34% 25% 28% 22% 24% 21% 28% 27%
AS 21% 23% 22% 28% 15% 24% 13% 16% 13% 13%
EU 21% 26% 29% 31% 32% 32% 22% 22% 20% 21%
LA 18% 19% 15% 19% 19% 25% 16% 23% 14% 21%
ME 40% 19% 34% 25% 32% 24% 24% 25% 25% 25%
NA 26% 24% 38% 33% 41% 32% 28% 32% 24% 23%
SW 12% 29% 12% 32% 17% 33% 13% 17% 12% 28%
All Regions 23% 24% 31% 30% 31% 30% 21% 24% 19% 20%
increases.7 The Southwest region has the highest average stage length, as
many of the long-haul services operated from the large hubs in Australia
and New Zealand connect to Europe, North America and Asia. By con-
trast, in other regions there are high numbers of short- and medium-haul
routes operating from the large hubs. The small hub and non-hub commu-
nities in most regions have shorter average stage lengths.
The growth of traffic movements was examined across hub classes and
average changes are presented for the 2009 to 2015 period in Table 12.8.
The number of communities losing air service and the number of new
Table 12.8 Average percentage change in departure movements from 2009
to 2015, and number of non-hub communities losing service
and receiving new service between 2009 and 2015
Region Large Medium Small Non- Non- Number of Number of

hub ‘A’ hub ‘B’ Non-hubs New Non-hub
Losing Communities
Service Receiving Jet
Services
AF 24.2 17.4 36.1 −61.4 20 47
AS 36.1 44.4 46.5 54.8 −8.5 21 120
EU 22.9 18.7 21.9 29.8 −74.5 27 82
LA 27.9 30.9 32.2 43.4 −11.8 15 54
ME 36.9 50.3 3.7 2.4 −197.6 12 15
NA 9.8 −7.6 −2.9 27.0 −91.7 12 69
SW 12.9 18.5 30.6 32.3 4 15
All regions 24.5 23.8 27.8 40.2 −51.2 111 402

M4371-FINGER_9781786431851_t.indd 234 14/11/2017 11:14

communities receiving service are recorded in the table. The large hubs
experienced strong growth in all regions over the period, and averaged
24.5 per cent increase in movements, with the North American market
experiencing the lowest growth. The Asian, Latin American and Middle
East regions had the highest growth. Growth was even stronger in these
regions at the medium and small hubs. The non-hubs in all regions had a
more varied experience. For European, Latin American and Asian non-
hubs type A, very strong traffic growth was recorded and many new com-
munities began receiving air services over the period (402 in total). Over
the period 111 non-hub communities lost jet air services, and these were
distributed across all regions.
SOCIAL AIR SERVICE PROVISION
In examining the trends in global air transport, a number of observations

may be made regarding the experience of small communities. The small
communities in all regions have small volumes of air traffic, and the service
provided uses smaller aircraft with typically just one carrier operating on
routes. The typical stage length is around 1250 km. These small communi-
ties are more vulnerable to significant traffic changes associated with the
economic cycle. The populations of the small and medium-sized air trans-
port communities are generally substantial and range between 100 000
and 250 000 outside of the North American and Southwest markets. With
deregulation in many domestic markets and liberalization of cross-border
air routes in all global regions, governments have recognized the vulner-
ability of smaller communities to traffic volatility, as carriers make com-
mercial decisions on the services to provide.
Social air services are air transport services identified and mandated
by regional or national governments, and are deemed to be essential
for reasons of social or economic development. Where carriers will not
provide an air service on a commercial basis, governments may identify
a requirement for an air service. Such services are being established in an
increasing number of jurisdictions worldwide. Social air service provision
typically involves the government offering exclusive concession and – if
necessary – financial support to an airline to provide air services to remote
or economically disadvantaged regions or communities. Legislation in
most cases sets out the basis for identifying communities, selecting airlines
and criteria for the support. The driving principles of transparency and
openness characterize social air service processes.
ICAO and WTO collaborated in developing a framework for the
establishment of social air services in domestic as well as in international

M4371-FINGER_9781786431851_t.indd 235 14/11/2017 11:14

markets where such services may help to drive the development of new
tourism products in the Least Developed Countries (LDCs) (ICAO, 2005).
The criteria set out in the framework are based on analysis of social air
service provision in developed-country markets. In their study, ICAO and
WTO set out a template for the design of social air service schemes under
the following criteria:
●● Route selection: identification of socioeconomic objectives and eco-

nomic justification/assessment;
●● Service level specification: determination of minimum service
standards;
●● Carrier selection: carrier eligibility, and competitive tendering
process and selection criteria;
●● Contract duration: review process, monitoring, audit and enforce-
ment considerations;
●● Subsidy payment: assessment of scheme costs and other considera-
tions (for example, efficiency of service provision);
●● Sources of financing;
●● Supplementary options: indirect subsidies, alternative incentives and
consideration of distortionary market effects.
In reviewing the provision of social air services in different global

regions, these criteria are focused on and the scale of the programmes is
indicated.
United States
The Airline Deregulation Act of October 1978 contained a provision

for social air services under the Essential Air Services Program, which
guaranteed – initially for a decade – continuity of service for small
communities that had been receiving air services. The programme was
initially funded for ten years, but has been extended and reorganized
on several occasions, most recently under the 2012 FAA Modernization
and Reform Act and the 2015 Consolidated and Further Appropriations
Act (Public Law No. 11 3–235). The programme mandates the US
Department of Transportation to provide qualifying communities with
access to the national air transport system, typically by subsidizing two
round trips per day with 30- to 50-seater aircraft to medium or large
hub airports (that is, airports serving at least 0.25 per cent of annual
passenger enplanements). The eligibility requirements for communities
have changed since 1978 and set out minimum distances for eligible com-
munities from larger commercial airports, as well as minimal passenger

M4371-FINGER_9781786431851_t.indd 236 14/11/2017 11:14

volumes on existing subsidized services (for a more detailed review of

the programme’s history see Tang, 2015). In 2016, 175 communities were
covered by the scheme, 60 of which are in the northern state of Alaska. A
maximum subsidy cap of $200 per passenger is imposed. The appropria-
tions for the programme have risen steadily from $68.9 million in 1979 to
$261 million in 2015. The scheme is funded from overflight fees paid by
non-US carriers using US airspace.
A second scheme, the Small Community Air Service Development
Program (SCASDP) was established in 2000 to promote new or enhanced
community air service initiatives, or to address higher-than-average air-
fares. This is a grant programme that supports small communities (air
transport communities receiving less than 0.05 per cent of annual passen-
ger enplanements) developing or maintaining air services through revenue
guarantees, grants for marketing, start-up expenses and research studies.
Priority is given to communities where airfares are higher than average,
where public–private partnerships have been established to facilitate air
service provision, where enhanced services are expected to bring benefits
to a wide range of users or where multiple communities can cooperate to
source a consolidated air service at a single airport. Annual allocations
vary ($7 million in 2014; $5.5 million in 2016), with typical grants of
$500 000 per community.
Canada
Canada does not operate an essential air service programme, but funds 13
remote airports through the National Airports Policy. Since the mid-1990s,
the federal government has moved to transfer ownership and operation
of regional and local airports to locally based authorities who take on the
responsibility for funding their maintenance (see Metrass-Mendes et al.,
2011 for a more detailed description of the process). At the same time,
the nationally funded Airports Capital Assistance Program funds projects
aimed at protecting the airport assets at about 200 regional and local
airports. A consultant report in 2015 to the Canadian Assembly recom-
mended against introducing an essential air service process (and thereby
subsidizing air carrier operations) in Northern Canada (RP Erickson &
Associates, 2015).
Australia
The Remote Air Service Subsidy Scheme (RASS) was introduced in 1983,
though the Australian government had been subsidizing air transport ser-
vices to remote regions since 1957. It is part of the federal government’s

M4371-FINGER_9781786431851_t.indd 237 14/11/2017 11:14

Regional Aviation Access Programme (RAAP). The RASS aims to ensure

access to scheduled air transport services for remote and isolated areas.
Communities apply to be included in the scheme, and through a com-
petitive tendering process carriers are selected to meet the various service
requirements, typically for a period of two years (though it can be for up
to four years). In 2015 the scheme covered 366 communities with popula-
tions of up to 200 persons and seven airlines providing the air services. The
budget allocation was AU$56.8 million in the 2015 to 2019 period,8 which
included grants for maintenance and upgrades of airstrips. The scheme
is funded from en-route air navigation charges levied by Airservices
Australia.
European Union
Europe’s air transport market was gradually liberalized between 1993

and 1997 under the so-called ‘Third Package’ of air transport liberaliza-
tion measures which came into effect in 1993 (Council Regulation No.
2408/92). As part of this package, Public Service Obligation (PSO) air
routes were permitted when allocated under competitive tendering pro-
cedures detailed in the regulation and were revised in 2008 (Council
Regulation (EC) 1008/2008, Articles 16–18). States nominate eligible
routes to communities that are peripheral or where such routes are neces-
sary for reasons of regional economic development and expect to have less
than 10 000 passengers per year. Each member state administers its own
scheme, but is subject to the terms of the European regulations. The 2008
regulation gave the European Commission legislative force to investigate
any tender competition or evaluate the basis for the imposition of a PSO.
States are required to consult with other states and publicly advertise
tender competitions through the Official Journal of the European Union.9
The states may limit access to the PSO routes and if necessary subsidize
the air service. States are required to submit detailed information on the
tender competition outcome and on the selection of carriers and service
arrangements. There were 237 PSO routes in operation in 13 EU member
states in December 2015.10
In addition to the PSOs, the European Union permits member state
governments to support airports and air carriers in line with EU state aid
guidelines,11 and there are three schemes through which state aid may be
allowed. These are: (1) state aid for investment in airport infrastructure,
which is permitted if there is ‘a genuine transport need and the public
support is necessary to ensure the accessibility of a region’; (2) operating
aid to regional airports with less than 3 million passengers per year, for up
to ten years to facilitate airports adjusting their business models towards

M4371-FINGER_9781786431851_t.indd 238 14/11/2017 11:14

fully commercial operations; and (3) start-up aid to airlines to launch new
air routes with the aim of increasing the connectivity of a region. These
guidelines regularize and update the 2005 guidelines introduced to harmo-
nize the public financing of airports and new route development funding.
Member states can also devise their own initiatives, as long as they
comply with EU regulations. For example, the UK government intro-
duced its Regional Air Connectivity Fund12 in November 2014 with a
view to expand UK regional airport routes within the UK and Europe,
while avoiding linkages to the large London airports because of capacity
constraints. These new routes are expected to be commercially viable after
three years, and 15 routes were initially selected.
India
Under the 1992 Route Dispersal Guidelines (RDG), the Indian Ministry
of Civil Aviation set out requirements for carriers to distribute their capac-
ity across three different categories of routes and thus cross-subsidize
air services to small and more remote communities. The 2016 National
Civil Aviation Policy rationalized the RDG policy and categorization of
routes, and will take effect from 2017. The route categories at present are:
(1) 12 high-density (Category I) routes (heavily trafficked routes between
the major urban centres of Mumbai, Kolkata, Hyderabad, Bangalore,
Trivandrum and Chennai); (2) routes to more remote parts of the country
in the Northeast, Jammu and Kashmir and island territories (Category
II routes); (3) routes within Northeast India and Jammu and Kashmir
(Category IIA); and (4) all other routes (Category III). The guidelines
require that 10 per cent of a carrier’s Category I capacity be deployed
on Category II routes, 1 per cent on Category IIA and 35 per cent on
Category III routes.13 This policy has forced carriers to connect small and
remote communities within their regions and to the main urban centres.
Latin America and Brazil
Airport privatization has taken place in several South American countries

during the 1990s and 2000s. Airport concessions are the approach adopted
in Colombia (1993), Mexico (1995), Chile (1997), Bolivia, Costa Rica,
Peru, Venezuela, Argentina (1997) and the Dominican Republic (Lipovich,
2008). In Latin America, military-operated airlines traditionally served
remote communities. As these airlines have been phased out or became
commercial operations, retaining the regional air services has become an
issue for government.
Brazil has experienced significant growth in air transport activity in the

M4371-FINGER_9781786431851_t.indd 239 14/11/2017 11:14

last ten years. In 2012 the Brazilian government set out plans to build or
adapt 70 airports for commercial use, as part of a strategic infrastructure
programme (Logistics Investment Program PIL). The plan envisaged an
investment of US$2 billion to develop a network of airports that would
serve remote regions (with the support of government funding) as well
as larger cities and tourist destinations (where private investors and con-
cessionaires would be sought). The plan aims to have 96 per cent of the
Brazilian population living within 100 km of an airport. In 2015, the
plan was renewed and extended to fund the development of regional hubs
through the concession of four state airports. An additional six state air-
ports were expected to be concessioned to the private sector in 2016 with
either 20- or 30-year terms (USTDA, 2015). Couto et al. (2015) described
the national network structure characteristics and identified regional
sub-networks with relatively low traffic volumes and connectivity. In
January 2015 the government enacted legislation to set up the Program
of Development of Regional Aviation (Programa de Desenvolvimento
da Aviação Regional – PDAR). This act relates to airports with less than
600 000 annual passenger throughput (embarking and disembarking),
which are designated as regional airports for the purposes of the pro-
gramme, with the exception of airports located in the Amazon, where the
passenger threshold is 800 000. There were 689 local and regional airports
in Brazil in 2016. The aims of the PDAR are to:
I – increase access of the population to the air transport system, with priority
to those living in less developed regions of the country, considering both the
increase in the number of municipalities and routes served by scheduled air
transport;
II – integrate isolated communities to the national civil aviation network to
facilitate the mobility of its citizens; and
III – facilitate access to areas with tourist potential, subject to the provisions of
section I. (Câmara Dos Deputados, 2015)
The act authorizes the payment of a subsidy to qualifying carriers of

up to 50 per cent of the capacity on direct domestic flights to regional
airports, to a maximum of 60 seats. Airport landing fees may also be
waived. The service contracts are granted for a five-year period, with
scope for extension to a second period of the same length. The subsidies
are financed through the National Civil Aviation Fund, with an estimated
maximum budget allocation ceiling of US$320 million (30 per cent of the
NCAF). The implementation of the PDAR programme was deferred in
mid-2015 because of budgetary constraints.14

M4371-FINGER_9781786431851_t.indd 240 14/11/2017 11:14

CONCLUSIONS
Air transport activity has grown steadily in all global regions over the last
two decades, and is expected to continue this growth trend as Asian, Latin
American and African regions develop economically. The urbanization of
the world’s population is leading to the spatial concentration of people
and of economies in relatively small spaces, and these act as anchor points
for the global air transport system. Changes in domestic and international
regulations governing air transport activities have led to a more market-
oriented industry and approach to providing air services.
The number of air transport communities has expanded substantially,
with the airports at the world’s largest population centres handling the
majority of air transport activity. Medium and small communities are
enjoying increasing levels of service with jet aircraft that enable them to
connect to national and international centres. The size of communities
with regular jet air services is still substantial in most regions, so that
many communities with populations under 100 000 do not have services.
For these communities and for much smaller and more remote locations,
social air service policies have been devised to enable access to national
air transport systems, and it can be expected that these kinds of policies
will be deployed in developing countries over the next couple of decades.
Based on experiences to date, these programmes can provide cost-effective
means of enabling accessibility for small communities, many of which may
become commercial over time. The approach to social air service provi-
sion makes explicit the need for cross-subsidization and for government
intervention to identify and provide transportation services in certain
circumstances.
NOTES
1. Steer Davies Gleave (2016).

2. FAA Hub classification system
1% or More Large
At least 0.25%, but less than 1% Medium
At least 0.05%, but less than 0.25% Small
More than 10 000, but less than 0.05% Non-Hub Primary
At least 2500 and no more than 10 000 Non-Hub Non-Primary
Primary Airports are Commercial Service Airports that have more than 10 000 passen-
ger boardings each year. Hub categories for Primary Airports are defined as a percent-
age of passenger boardings within the United States in the most current calendar year
ending before the start of the current fiscal year.

M4371-FINGER_9781786431851_t.indd 241 14/11/2017 11:14

Non-primary Commercial Service Airports are Commercial Service Airports that

have at least 2500 and no more than 10 000 passenger boardings each year.
3. The last printed edition is the IATA Airline Coding Directory 2012; the ACD is now
licensed electronically on a monthly subscription basis.
4. Moscow had four airports in 2010, but three in 2015.
5. UN city population data is available at https://esa.un.org/unpd/wup/. Tableau is a busi-
ness intelligence service offering data analytics capabilities and is available at http://
www.tableau.com/. The World Bank data sets are available at http://data.worldbank.
org/.
6. Data for the populations of many small island communities was not available in the UN
and World Bank databases.
7. The non-hub type C communities have fewer than ten movements per year. The com-
munities in this category change significantly from year to year, as operating carriers
can add or cut services and change the classification. Several of these communities are
remote islands with ad hoc service, but are located at substantial distances from the
mainland.
8. Press Statement by Warren Truss MP http://minister.infrastructure.gov.au/wt/releases/
2015/May/wt133_2015.aspx.
9. The Official Journal of the European Union is an online daily gazette record for the
European Union and includes invitations to tender, information notices, as well as regu-
lations, directives, decisions, recommendations and opinions from the EU institutions.
10. The European Commission Transport Directorate maintains a listing of current PSO
contracts at http://ec.europa.eu/transport/modes/air/internal-market/public-service-obli
gations-psos_en.
11. See Communication from the Commission, ‘Guidelines on State aid to airports and
airlines (2014/C 99/03)’, February 2014.
12. See the UK Department for Transport (2014).
13. Prior to 2016, carriers were required to deploy 50 per cent of their Category I capacity
on Category III routes.
14. www.ch-Aviation.com, June 2015.
REFERENCES
Airbus Industries (2016), ‘Global market forecast: mapping demand: 2016–2035’,

Airbus SAS, 2016. Accessed December 2016 at http://www.airbus.com/company/
market/global-market-forecast-2016–2035/.
Boeing (2016), ‘Current market outlook’, Boeing, 2016. Accessed December 2016
at http://www.boeing.com/resources/boeingdotcom/commercial/about-our-mar
ket/assets/downloads/cmo_print_2016_final_updated.pdf.
Câmara Dos Deputados (2015), Law No. 13.097, Chapter VIII, Art. 116, 19
January, Government of Brazil.
Couto, G.S., A.P. Couto da Silva, L.B. Ruiz and F. Benevenuto (2015), ‘Structural
properties of the Brazilian Air Transportation Network’, Anais da Academia
Brasileira de Ciências (Annals of the Brazilian Academy of Sciences), 87(3) 1653–74.
Decurtins, C.G. (2007), ‘The air transport review at the WTO: bilateralism versus
multilateralism’, PhD Thesis, Université De Genève Institut Universitaire de
Hautes Études Internationales, Geneva.
ICAO (2005), ‘A study of an essential service and tourism development route
scheme’, International civil Aviation Organisation, May. Accessed October 2016 at
http://www.icao.int/sustainability/EssentialServicesStudy/EssentialServicesStudy_
en.pdf.

M4371-FINGER_9781786431851_t.indd 242 14/11/2017 11:14

Lipovich, G.A. (2008), ‘The privatization of Argentine airports’, Journal of Air

Transport Management, 14(1), 8–15.
Metrass-Mendes, A., R. de Neufville and A. Costa (2011), ‘Air accessibility in
northern Canada: prospects and lessons for remote communities’, paper pre-
sented at Cranfield University. Accessed October 2016 at http://ardent.mit.edu/
airports/ASP_papers/Alda%20Canada_AMM_RDN_AC_V02.pdf.
Reynolds-Feighan, A.J. (2007), ‘Competing networks, spatial and industrial con-
centration in the US airline industry’, Spatial Economic Analysis, 2(3), 237–57.
R.P. Erickson & Associates (2015), ‘Comparison of approaches for supporting,
protecting & encouraging remote air services’, report prepared for CTA Review
Secretariat Transport Canada, RP Erickson & Associates, Calgary.
Steer Davies Gleave (2016), ‘Study on airport ownership and management and the
ground handling market in selected non-EU countries’, Final Report June 2016,
European Commission DG MOVE (ref MOVE/E1/SER/201 5–24 7–3).
Tang, R. (2015), ‘Essential Air Service (EAS)’, Congressional Research Service
paper R44176, September 2015. Accessed October 2016 at https://www.fas.org/
sgp/crs/misc/R44176.pdf.
UK Department for Transport (2014), ‘Appraisal framework for UK startup aid
for airports with fewer than 3 million passengers per annum’, Department for
Transport, London. Accessed October 2016 at https://www.gov.uk/government/
uploads/system/uploads/attachment_data/file/396896/start-up-aid-appraisal-
framework-2a.pdf.
US Trade & Development Agency (2015), ‘Brazil’s priority transportation projects –
a resource guide for US industry: aviation’, accessed October 2016 at https://
www.ustda.gov/sites/default/files/pdf/program/regions/lac/brazilresourceguide/
Brazil Resource Guide – 1 – AVIATION Section.pdf.

M4371-FINGER_9781786431851_t.indd 243 14/11/2017 11:14

13. Oligopolization of markets
Sveinn Vidar Gudmundsson
We in the airlines make up the most inefficient oligopoly in the world.

Robert Crandall (1976)1
INTRODUCTION
Although oligopoly is often associated with the airline industry, the indus-
try is characterized by many firms, different business models and a strong
international dimension. For this reason, any analysis must answer the
question of whether the industry has strong or weak oligopolistic charac-
teristics, and if these characteristics differ between domestic and interna-
tional markets and influenced by strategic groups.
Since deregulation of the US airlines in 1978 there have been several
waves of mergers in the US airline industry, usually followed by concerns
about intensified oligopolization in the market. Underpinning these con-
cerns is the common definition of an oligopoly. Practitioners have sug-
gested a rule of thumb that says if four or fewer companies have 50 per
cent or more of a market then there is effectively an oligopoly. Using this
benchmark for the US domestic market, the top four airlines had 60.2
per cent passenger market share in 2015 (Southwest Airlines 20.4 per
cent, Delta Air Lines 16.5 per cent, American Airlines 13.4 per cent and
United Airlines 9.9 per cent), telling us that indeed the industry has the
characteristics of an oligopoly. However, in the US international market
the story is altogether different, with the top four airlines having only 39.9
per cent market share (United Airlines 12.7 per cent, American Airlines
12.4 per cent, Delta Air Lines 11.3 per cent, and British Airways 3.5 per
cent).
Nonetheless, simple rules of thumb and even more sophisticated con-
centration measures can be misleading as to the true state of competition
intensity in an industry. Different strategic groups in an industry may
behave differently within the group and across the groups. In this context
it is important to point out that the leader in the US domestic market
is a low-cost airline that has grown slowly but surely since deregulation
(Gudmundsson, 1998), while the other three are legacy carriers that have
maintained their position mostly through mergers and acquisitions.
244

M4371-FINGER_9781786431851_t.indd 244 14/11/2017 11:14

Oligopolization of markets 245
Therefore, the question is: would the competition and interdependencies

be the same within the top four airlines in the presence or absence of an
airline belonging to a different strategic group? In other words, Southwest
is a clear example of a carrier with a sustainable competitive advantage,
growing profitably over a long period of time, with only a few minor
acquisitions (measured in terms of proportional size) over the last four
decades (Morris Air, Muse Air and AirTran Airways). In contrast, Delta,
American and United, the other top airlines in the US domestic market,
have all filed bankruptcies over the same period, and have all grown
through major mergers as well as minor acquisitions. That is, the history
of the top four airlines on how they got there is very different. We can
therefore ask another question: will concentration in the domestic market
continue to increase as Southwest Airlines continues its growth pattern
of past years, and will the three legacy carriers continue to maintain their
position through mergers and acquisitions?
As is commonly known, oligopoly can arise due to mergers and acquisi-
tions, entry barriers, economies of scale, density and scope, and sustainable
competitive advantage. Observing both the US domestic market and the
European market, it is clearly evident that Southwest Airlines, easyJet and
Ryanair have raised barriers to entry at the route level. All of these airlines
have increased their market share greatly, not only by winning passengers
from the legacy airlines, but also by creating new customers that would
not have flown in the absence of the service. What is vital to point out is
that these airlines have demonstrated sustainable competitive advantage
evidenced by their long-term profitability and steady increase in market
share. In other words, the composition of oligopolistic firms and their
competition dynamics appear ever more complex and not subject to simple
rules of thumb.
In the analysis section of this chapter, oligopoly in domestic versus
international markets will be treated separately to point out an important
difference that shapes to some extent the competitive behaviour of the
dominant firms and strategic groups. Another contribution in this chapter
is how concentration will be quantified. Market share in the airline indus-
try is usually measured based on available seat miles (output) or turnover
(sales), both of which are appropriate when firms belong to the same
strategic group. However, using these measures and combining different
business models such as low-cost and legacy airlines into the same group
will enlarge the share of legacy airlines that attract higher-yield passengers
over longer average distances, while low-cost airlines gather to lower-yield
passengers over shorter average distances. For these reasons, oligopoly
measures in this chapter are based on the passenger market shares.
The remainder of this chapter is organized as follows. Following this

M4371-FINGER_9781786431851_t.indd 245 14/11/2017 11:14

introduction, strategic groups are introduced; in the third section oligopo-

listic interdependence is discussed; in the fourth section industry entry and
exit are covered; in the fifth section mergers and acquisitions are discussed;
in the sixth section pricing implications are introduced; in the seventh
section different approaches and limitations of measuring concentration
are introduced; in the eighth section the industry concentration level in the
US airline industry is assessed; and the final section is the conclusion of
the chapter.
STRATEGIC GROUPS
Unlike aircraft manufacturing, the airline industry has comparably low

barriers to entry, more firms in the industry, and therefore greater chances
of new entrant firms rising through the ranks. Good examples would be
Southwest Airlines (est. 1967), JetBlue (est. 1998) and Virgin America (est.
2007). Yet in the airlines it is well known that there is a strong mobility
barrier between strategic groups (Caves and Porter, 1977; Mascarenhas
and Aaker, 1989), namely the ability of predominantly domestic low-cost
carriers (strategic group: low-cost carriers) to enter international long-haul
operations (strategic group: network/legacy carriers). In fact, McGee and
Thomas (1986) saw mobility barriers as a useful way to classify strate-
gic groups. Mobility barriers within industries differ substantially from
one industry to another, but their specificities do not only explain intra-
industry competition dynamics but also barriers of entry for new entrants.
To give an idea of the specificities of mobility barriers one can cite
factors that preclude the ability of low-cost airlines to embark on long-
haul flights – so far an exclusive domain of network carriers and charter
airlines. Morrell (2008) cited factors such as higher proportion of fuel
costs in total trip costs, difficulty to maintain the same high labour and
aircraft productivity, higher station costs because of greater distances
(crew layovers and administrative costs), more complex scheduling (time-
belts and night curfews), and more costs associated with in-flight services.
Addressing the same question, Gudmundsson (2015) argued that differ-
ent network factors (economies of density, optimized stage lengths and
feeding) were required such as market protection (use of secondary air-
ports to avoid head-on competition).
Although the strategic group of low-cost airlines is little by little moving
into international markets, Southwest Airlines has 1.1 per cent market
share in the US international passenger market, and JetBlue 3.3 per cent,
a low comparable share and for a very good reason. The primary reason
is that these carriers do not enjoy the same network advantages (Pustay,

M4371-FINGER_9781786431851_t.indd 246 14/11/2017 11:14

1980) generated by the legacy carriers through branded alliances and

hub-and-spoke operations, crucial to fill larger aircraft on long-haul inter-
national flights. Low-cost carriers replicating such effects by connecting
domestic and international networks over hubs undermine the strategic
fit of activities that help keep costs low (Porter, 1996). For this reason the
two strategic groups are like comparing apples and oranges: both are air-
lines, but the underlying genetics of one strategic group thwarts effective
competition with the other strategic group on the international long-haul
segment of the market.
INTERDEPENDENCE
The assumption of interdependence in oligopolies implies that each firm’s

actions affect other firms, and therefore also affect market conditions
(Stigler, 1964). Consequently, each firm evaluates the possible reactions
of other competing firms and what countermoves to take. The theory
assumes price rigidity from the anticipation of countermoves, as firms will
only adjust prices and quantity of output in accordance with a market
leader.
Interdependence of two strategic groups in an oligopolistic market
is less than if all market players operated the same business model. The
reason for this is that low-cost airlines like Southwest have a different
route structure, point-to-point, as opposed to hub-and-spoke, causing low
direct route competition. We can also argue that hub-and-spoke networks
are separated due to the dominance of a particular carrier in each hub. In
other words, the market power of oligopoly firms in the airline industry is
to some degree separated, so customers’ access to different prices and sat-
isfactory services varies considerably from one market segment to another.
Thus, direct and indirect competition matters, and different strategic
groups have markedly different dynamics in how they compete within the
group and across the groups.
The concept of multimarket contact is one way to assess how interde-
pendent firms are in various markets. The greater the multimarket contact,
the greater the interdependence and the less vigorous the competition
among the firms in oligopolistic markets. Kahn (1950), Edwards (1955)
and Singal (1996) showed that multimarket contact facilitates an oligopo-
listic collusion and market power if collusion is unsustainable. Thus, in an
oligopolistic airline industry we expect fares to be higher in markets with
a high degree of multimarket contact. Much research has investigated the
hypothesis showing price increases in such markets (Evans and Kessides,
1994; Singal, 1996). For example, Evans and Kessides (1994) found that

M4371-FINGER_9781786431851_t.indd 247 14/11/2017 11:14

legacy airlines follow the golden rule, avoiding price changes in one route in
order to avoid competitors’ retaliation in other contested routes. However,
this relationship does not necessarily apply across strategic groups.
Southwest Airlines has a large proportion of monopoly routes but
refrains from charging monopoly prices for two reasons: first, the need to
create new demand for its services; second, to compete with other trans-
port modes such as cars and buses, rather than with other airlines. For
this reason, a price change by a dominant airline in a nearby hub airport
is unlikely to affect Southwest’s pricing on a nearby point-to-point route.
In support of this hypothesis, Brueckner et al. (2013) showed that legacy
airline competition had weak effects on average fares, while low-cost airline
competition, no matter how it occurs, had major impacts on legacy car-
riers’ average fares, implying different behaviour of strategic groups in
oligopolistic markets.
INDUSTRY ENTRY AND EXIT
New entry and exit to the industry has fluctuated considerably and has
been subject to various internal and external conditions. In an attempt
to explain some of the influencing factors, Gudmundsson and Van
Kranenburg (2002) studied entry rates into the US airline industry. They
found that political pro-entry interventions and announcements of inter-
ventions were significantly related to entry rates. Other important positive
factors were reduction in acquisition and leasing costs of aircraft, and the
rate of GDP growth.
To give an idea as to the entry and exit dynamics in the industry, 176
operating certificates for jet services were issued in the US from 1979 to
1999; of these, 123 (70 per cent) started operations and 37 (21 per cent)
airlines were still operating in 1999 (Gudmundsson and Van Kranenburg,
2002). Similarly, ATA (2008) examining exit (Chapter 11 and 7 bankrupt-
cies) of air carriers from 1979 until 2008, reported that 185 carriers had
disappeared.2
Thus, we can talk of two sources of concentration, namely mergers and
acquisitions and exit. However, if entry of new airlines to the industry is
taking place, it acts to some extent as a countervailing force against con-
centration. At the same time, industry concentration and new entry have
occurred in waves. In other words, natural market equilibrium appears to
be non-existent (Button, 1996). Thus, the common assumption that only
a few airlines will survive in the market is not an inevitable conclusion of
market deregulation unless the focus is on the legacy carriers, but exclud-
ing new entrants (Swan, 2005).

M4371-FINGER_9781786431851_t.indd 248 14/11/2017 11:14

Entry barriers have been studied to a considerable extent in the context

of the air transport industry. The argument at the dawn of deregulation
was that entry costs for new airlines were not detrimental as the aircraft
asset could be readily acquired and disposed of due to an efficient market
for aircraft (Kahn, 1988). What is more, economies of scale were consid-
ered non-existent or weak at the firm level, although clearly present at the
aircraft level (Bailey and Panzar, 1981; Caves, 1962). In other words, the
existence of economies of density and scope through airport hubs enabled
airlines to operate larger aircraft with lower seat costs (Levine, 1987),
creating an access barrier for new entrants. However, the low-cost model
effectively got around this barrier by focusing on secondary markets and
by forging efficiency through high productivity of the aircraft asset (fast
turnarounds) and employees (simple operating model).
Air transport markets have relatively low entry and exit costs compared
to many other capital-intensive industries due to the mobility of the key
asset: the aircraft (Kahn, 1971). Regardless, history has told us that the
number of start-up filings rises in the wake of merger waves and reces-
sions due to better aircraft availability and airport access in these periods.
Following the most recent wave of mergers, there are a number of start-ups
going through licensing at the DoT, some of which are already operating
on-demand services (Bachman, 2016). However, over time the industry
has matured and the density of the route system has increased, raising
barriers for new carriers using conventional business models like low cost.
Although the cost of entry may hardly have changed, other competition-
related barriers such as route access may have increased with time, calling
for new innovative business model ideas.
MERGERS AND ACQUISITIONS
Frequent mergers in the airline industry have raised the question of

whether the industry has become more oligopolistic. It is often stated that
consolidation is inevitable due to scale economies and cost efficiencies
to firm scale. We know that, in the presence of strong scale economies,
consolidation will form a long-run oligopolistic equilibrium in the indus-
try (Bain, 1954). However, little research evidence exists to back up such
claims for the airline industry. On the contrary, most academic studies
show constant returns to firm scale (Caves et al., 1984; Keeler and Formby,
1994; Oum and Zhang, 1997), but significant scope and density economies
(Keeler and Formby, 1994; Levine, 1987).
In the absence of scale economies, but in the presence of cyclical vari-
ations in excess capacity, the industry is likely to be inherently unstable

M4371-FINGER_9781786431851_t.indd 249 14/11/2017 11:14

(Button, 1996). Under this assumption the industry concentration rises

and falls, so far as entry and exit costs (and barriers) remain stable. The
evidence tells that since deregulation in 1978, mergers and acquisitions
have taken on a waveform. Table 13.1 shows mergers in the industry from
1990 to 2013 and the merger waves that took place in 1987–93, 1997–2001,
2005–07 and 2009–13.
Antitrust legislation is a tool to assure market openness to competi-
tion, ease of entry for new competitors, and competitive prices for con-
sumers. In the spirit of these aims, in 2013, the Department of Justice
(DoJ) attempted to block the merger between American and US Airways,
implying a harder stand on major mergers seen as hurting competition.
Although few assume that the industry will consolidate into one giant
airline and a virtual monopoly, pressure on the regulator rises after merger
waves and short-run price increases. In such periods airlines will continue
to file for Chapter 11 bankruptcies and seek mergers to avoid outright
exit (Chapter 7), options that are increasingly under pressure from the
regulator.
In fact, past merger waves have shown that mergers have not brought
about the expected cost savings from synergies and scale economies as
expected (Gudmundsson et al., 2016), pushing attention to the revenue side
and competition-reduction effects. In other words, competitive advantage
from larger size is not guaranteed in the long run, at least on the cost side,
pointing to other reasons behind mergers and acquisitions.
PRICING IMPLICATIONS
Airfares in the US have fallen in constant dollars since the start of deregu-
lation. However, annual change in airfares has fluctuated considerably over
the same period (see Figure 13.1). The average fare decreases over longer
periods only give us a bird’s-eye view, as concentration in the industry may
penalize certain market segments and benefit others; it may also change
pricing behaviour of the oligopolist. Finally, temporal fare effects may take
place with short periods of increases following waves of concentration
without changing the overall downward trend in average fares, assuming
there is no long-run stability following periods of concentration.
In other words, the fluctuating concentration index in the industry causes
temporary benefits at best when the index is high; these diminish gradu-
ally as efficiency-driven mergers and acquisitions open space for emerging
airlines to gain a foothold in new markets while unprofitable routes are
shed by the merged airlines trying to extract merger benefits. As previously
mentioned, this relationship should hold for merger-driven concentration,

M4371-FINGER_9781786431851_t.indd 250 14/11/2017 11:14

Table 13.1 Mergers and acquisitions in the US, 1990–2013
M&A Year Type

American Airlines / Eastern Air Lines: Latin Routes 1990 L/L
American Airlines / TWA: Heathrow Routes 1991 L/L
M4371-FINGER_9781786431851_t.indd 251
Delta Air Lines / Pan Am Airlines (Shuttle and Atlantic Routes) 1991 L/L
Air Wisconsin / Aspen Airways 1991 R/R
United Airlines / Pan Am: Latin and Caribbean Routes 1992 L/L
Southwest Airlines / Morris Airlines 1993 LC/LC
AirTran Airways / Valujet 1997 LC/LC
American Airlines / Reno Air 1999 L/LC
Delta Air Lines / Atlantic Southeast Airlines 1999 L/R
Delta Air Lines / Comair 1999 L/R
251
American Airlines / TWA 2001 L/L
US Airways / America West 2005 L/LC
Republic Airways / Shuttle America 2005 L/R
SkyWest / Atlantic Southeast 2005 R/R
Pinnacle Airlines / Colgan Air 2007 R/R
Delta Air Lines / Northwest 2009 L/L
United Airlines / Continental 2010 L/L
Southwest Airlines / AirTran 2010 LC/LC
US Airways / American 2013 L/L
Note: LC 5 low cost; L 5 legacy, R 5 regional.
Source: Compiled by the author.

14/11/2017 11:14
but may not for competitive advantage-driven concentration (for example,

Southwest Airlines) that may continue eating into market share from the
legacy airlines, so far as the underlying competitive advantages stay intact.
Although the oligopoly tendencies of the industry have not led to long-
run increases in fares on average since deregulation, airport concentration
has strengthened market power in hub airports, leading to a so-called hub
premium. A hub premium occurs when a higher fare is charged by the
dominating hub airline for flights originating and terminating at the hub.
Research in the early 1990s revealed that in the event of hub dominance,
dominant airlines gained a fare premium ranging from 31.5 per cent
(Minneapolis) down to −4.0 per cent (St Louis), with an overall average of
17.7 per cent for the top 10 airports in the US (Borenstein, 1992). However,
one of the consequences of mergers among the legacy carriers was de-
hubbing (for example, Memphis), as the merged airlines sought efficiency
in the combined network. Empirical research suggests (Tan and Samuel,
2016) that when de-hubbing takes place, airfares decrease if a low-cost
carrier is present at the de-hubbed airport, whereas airfares go up if the
de-hubbed airport is only served by a legacy carrier.
Similar findings were unravelled for Europe; for example, Lijesen et al.
(2001) reported an average hub premium of 15 per cent for Lufthansa, Air
France and Swissair at their primary hubs (Frankfurt, Paris and Zurich).
However, international alliances somewhat skewed this relationship, as
code-sharing agreements (through international alliances) beyond the hub
led to a 27 per cent reduction in fares (Brueckner, 2001). Thus, on average,
the dominant airline enjoyed a hub premium, that was to some extent
offset by lower ex-hub airfares. In other words, the early work showed
that despite the market power of the largest carriers, alliance code-shares
appeared to moderate the market power effects of dominant airlines at
hubs.
Later research on hub airports has shown a complicated relationship
that is not perfectly in line with economic theory. Alderighi et al. (2015),
researching European airlines, argued that although unilateral code shares
should, according to theory, be welfare enhancing by reducing double mar-
ginalization, the relationship only holds for the highest fare classes, but not
for the lowest fare classes that are not subject to the double marginalization
problem to the same degree. In other words, the impact of market power
at hub airports on fares stemming from industry concentration may be
moderated to some extent through alliance relationships beyond the hubs
as pointed out before, but not for all market segments.
What is clear is that legacy carriers face dynamic competition regard-
less of the level of industry concentration. For example, if a low-cost
carrier enters a market dominated by a legacy carrier, its airfares

M4371-FINGER_9781786431851_t.indd 252 14/11/2017 11:14

decrease dramatically both at the low end and the high end of the fare
dispersion (Tan, 2016). Thus, there is evidence that the proportional
pricing impact of smaller players in the industry – most of which are
low-cost or low-fare airlines – is proportionally more important than
their market share indicates, due to the strong reaction of legacy carriers,
but not vice versa.
Thus, concentration among low-cost airlines versus legacy airlines
may have markedly different impact on airfares. Dobson and Piga (2013)
argued that the EasyJet–Go Fly and Ryanair–Buzz acquisitions had
impacted pricing on the affected routes. In fact, they showed that early-
booking fares decreased, although late-booking fares increased. In other
words, the acquisitions had a net benefit for price-conscious consumers
who booked early. Conversely, research focusing on mergers among legacy
airlines in the US showed that mergers, on average, caused short-run fare
increases of 2.3 to 5.9 per cent (Jain, 2015). Although few studies have been
performed on the difference in price effects between legacy carrier mergers
and low-cost airline mergers, it nevertheless raises the question of whether
the pricing effects are different depending on strategic groups.
Figure 13.1 shows the general trend in airfares in the US domestic
market from 1995 to 2016. The figure clearly shows that, on average, fares
increased substantially during the financial crisis period in 2009, but then
began to decline again (3.8 per cent in 2015). Although a global measure
like this does not unravel all the nuances of individual route impacts, it is
nevertheless an indicator of the general trend in the industry.
Jain (2015) showed that although fares increased following mergers, such
effects are just a temporary halt in the long-run downward trend. Thus,
fluctuation in industry concentration does not have a negative long-run
impact on average price trends. On the contrary, the industry shows sub-
stantial decrease in long-run average airfares despite periods of concen-
tration. This being said, the price impact of concentration on individual
routes may differ based on the close-substitutes hypothesis (Lijesen, 2004).
In this regard, close substitutes constitute two routes that exhibit high
cross-elasticity of demand.
There is another unique airfare related development in air transport
markets, namely the increased separation of the ticket price from growing
ancillary revenues. For example, in 1990 fares composed 88.5 per cent of
the airlines’ total revenue from passenger transport, whereas in 2015 this
component represented only 74.7 per cent and 73.8 per cent in 2016 (BTS,
2016). For this reason, the fare component is likely to continue to decrease
in the long run as airlines develop more sophisticated sources for ancillary
revenues.

M4371-FINGER_9781786431851_t.indd 253 14/11/2017 11:14

500 8.0
450 6.0
400 4.0
350 2.0
Annual Change (%)

Average Fare ($)
300 0.0
250 –2.0
200 –4.0
150 –6.0
100 –8.0
50 –10.0
0 –12.0
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Average Fare ($) Annual Change (%) Linear (Average Fare ($))
Note: 2016 constant $ using Bureau of Labor Statistics Consumer Price Index.
Source: Compiled by the author from Bureau of Transportation Statistics (2016).
Figure 13.1 Average fares in US domestic market
MEASURING CONCENTRATION
Two questions have been central in the quest to measure concentration

in an industry: how to measure the market power of the largest firms,
and how to measure the disparity in market shares among all firms in an
industry. Obviously there is a large difference between an oligopolistic
industry like civil aircraft manufacturing, with two large firms in one
relatively open world market (Airbus and Boeing) and few smaller firms
(Bombardier, Embraer, ATR, etc.) and oligopoly in the airline industry,
with one to several large firms and many small firms in each of many
sheltered national markets (predominantly restricted bilaterals and no
cabotage rights). Measuring concentration under these latter conditions
is less obvious, and current methods are increasingly under scrutiny by
failing to take into account the wider network effects (Lijesen, 2004;
Roberts, 2014).
Various methods have been suggested over time, especially to aid
authorities in anti-trust cases. To give a brief insight to the development

M4371-FINGER_9781786431851_t.indd 254 14/11/2017 11:14

of these measures, the following section will cover the most important
methods and their limitations.
The most notable industry concentration measures are the Lorenz–Gini
index, the concentration ratio (C4) (see Finkelstein and Friedberg, 1967)
and the Herfindahl–Hirschman Index (HHI) (Hirschman, 1969). The
latter has generally replaced the former two, although the concentration
ratio is considered useful to measure concentration, and the Lorenz–Gini
index useful to measure industry dispersion. The Lorenz–Gini approach
works well when applied to a large number of observations, but is less
practical for industry concentration as it only measures disparity in market
shares and fails to take into account the size of a firm’s market share
(Roberts, 2014). However, the industry concentration ratio (CR) provides a
more straightforward measure of concentration, making up for this short-
coming. The CR is expressed as follows:
CR m 5 a si
m
i51
where si is the market share and m defines the mth firm in the industry.
Using the CR, cut-off values were established as between 0 and 50 per
HH805toa100
cent market share for a low-concentration N industry, 50 to 80 per cent for
medium concentration, and from s2i 3per100cent for high concentra-
i51
tion. Since the CR has been criticized for revealing little about the inequal-
ity among the size distribution of firms, researchers sought other measures
a si a(1969),
to aid analysis for anti-trust court cases.
m
n
One solution was suggested CR mby5Hirschman f (i) jt,k
who further devel-
i51
j i51
oped the work of Herfindahl, SL t,k 5 1
culminating 2 in the HHI. The HHI has
nt,k
become the standard in applied concentration analysis, and is expressed
as follows:
HHT^5 a s2i 3 100

N
i51
where si is the market share of firm i in the market, and N is the number
a i51 f (i) t,k of when the indus-

of firms. n
j
An important aspect of the HHI T^ is the determination
SL jt,k 5 1Although
try can be considered concentrated. 2 cut-off values must not be
nt,k
applied blindly given the unobserved factors affecting competition in the
industry, certain ranges have been laid out. The US Department of Justice
0C/0T^
(DoJ) and the Federal Trade Commission (FTC) (US DoJ and FTC, 2010)
^
T analysis of merger cases: HHI < 1500 is a
set the following values in their
competitive market, an HHI of 1500 to 2500 is a moderately concentrated
market, and an HHI of ≥ 2500 is a highly concentrated marketplace.
0B 0B 0B
Furthermore, in merger analysis the , DoJ
0, normally
, 0,and uses ,a0change of 100
T^ 0T̂ 0ts 0ts0
0C/0T^ Finger and Kenneth Button - 9781786431851

Matthias
0B 0B 0B
M4371-FINGER_9781786431851_t.indd 255 , 0, , 0,and s
,0 14/11/2017 11:14
in the HHI following a merger as a cut-off factor normally provoking

investigation.
In his original work, Hirschman (1969) pointed out pros and cons of
the index, for example that it might be a useful cost-effective measure of
concentration but not the only criterion to use, as some industries might
show the characteristics of high concentration and yet be highly com-
petitive. For this reason he suggested several other variables to practition-
ers, such as locational distribution, psychology of officers and product
substitutability (Hirschman, 1969; Roberts, 2014). In addition, Roberts
(2014) pointed out that the HHI misses out industry-specific factors that
influence competition intensity, such as market power, barriers to entry,
economies of scale or scope, technology changes and firm-specific charac-
teristics. In the context of airlines, network effects and strategic groups can
be added to the list.
In line with these observations, several researchers have suggested
adjustments to the HHI. One particular adjustment tested on airline data
proposed by Lijesen (2004) is of interest. He argued that the HHI is sensi-
tive to geographical boundaries and product homogeneity, and suggests a
HHI measure that accounts for close substitutes. In this way the HHI is
adjusted by weighing market shares by a quality indicator (relative travel
time), providing a better approximation of concentration in markets with
close substitutes.
INDUSTRY CONCENTRATION LEVEL
Method and Data
The primary data used in this chapter is the US Department of

Transportation’s Form 41 Traffic data US carriers and all carriers. That
includes full data for all US carriers in the domestic market and full data
for all carriers on international routes to and from the US. The Domestic
Airline Consumer Airfare Report was used to extract annual changes
in average airfares in the US domestic market. However, no fare data is
available for international routes to and from the US. In the data-set the
first year of available passenger data was 1990, and 1995 for domestic fare
data.
The HHI was calculated first globally for the domestic market, and then
for the international market. All the sub-indices were calculated based on
the two global indices.

M4371-FINGER_9781786431851_t.indd 256 14/11/2017 11:14

Overall
500 US Domestic and International HHI 8.0
450 6.0
Using the DoJ and FTC definition of industry concentration, the US
400 4.0
airline industry is competitive both in the domestic and the international
markets.
350 Figure 13.2 shows that the HHI index (HHI-DOM) for domestic 2.0
Annual Change (%)

Average Fare ($)
air transport
300
in the US ranges from 1026 in 1990 to 1061 in 2015. However,
0.0
both the beginning (1990) and the end of the period (2015) have high HHI
250
compared to 2007 when the HHI went down to 673. Using the HHI–2.0 cut-off
values
200we can see that the domestic US industry has shown moderate –4.0 con-
centration
150
in two periods: until 1992, and following the Financial –6.0 Crisis in
2008. For the rest of the period since deregulation, the domestic industry
has 100
shown an unconcentrated index. –8.0
The50 level of concentration is even lower when examining the –10.0HHI
index for international air transport (HHI-INT) from and to the US,
0 –12.0
ranging from 414 in 1990 to 518 in 2015, with the highest index values
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
in 2012, at 564. In other words, the industry – according to the HHI
index forAverage
the international
Fare ($) marketChange
Annual – is (%)
far fromLinear
being concentrated.
(Average Fare ($))
However, the international trend line is rising whilst the trend line for
domestic concentration is falling, regardless of a substantial rise in the
HHI-DOM from 2009 until now. In other words, although the fluc-
tuations in the HHI-DOM are greater, the HHI-INT is more stable but
gradually rising.
1400
1200
1000
800
600
400
200
0
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
HHI-INT HHI-DOM Linear (HHI-INT) Linear (HHI-DOM)
Figure 13.2 Herfindal–Hirschman Index: US international and domestic

markets

M4371-FINGER_9781786431851_t.indd 257 14/11/2017 11:14

US Domestic HHI
Table 13.2 and Figure 13.3 show clearly the remarkable rise of the low-cost
airlines in the US domestic industry (HHI-LC10 and HHI-LC4), effec-
tively challenging the legacies (HHI-L10 and HHI-L4). However, in the
latter part of the period the legacies (HHI-L10) have somewhat increased
the concentration through mergers and acquisitions, although far from the
1990 level (HHI-L10 in 1990 5 973; HHI-L10 in 2010 5 433; HHI-L10
in 2015 5 571). What is clear is that the concentration in the early 1990s
is different from today, due to a stronger position of low-cost carriers and
Southwest Airlines in particular (HHI-LC10 in 1990 5 42; HHI-LC10 in
2015 5 466). What is more, based on the past trend Southwest is likely to
increase its market share due to its sustainable competitive advantage over
the legacy carriers, while the latter are likely to maintain their position
through mergers and acquisitions.
US International HHI
Table 13.3 and Figure 13.4 show the HHI values of US carriers and inter-
national carriers in the top four (HHI-4, only one international carrier is
present in this group), top 10 (HHI-10) and top 20 (HHI-20) carriers in
the international market. It is clear that US carriers have increased their
share in this market (HHI-US10 in 1990 5 299; HHI-US10 in 2015 5 451),
Table 13.2 Market shares (passenger traffic) and concentration (HHI):

US domestic market
1990 2000 2010 2015

Top 4 56.60% 51.90% 48.60% 60.20%
HHI-4 807 689 645 966
Legacy 56.60% 39.80% 31.80% 39.80%
HHI-L4 807 542 363 549
LC 0.00% 12.10% 16.80% 20.40%
HHI-LC4 0 146 282.6 416
Top 10 90.70% 84.90% 73.50% 80.60%
HHI-10 1015 912 760 1037
Legacy 81.70% 69.50% 43.50% 46.40%
HHI-L10 973 755 433 571
LC 9.00% 15.30% 30.00% 34.20%
HHI-LC10 42 157 327 466

M4371-FINGER_9781786431851_t.indd 258 14/11/2017 11:14

100.00% 1200
90.00%
1000
80.00%
70.00% 800
60.00%
M4371-FINGER_9781786431851_t.indd 259
50.00% 600
40.00%
30.00% 400
20.00% 200
10.00%
0.00% 0
1990 2000 2010 2015 1990 2000 2010 2015
(a) Top 4 Top 10 (b) HHI-4 HHI-10
90.00% 1200
259
80.00%
1000
70.00%
60.00% 800
50.00%
600
40.00%
30.00% 400
20.00%
200
10.00%
0.00% 0
1990 2000 2010 2015 1990 2000 2010 2015
(c) Legacy-4 LC-4 Legacy-10 LC-10 (d) HHI-L4 HHI-C4 HHI-L10 HHI-LC10

Figure 13.3 Market shares (passenger traffic) and concentration (HHI): US domestic market
14/11/2017 11:14
Table 13.3 Market shares (passenger traffic) and concentration (HHI):

US international market
1990 2000 2010 2015

Top 4 30.3% 32.2% 39.1% 39.8%
HHI-4 242 290 413 453
Top 10 57.6% 53.5% 57.1% 53.2%
HHI-10 370 372 471 485
Top 10 US 45.4% 37.6% 43.3% 39.4%
HHI-US10 299 319 431 451
Top 10 International 12.3% 15.9% 13.9% 13.8%
HHI-INT10 51 53 40 33
Top 20 74.50% 68.40% 69.7% 66.0%
HHI-20 404 396 488 502
Top 20 US 48.90% 42.10% 48.30% 43.3%
HHI-US20 325 326 438 456
Top 20 International 25.60% 26.40% 24.70% 22.70%
HHI-INT20 78 70 50 46
while the international carriers show reduced concentration in the top 10

(HHI-INT10 in 1990 5 51; HHI-INT10 in 2015 5 33) and in the top 20
(HHI-INT20 in 1990 5 78; HHI-INT20 in 2015 5 46).
In other words, despite much debate in the US regarding the rise of the
Middle East carriers, these carriers have had little impact on concentration
in the US international market. Emirates had 1.5 per cent share in 2015,
Qatar 0.6 per cent, Etihad 0.6 per cent and Saudi Arabian 0.2 per cent. In
comparison the top 4 carriers were United 12.7 per cent, American 12.4
per cent, Delta 11.3 per cent, and British Airways 3.5 per cent.
CONCLUSION
In this chapter, concentration was measured by separating domestic and
international markets, revealing different portrayals of the level of con-
centration. In the calculations, HHI market share for passenger traffic was
used (instead of the usual available seat-miles or total revenues) as a more
viable method to account for different business models of low-cost and
legacy carriers.

M4371-FINGER_9781786431851_t.indd 260 14/11/2017 11:14

80.0% 600
70.0%
500
60.0%
400
50.0%
40.0% 300
30.0%
200
M4371-FINGER_9781786431851_t.indd 261
20.0%
100
10.0%
0.0% 0
1990 2000 2010 2015 1990 2000 2010 2015
(a) Top 4 Top 10 Top 20 (b) HHI-4 HHI-10 HHI-20
60.0% 500
50.0%
261
400
40.0%
300
30.0%
200
20.0%
10.0% 100
0.0% 0
1990 2000 2010 2015 1990 2000 2010 2015
Top 10 US Top 10 International HHI-US10 HHI-INT10
(c) Top 20 US Top 20 International (d) HHI-US20 HHI-INT20

Figure 13.4 Market shares (passenger traffic) and concentration (HHI): US international markets
14/11/2017 11:14
The oligopolistic market position of a firm can arise from competitive

advantage, and in this chapter it has been argued that Southwest Airlines,
which has the highest market share, is in that position. Meanwhile, the
other large US carriers among the top four have all gone through bank-
ruptcies and maintained their market position through major mergers.
Because of this competitive advantage, Southwest Airlines is likely to
continue to grow in domestic market share, but faces a mobility barrier
embarking on international long-haul routes.
Based on the data, industry concentration in the US domestic market
shows substantial fluctuations – usually increasing in the years following
recessions, when the number of financially weak carriers in the industry
increases. However, concentration – as measured by the global HHI – has
receded during economic prosperity.
Overall, the oligopolistic characteristics of the domestic market are
stronger than in international markets, yet both markets do not show
strong concentration using the DoJ criteria. One explanation is that the
industry has had a large number of new entrants since deregulation that
have challenged the legacy carriers in the domestic market in a sustainable
way. The US international market has been competitive over many decades
but shows sign of concentration driven by M&As among US carriers..
Thus, one must conclude that the gradual rise of market shares among
Middle East airlines (although proportionally low) has higher impact on
the competition dynamics among international airlines in the US market
than on US airlines per se.
The competitiveness of the industry depends on new entry that tends
to fluctuate substantially subject to entry barriers and political pro-entry
interventions (Gudmundsson and Van Kranenburg, 2002). So far as entry
barriers are not overwhelming, industry concentration is likely to fluctuate
according to economic cycles and shocks. However, domestic concentra-
tion indices may continue to increase, as low-cost airlines with sustainable
competitive advantage continue to grow, while legacy airlines strive to
maintain position through mergers and acquisitions.
NOTES
1. American Airlines’ Crandall says, ‘Airlines make up world’s most inefficient oligopoly.’
Marketing News, San Diego, 23 April 1976, p. 9.
2. At the dawn of EU liberalization (1993 to 1999) there were 131 start-up airlines in the
15 EU countries; of these, 74 had disappeared by 1999 (77 per cent failure rate). Of
124 incumbent airlines operating at the beginning of 1993, 55 had disappeared by 1999
(44 per cent failure rate) (Lyons, 2000). When market shares are examined between the
incumbent carriers and low-cost carriers, we see that in Europe, between 2003 and 2006,

M4371-FINGER_9781786431851_t.indd 262 14/11/2017 11:14

low-cost carriers’ market share increased by 40 per cent to 20 per cent overall share, but
the incumbent airlines lost 10 per cent, bringing it down to 55 per cent overall market
share. Another parallel development, stimulated by the advent of low-cost airlines, was
the reduction in premium intra-EU passenger traffic that declined by 33 per cent between
2001 and 2005 (AEA, 2007).
REFERENCES
Air Transport Association of America (ATA) (2008), ‘Airline bankruptcy and
service sessations’. Available at: www.airlines.org/economics/specialtopics.
Alderighi, M., A.A. Gaggero and C.A. Piga (2015), ‘The effect of code-share
agreements on the temporal profile of airline fares’, Transportation Research
Part A: Policy and Practice, 79, 42–54.
Association of European Airlines (AEA) (2007), ‘Operating Economy of AEA
Airlines’, December.
Bachman, J. (2016), ‘Where are all the startup airlines to save you money?’,
Bloomberg Technology News, 20 April.
Bailey, E.E. and J.C. Panzar (1981), ‘The contestability of airline markets during
the transition to deregulation’, Law and Contemporary Problems, 44(1),
125–45.
Bain, J.S. (1954), ‘Economies of scale, concentration, and the condition of entry in
twenty manufacturing industries’, The American Economic Review, 44(1), 15–39.
Borenstein, S. (1992), ‘The evolution of U.S. airline competition’, Journal of
Economic Perspectives, 6(2), 45–73.
Brueckner, J.K. (2001), ‘The economics of international codesharing: an analysis of
airline alliances’, International Journal of Industrial Organization, 19, 1475–98.
Brueckner, J.K., D. Lee and E.S. Singer (2013), ‘Airline competition and domestic
US airfares: A comprehensive reappraisal’, Economics of Transportation, 2(1),
1–17.
Bureau of Transportation Statistics (BTS) (2016), Airfares P-1.2.
Button, K. (1996), ‘Liberalising European aviation: is there an empty core
problem’, Journal of Transport Economics and Policy, 30(3), 275–91.
Caves, D.W. (1962), Air Transport and its Regulators, Cambridge, MA: Harvard
University Press.
Caves, D.W., L.R. Christensen and M.W. Tretheway (1984), ‘Economies of density
versus economies of scale: why trunk and local service airline costs differ’, The
Rand Journal of Economics, 15, 471–89.
Caves, R.E. and M.E. Porter (1977), ‘From entry barriers to mobility barriers: con-
jectural decisions and contrived deterrence to new competition’, The Quarterly
Journal of Economics, 91(2), 241–61.
Dobson, P.W. and C.A. Piga (2013), ‘The impact of mergers on fares structure:
evidence from European low-cost airlines’, Economic Inquiry, 51(2), 1196–217.
Edwards, C.D. (1955), ‘Conglomerate bigness as a source of power’, in the NBER
conference report, Business Concentration and Price Policy, Princeton, NJ:
Princeton University Press.
Evans, W.N. and I.N. Kessides (1994), ‘Living by the “golden rule”: multimarket
contact in the US airline industry’, The Quarterly Journal of Economics, 109(2),
341–66.

M4371-FINGER_9781786431851_t.indd 263 14/11/2017 11:14

Finkelstein, M.O. and R.M. Friedberg (1967), ‘The application of an entropy

theory of concentration to the Clayton Act’, Yale Law Journal, 76(4), 679–83.
Gudmundsson, S.V. (1998), Flying Too Close to the Sun: the Success and Failure of
the New-entrant Airlines, Aldershot: Ashgate Publishing.
Gudmundsson, S.V. (2015), ‘Limits to the low-cost niche? Finding sustainable
strategies for low-cost long-haul airlines. Discussion Paper, Paris Air Forum, 22
June.
Gudmundsson, S.V. and H.L. van Kranenburg (2002), ‘New airline entry rates in
deregulated air transport markets’, Transportation Research Part E, 38, 205–19.
Gudmundsson, S.V., R. Merkert and R. Redondi (2016), ‘Cost functions and deter-
minants of unit cost effects in horizontal airline M&As’, Working Paper.
Hirschman, A.O. (1969), National Power and the Structure of Foreign Trade, origi-
nally published in 1945, Berkeley, CA: University of California Press.
Jain, V. (2015), ‘What did the wave bring?: Short-term price effect of the US airline
merger wave (2009–2012)’, Journal of Economic Policy and Research, 10(2),
49–68.
Kahn, A.E. (1950), ‘The Chemicals Industry’, in Walter Adams (ed.), The Structure
of American Industry: Some Case Studies, New York: Macmillan.
Kahn, A. (1971), The Economics of Regulation, II: Institutional Issues, New York:
Wiley & Sons.
Kahn, A. (1988), ‘Surprises of airline deregulation’, American Economic Review,
Papers and Proceedings, May, pp. 316–22.
Keeler, J.P. and J.P. Formby (1994), ‘Cost economies and consolidation in the US
airline industry’, International Journal of Transport Economics, 21(1), 21–45.
Levine, M.E. (1987), ‘Airline competition in deregulated markets: theory, firm
strategy, and public policy’, Yale Journal on Regulation, 4, 393–494.
Lijesen, M.G. (2004), ‘Adjusting the Herfindahl index for close substitutes: an
application to pricing in civil aviation’, Transportation Research Part E: Logistics
and Transportation Review, 40(2), 123–34.
Lijesen, M.G., P. Rietveld and P. Nijkamp (2001), ‘Hub premiums in European
civil aviation’, Transport Policy, 8(3), 193–9.
Lyons, P.K. (2000), Transport Policies of the European Union, January, EC-Inform.
Mascarenhas, B. and D.A. Aaker (1989), ‘Mobility barriers and strategic groups’,
Strategic Management Journal, 10(5), 475–85.
McGee, J. and H. Thomas (1986), ‘Strategic groups: theory, research and tax-
onomy’, Strategic Management Journal, 7(2), 141–60.
Morrell, P. (2008), ‘Can long-haul low-cost airlines be successful?’, Research in
Transportation Economics, 24(1), 61–7.
Oum, T.H. and Y. Zhang (1997), ‘A note on scale economies in transport’, Journal
of Transport Economics and Policy, 31, 309–15.
Porter, M.E. (1996), ‘What is strategy?’, Harvard Business Review, November.
Pustay, M.W. (1980), ‘Airline competition and network effects’, Transportation
Journal, 19(4), 63–72.
Roberts, T. (2014), ‘When bigger is better: a critique of the Herfindahl–Hirschman
Index’s use to evaluate mergers in network industries’, Pace Law Review, 34(2),
894–946.
Singal, V. (1996), ‘Airline mergers and multimarket contact’, Managerial and
Decision Economics, 17(6), 559–74.
Stigler, G.J. (1964), ‘A theory of oligopoly’, Journal of Political Economy, 72(1),
44–61.

M4371-FINGER_9781786431851_t.indd 264 14/11/2017 11:14

Swan, W.M. (2005), ‘Consolidation in the airline industry’, working paper, accessed
10 September 2016 at www.sauder.ubc.ca/. . ./consolidation%20paper.ashx.
Tan, K.M. (2016), ‘Incumbent response to entry by low-cost carriers in the US
airline industry’, Southern Economic Journal, 82(3), 874–92.
Tan, K.M. and A. Samuel (2016), ‘The effect of de-hubbing on airfares’, Journal of
Air Transport Management, 50, 45–52.
US Department of Justice and the Federal Trade Commission (DoJ and FTC)
(2010), Horizontal Merger Guidelines, 19 August.

M4371-FINGER_9781786431851_t.indd 265 14/11/2017 11:14

14. Domination of hub-and-spoke
systems
Marc C. Gelhausen and Peter Berster
INTRODUCTION
A hub-and-spoke (HS) system requires a smaller number of flights for con-

necting each node compared with a pure origin–destination (OD) system,
thus leading to a more efficient system in terms of resources employed
(‘economics of density’). Thus, an HS can generate more routes and
frequencies compared with an OD system because of the higher traffic
volumes and the high share of transfer passengers (Redondi et al., 2011,
2012). This is particularly true for intercontinental destinations, where
many OD markets are less dense and cannot be viably served at a reason-
able frequency. Therefore, local passengers benefit from a hub system with
high transfer passenger volumes as well.
Airline deregulation started almost 40 years ago when the US Airline
Deregulation Act was approved by the US Congress (Goetz and Vowles,
2009) and more countries have followed in subsequent years. For example,
deregulation began in Australia in the early 1980s (e.g. Hooper, 2005) and
in Europe in the late 1980s (e.g. Button et al., 1998; Ehmer et al., 2000).
Deregulation fostered HS systems, which were more spatially – but also
more temporally – concentrated on coordinated incoming and outgo-
ing flights (‘wave system’). Before deregulation, airline networks were
already spatially concentrated because of the system of bilateral agree-
ments and national carriers. However, these star-shaped networks were
not temporally coordinated, and transfer opportunities existed rather by
accident (Burghouwt and De Wit, 2005). Nevertheless, deregulation led
to intense competition between hubs, as many hubs compete for the same
OD markets. In this regard, European hubs have a geographical advantage
compared with American and Asian hubs relative to the main markets, but
this might change in the future with the emerging markets in East Asia and
the rise of the Gulf carriers (Redondi et al., 2011). Deregulation led many
legacy carriers in the US and Europe to switch from point-to-point (P2P)
systems to temporal coordinated HS systems (Burghouwt and De Wit,
266

M4371-FINGER_9781786431851_t.indd 266 14/11/2017 11:14

Domination of hub-and-spoke systems 267
2005). However, as mentioned, European networks were already spatially

concentrated, but only because of bilateral agreements and national flag
carriers being tied to their home base. Temporal coordinated HS systems
have cost- and demand-side advantages (e.g. Button, 2002) that airlines
tried to realize by reorganizing their network more towards HS.
However, deregulation also gave rise to the low-cost carriers (LCCs)
which are operating on P2P systems (e.g. Berster et al., 2012; Reynolds-
Feighan, 2001; Williams, 2001). In fact, the strategy of hub concentration
of legacy carriers opened the door for LCCs (Dobruszkes, 2006). These
markets are mainly in short- and medium-haul travel, for example, within
Europe or between Europe and North Africa, which are rather dense or
where sufficient demand can be generated by direct services and low fares.
While there is still an ongoing discussion as to whether the LCC model is
transferable to long-haul routes (see, for example, Daft and Albers, 2012),
hubbing makes many markets viable that are usually less dense compared
with short- and medium-haul travel. Demand for long-distance air travel
is less price-elastic than demand for short and medium air travel, so that
the potential for demand generation is limited and dense markets or feeder
traffic are needed (Morrell, 2008). Pels (2008) concluded that a sufficient
level of demand is mandatory for long-haul LCCs. Yet in many cases
LCCs would compete on such routes with well-established hub carriers
and they cannot play fully to their strengths, as operational efficiency as a
major competitive advantage is less important in long-haul travel than in
short- or medium-haul markets, due to longer flights and therefore fewer
turnarounds (Wensveen and Leick, 2009). Estimates of the cost advantage
of LCCs in long-haul travel range from up to 10 per cent (Moreira et al.,
2011) to 60 per cent at most (Morrell, 2008). A random connecting system
or ‘self-help hubbing’ (Malighetti et al., 2008) may generate some transfer
passengers, although far fewer than at a fully-fledged hub. For example,
the share of transfer passengers of Ryanair at London Stansted was
17.2 per cent (O’Connell and Williams, 2005). However, compared with
services of legacy carriers, ‘self-help hubbing’ is typically less convenient
for passengers, as the costs of the search for connecting flights, baggage
handling during stopover and the risk of missed connections are entirely
borne by the passenger (Grimme, 2011). Hubbing is necessary for airlines
to serve thin markets, in particular in long-haul travel, at a relatively high
frequency. Furthermore, larger aircraft are typically employed in inter-
continental long-haul markets for technical and economic reasons, which
intensifies the problem of maintaining attractive flight schedules on thin
routes and makes hubbing necessary (Wilken et al., 2016).
One of the negative effects of increased hubbing involves capacity
constraints especially during peak hours. While only a relatively small

M4371-FINGER_9781786431851_t.indd 267 14/11/2017 11:14

number of about ten airports were heavily capacity constrained, that is,
during most daytime hours, in 2008 (the year before the global economic
downturn) there was a high degree of average capacity utilization among
the largest 100 airports, which accounted for more than 50 per cent of
flights globally. About 15 per cent of all flights were operated under
capacity-constrained conditions (Gelhausen et al., 2013). Key measures
to mitigate capacity constraints at airports include increasing capacity by
adding new runways, reorganization of traffic either to off-peak times or
less-congested airports and using aircraft with higher seat capacity. The
most straightforward option is investing in additional runways to increase
airport capacity. However, in particular at large airports this requires
lengthy planning procedures because of the involvement of the public,
who are typically opposed to airport capacity enlargements, especially
in highly developed countries. These planning procedures can take up to
between 10 and 20 years at large hub airports, and in some cases adding
new runways to an airport is an insurmountable task. Examples include
Frankfurt and Munich airports and London Heathrow. Long-term capac-
ity bottlenecks are therefore to be expected, especially at some large hub
airports (Gelhausen, 2013).
The other three options rely more on airline behaviour to mitigate the
effects of airport capacity shortages. However, reorganizing traffic to off-
peak times might conflict with the airlines’ wave system of transfer con-
nections and at some airports, like London Heathrow, there is a shortage
of available slots during most daytime hours. Moving flights with a high
share of transfer passengers to less congested airports is unfavourable
from the viewpoint of economics of density, while this might be an option
in the case of flights dominated by OD passenger volumes. Employing
larger aircraft is a trend that has been observed for a while at both large
and small airports. However, employing larger aircraft is not only caused
by airport congestion, but also – and mainly – because average flight
length has increased over time. Here, technical and economic reasons play
an important role. Yet this trend threatens airports of low-demand routes
(Berster et al., 2015).
GLOBAL TRAFFIC GROWTH SINCE 2000 AND BY

REGION
Between 2000 and 2014, the number of global take-offs at airports

increased from just under 25 million to more than 31 million. In 2007, the
year the global financial crisis began, global take-offs totalled almost 30
million; that is, the growth of flights decreased considerably thereafter.

M4371-FINGER_9781786431851_t.indd 268 14/11/2017 11:14

The compound annual growth rate of global flights over the period was
1.70 per cent, which means that flights grew on average by 1.70 per cent
a year. However, if we split the period into two sub-periods (2000–07 and
2007–14) we detect a rapid decline of the compound annual growth rate:
while the compound annual growth rate is 2.44 per cent from 2000 to 2007,
it decreases to only 0.97 per cent from 2007 to 2014. Looking at the cor-
responding annual growth rate of global gross domestic product, we find
a close relationship: compound annual growth rate of global GDP being
2.79 per cent from 2000 to 2014, rises to 3.43 per cent from 2000 to 2007
and then decreases to 2.15 per cent from 2007 to 2014 (World Bank, 2016).
Turning to the traffic distribution between world regions in terms of
take-offs, North America was still the largest market in 2014, but its global
share fell throughout from 48 per cent to 31 per cent. This was mainly
due to a stagnating US market. On the other hand, the share of the Asian
market more than doubled: in 2000, 11 per cent of global take-offs were
handled at Asian airports, but this share increased to 18 per cent in 2007
and 25 per cent in 2014. The third-largest market in 2014 was Europe,
and its share fluctuated between values of 23 per cent (2000) and 25 per
cent (2007). The remaining four regions had a fairly small share of global
flights; however, the airports in Africa and in particular the Middle East
showed a continued positive traffic growth.
Considering the compound annual growth rate for take-offs by world
region between 2000 and 2014, as well as the sub-periods 2000–07 and
2007–14, with the exception of the rather small market of South America,
we find a stagnating or a decreasing growth rate across all world regions.
While growth rates remain positive in four out of seven regions from
2007 to 2014, we find a negative trend in Europe and North America.
Correspondingly, there is weak economic growth in North America and
the European Union: while GDP grew by 2.45 per cent and 2.29 per cent a
year on average during the period from 2000 to 2007, these rates fell to 1.01
per cent and 0.12 per cent a year for 2007 to 2014.
In summary, air traffic growth was strong in most world regions from
the year 2000 until 2007; however, growth rates have declined considerably
since then. This appears in many cases to be due to weak economic growth
since the global financial crisis in 2007, as GDP is typically a main driver
of air traffic development (IATA, 2008). In particular, well-developed
markets such as North America and Europe suffer from restricted growth,
while emerging markets such as Asia and the Middle East are still going
strong, which again coincides with economic development. However, in
the case of the US market a number of airline mergers due to economic
problems also play an important role in explaining the traffic development
there.

M4371-FINGER_9781786431851_t.indd 269 14/11/2017 11:14

TRAFFIC CONCENTRATION BETWEEN AIRPORTS

GLOBALLY AND BY REGION
Figure 14.1 shows the Lorenz curve of global take-offs for 2000, 2007 and
2014. Airports are ranked in descending order relating to their number of
flights in the appropriate year and their shares are cumulated so that the
last value ends up at 100 per cent. The curve is based on more than 3000
airports for 2000, 2007 and 2014 that have at least one scheduled flight
a year. Concentration of traffic at very large airports, and especially at
hubs, is large: in 2000 there were 3076 airports and the value of the Gini
coefficient was 0.807, in 2007 with 3258 airports it was 0.815, and in 2014
with 3349 airports it was 0.812. Thus, traffic concentration increased only
marginally on a global scale. The largest four airports in 2014 were in the
US (Chicago O’Hare, Atlanta Hartsfield-Jackson, Dallas/Fort Worth and
Los Angeles), followed by Beijing Capital airport. The largest European
airport is London Heathrow, in ninth place. The world’s 30 largest air-
ports handled more than 21 per cent of the global take-offs, and 50 per
cent of the flights globally were handled by the 120 largest airports in
2014.
If we limit the sample to airports that have more than 1000 take-offs
100
90
80
Share of global take-offs (%)
70
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100
Share of airports (%)
2000 2007 2014
Source: Based on OAG (2015).
Figure 14.1 Lorenz curve for take-offs at airports worldwide

M4371-FINGER_9781786431851_t.indd 270 14/11/2017 11:14

Table 14.1 Gini-coefficients and number of airports by region for 2014
World Region Number of Airports (2014) Gini-coefficient (2014)

North America 880 0.847
South America 454 0.734
Europe 622 0.780
Asia 684 0.789
Middle East 104 0.805
Southwest Pacific 298 0.824
Africa 307 0.742
World 3349 0.812
Source: Based on OAG (2015) and Sabre ADI (2015).
a year, concentration of traffic at a small number of airports is still very

high: the values of the Gini-coefficients for 2000, 2007 and 2014 are 0.729,
0.735 and 0.738, respectively. The number of airports that have more
than 1000 flights a year is 1945, 1986 and 2151, respectively. While traffic
concentration increased very little between 2000 and 2014, the number
of airports, and airports with more than 1000 annual take-offs, increased
yearly by 8 per cent and 11 per cent, respectively. Thus, the results remain
basically unchanged; hence we will stick with the existing sample of air-
ports for the time being.
Table 14.1 summarizes the results for the airports of each region and
their corresponding value of the Gini-coefficient for the year 2014. As
described earlier, the global average value of the Gini-coefficient is 0.819
for the year 2014. Values fluctuate for 2014, from 0.847 for North America
to 0.734 for South America. The Middle East and Southwest Pacific
regions also show high values of the Gini-coefficient, at 0.805 and 0.824,
respectively. On the other hand, Africa and South America have a lower
concentration of traffic. Europe and Asia are about in the middle of the
pack compared with the other regions. As we will see later in more detail,
regions with higher Gini-coefficients – that is, higher traffic concentration
– tend to have higher shares of airline alliance traffic, while higher shares
of LCC and regional carriers tend to have damped traffic concentration.
An exception to the rule is the Middle East, with a high share of airlines
not belonging to an airline alliance (such as Emirates), but they tend to
hub their flights, too. Europe is another case, with a high share of airline
alliance traffic and a high share of LCC traffic, resulting in a value of the
Gini-coefficient that tends to be in the middle of the pack compared with
other world regions. The North American region has not only the highest

M4371-FINGER_9781786431851_t.indd 271 14/11/2017 11:14

0.86
0.84
0.82
y = 4E-05x + 0.7673
R2 = 0.0841
0.80
0.78
0.76
0.74
0.72
0 100 200 300 400 500 600 700 800 900 1000
Gini-coefficient (2014) Linear (Gini-coefficient (2014))
Figure 14.2 Correlation analysis between number of airports in a world

region and the corresponding value of the Gini-coefficient
degree of traffic concentration, but also the highest number of airports.

A simple correlation analysis shows that there is only a very weak posi-
tive (Figure 14.2) but not statistically significant relationship between the
number of airports of a region and their Gini-coefficient value. However,
the sample size is very small for conclusions of statistical significance and
is presented only for illustrative purposes.
These findings seem to support the hypothesis that hubbing, and there-
fore a higher concentration of traffic, increases the viability of routes and
thus airports, as in the case of North America. However, these results
should not be overstated, because these seven world regions are heteroge-
neous with regard to geographical, economic and population issues, for
instance.
Comparisons with the various regional Lorenz curves show that in
emerging markets traffic concentration increases with traffic growth. This
is especially evident for the Middle East, South America and Asia regions,
while traffic concentration remains largely unchanged in more mature
markets such as North America and Europe. However, exceptions to the
rule are Africa and the Southwest Pacific: there has been strong traffic
growth in the past in Africa, but traffic concentration has not changed
much, while we see an increase in traffic concentration in the Southwest

M4371-FINGER_9781786431851_t.indd 272 14/11/2017 11:14

Pacific region, although traffic has decreased over the past 14 years. We
now briefly discuss the Lorenz curve for each world region.
The Lorenz curve for North America comprised 880 airports for 2014,
which is the highest number of airports and, with a Gini-coefficient of
0.847, it exhibits the greatest degree of concentration of any world region.
However, concentration increased little between 2000 and 2014, although
take-offs decreased by 1.30 per cent a year. South America, which had 454
airports in 2014 and is one of the smaller world regions, has the lowest
degree of concentration of any of the seven world regions, with a Gini-
coefficient of 0.734. Concentration increased between 2000 and 2014, and
especially before 2007. Furthermore, take-offs increased by 1.75 per cent a
year between 2000 and 2014, and even by 3.31 per cent a year from 2007 to
2014. However, concentration increased less between 2007 and 2014. The
622 European airports in 2014 was the third-largest number in the world
and, with an associated Gini-coefficient of 0.780, was in the middle of the
global range. This would seem to reflect the high degree of airline allianc-
ing in the region and large LCC market. The airline alliances prefer hub
airports for their operations but LCCs tend to choose regional airports,
although they are increasingly offering flights from larger or some hub
airports. As a result, concentration remained roughly constant between
2000 and 2014. There has been a significant rise in take-offs, however, of
3.58 per cent a year between 2000 and 2007, although this stagnated over
the next seven years.
There were 684 airports in Asia in 2014, making this the second-largest
of the global regions. The degree of concentration is akin to Europe’s with
a Gini-coefficient of 0.789, with concentration increasing between 2000
and 2007 as traffic grew annually by 10.31 per cent, although the concen-
tration declined in subsequent years, with growth in take-offs falling to
6.13 per cent a year.
The Lorenz curve for the Middle East, which with just 104 airports is by
far the smallest, reflects a mid-range concentration with a Gini-coefficient
of 0.805. Concentration increased between 2000 and 2007, which was
accompanied by significant annual traffic growth of 6.37 per cent, and
even greater concentration increased thereafter, with traffic growth of 6.63
per cent a year.
The Southwest Pacific had 298 airports for 2014, the second-smallest in
the world. However, associated with a fall of 0.18 per cent a year in take-
offs between 2000 and 2014 concentration increased consistently, and for
2014 the region exhibited the second-largest Gini-coefficient (0.824).
Africa in 2014 comprised 307 airports and is similar in size to the
Southwest Pacific region. Its take-offs increased by 4.11 per cent a
year over the period from 2000, but the concentration remained almost

M4371-FINGER_9781786431851_t.indd 273 14/11/2017 11:14

c onstant. It had a Gini-coefficient of 0.742, the second-lowest coefficient

among the seven world regions. The low degree of concentration seems to
have been the result of less-developed markets and airlines, and of gener-
ally much poorer aviation infrastructure than most other areas.
As an intermediate result, concentration of traffic is very high on a
global level. However, traffic concentration varies considerably between
different world regions and depends on factors that include rather regional
circumstances. However, the share of airline alliance, LCC and regional
airline traffic seems to play an important role: airline alliances depend more
on hub traffic than LCC and regional airlines, which are more focused on
smaller and regional airports. However, this could change in the future
if LCCs increasingly offer flights from hub airports, as for example the
new strategy of Ryanair in Europe strongly suggests (Independent, 2016).
On the other hand, traffic growth seems to be less suitable for explaining
traffic concentration in various cases, especially in more mature markets.
TRAFFIC STRUCTURE DEVELOPMENT SINCE 2000

GLOBALLY AND BY REGION
The process of airline liberalization has fostered two very different airline
business models: on the one hand, there are the full-service network carri-
ers (FSNCs) with a complex hub system, so that airlines enjoy economics
of density, making thin routes viable, especially in long-haul travel. On
the other hand, there are the LCCs with a less complex business model,
mainly serving point-to-point routes on denser routes. At the beginning
of liberalization, LCCs mainly used secondary and regional airports,
but in recent years they have also been serving larger airports as well
as hub airports that are closer to agglomerations. This development is
supported by the tendency of FSNCs to establish their own, largely inde-
pendent LCC subsidiary, such as Germanwings/Eurowings in the case of
Lufthansa, Vueling (International Airlines Group, IAG) and Transavia
(Air France/KLM). Membership in an airline alliance is a big competitive
advantage, giving airlines access to larger markets. Nevertheless, there
are still large carriers, such as Emirates, that do not belong to an airline
alliance.
In terms of these business models, while the shares of global take-offs
of FSNCs not belonging to any airline alliance decreased rapidly between
2000 and 2014, airlines of the three strategic alliances and LCCs grew
considerably. The share of regional and charter airlines decreased slightly
during this period.
The positioning and success of the carrier types vary across various

M4371-FINGER_9781786431851_t.indd 274 14/11/2017 11:14

30
2000
2007
Share of take-offs at European airports (%)
25 2014
20
15
10
0
N t
ce
rld
er
t
na
FS den
gh
LC
ea
rt
an
C
wo
io
ei
ha
yt
n
lli
Fr
eg
ne
pe
Sk
C
A
R
O
de
ar
In
St
Figure 14.3 Traffic structure development between 2000 and 2014 at

European airports
geographical regions. Increasingly former independent FSNCs have joined

one of the three remaining airline alliances over time, so that their share of
flights decreased rapidly in all world regions between 2000 and 2014. The
largest airline alliance varies from region to region; for example, Europe
is dominated by Star Alliance (Figure 14.3), while Oneworld in Asia has
the largest share of flights of all airline alliances (Figure 14.4). Shares of
take-offs are split more evenly between the three alliances (Star Alliance,
Oneworld and Skyteam) in North America (Figure 14.5). There is a com-
paratively large share of independent FSNCs at Middle East and African
airports; this can be attributed to individual airline strategies in the case of
the Middle East, with Emirates as an example, and less-developed markets
and airlines in the case of Africa. The South American, Asian and African
markets are still characterized by a relatively large share of regional carri-
ers. With the exception of the Middle East and Africa, LCCs have rapidly
increased their share of flights.

M4371-FINGER_9781786431851_t.indd 275 14/11/2017 11:14

60
2000
2007
Share of take-offs at Asian airports (%)
50 2014
40
30
20
10
0
ld
N t
ce
er
t
na
FS den
gh
LC
r
ea
rt
an
C
wo
io
ei
ha
yt
n
lli
Fr
eg
ne
pe
Sk
C
A
R
O
de
ar
In
St

Asian airport
AIRPORT CAPACITY UTILIZATION AND CAPACITY

CONSTRAINTS
Due to traffic concentration and the need for synchronization of incom-

ing and outgoing flights to provide attractive stopover connections, high
levels of capacity utilization and capacity constraints are more likely to
appear at hubs than at secondary airports, which typically have high shares
of point-to-point traffic. However, there are various measures to mitigate
capacity constraints, such as shifting traffic to less congested hours (‘peak
hour pricing’), shifting traffic to less congested neighbouring airports,
employing larger aircraft and the investment option: building new runways
or terminals.
While the last option seems to be the most attractive one, because it
does not impair the scheduling of airlines and the hub function (that is,
generating stopover connections), it is difficult to implement in some
cases: in economically highly developed countries with a high population

M4371-FINGER_9781786431851_t.indd 276 14/11/2017 11:14

40
Share of take-offs at North American airports (%)
2000
35 2007
2014
30
25
20
15
10
0
t
ld
ht
na
FS den
rte
nc
LC
or
ea
ig
C
io
ha
lia
e
w
N
n
yt
eg
Fr
ne
pe
Al
C
Sk
R
O
de
ar
St
In

North American airports
density and level of participation, airport development plans take a long

time to implement in many cases, since the population surrounding the
airport typically opposes such plans because of noise emissions and pollu-
tion. For example, adding a new runway can take more than ten years (as
with Frankfurt airport, for example) or even be an insurmountable task.
Nevertheless, while large hub airports generally seem to be prone to con-
gestion because of traffic volumes, the level of airport capacity constraints
varies from region to region and depends even more on peripheral condi-
tions of an airport, that is, the level of noise exposure and pollution of the
population in the neighbourhood of an airport. This is true for very large
airports in terms of take-offs (Gelhausen, 2013).
We have analysed the congestion level of the 30 largest airports glob-
ally by means of the capacity utilization index (CUI): see Table 14.2. The
CUI is based on so-called ranking curves (Wilken et al., 2011): for each
airport, hourly traffic volume of a year is ranked in descending order.
From the ranking curve of an airport, the 5 per cent peak hour volume
and the adjusted mean hour volume are derived (Gelhausen et al., 2013;

M4371-FINGER_9781786431851_t.indd 277 14/11/2017 11:14

Table 14.2 Take-offs and CUI values of the largest 30 airports in 2014
Rank Airport name IATA Code Region Take-offs CAGR CUI (2014)
M4371-FINGER_9781786431851_t.indd 278
(2014) (2000–2014)
1 Chicago O’Hare International ORD North America 446 213 –0.23% 0.66
2 Atlanta Hartsfield-Jackson International ATL North America 427 603 –0.23% 0.64
3 Dallas/Fort Worth International DFW North America 333 328 –1.47% 0.71
4 Los Angeles International LAX North America 299 946 –1.73% 0.69
5 Beijing Capital PEK Asia 292 313 9.22% 0.75
6 Denver International DEN North America 276 065 0.84% 0.60
7 Charlotte Douglas International CLT North America 257 002 1.93% 0.74
278
8 Houston George Bush Intercontinental IAH North America 243 278 0.48% 0.65
9 London Heathrow LHR Europe 236 295 0.13% 0.84
10 Frankfurt International FRA Europe 230 031 0.60% 0.70
11 Paris Charles de Gaulle CDG Europe 219 016 –0.70% 0.69
12 Amsterdam A MS Europe 211 795 0.62% 0.62
13 Tokyo Haneda HND Asia 211 649 3.87% 0.68
14 Istanbul Ataturk 1ST Europe 209 395 8.71% 0.75
15 New York J F Kennedy International JFK North America 208 964 0.82% 0.68
16 Shanghai Pudong PVG Asia 205 541 16.98% 0.40
17 San Francisco International SFO North America 203 656 0.03% 0.67
18 Guangzhou CAN Asia 203 159 9.51% 0.75
19 Toronto Lester B Pearson International YYZ North America 201 002 0.24% 0.62
20 Jakarta Soekarno-Hatta CGK Asia 200 371 10.45% 0.71

14/11/2017 11:14
M4371-FINGER_9781786431851_t.indd 279
21 Philadelphia International PHL North America 197 791 –0.52% 0.70
22 Detroit Wayne County DTW North America 191 413 –1.59% 0.60
23 Newark Liberty International EWR North America 190 801 –0.88% 0.70
24 Phoenix Sky Harbor International PH× North America 189 229 –1.64% 0.67
25 Minneapolis International MSP North America 188 906 –1.46% 0.57
26 Hong Kong International HKG Asia 188 539 5.69% 0.69
27 New York La Guardia LGA North America 187 098 –0.45% 0.76
28 Mexico City International ME× South America 185 964 2.22% 0.68
279
29 Munich International MUC Europe 177 107 1.66% 0.60
30 Kuala Lumpur International KUL Asia 173 575 4 8.83% 0.67
North America 4 042 295 0.665
South America 185 964 0.680
Europe 1 283 639 0.704
Asia 1 475 147 0.668
Top 30 airports 6 987 045 0.674

14/11/2017 11:14
Reichmuth et al., 2011). Therefore, the CUI is the ratio of the mean hour
volume divided by the 5 per cent peak hour volume. Airports with CUI
values in the range of more than 0.65 to 0.70 can be considered as con-
gested already, while values of more than 0.80 indicate very high levels of
congestion. However, everything else being about equal, Asian airports
typically tend to have slightly higher CUI values compared with other
regions because traffic tends to be more evenly distributed.
Table 14.2 shows the world’s 30 largest airports with their number of
take-offs and CUI value for 2014, as well as the compound annual growth
rate between 2000 to 2014. The top 30 are dominated by North American
airports; however, while many North American and European airports
basically stagnated between 2000 and 2014, Asian airports grew strongly
during the same period. The top 30 airports have an average CUI value
(weighted by take-offs for 2014) of 0.67, along with the North American
and Asian airports (CUI values were rounded by two decimal places). The
European airports have on average a CUI value of 0.70, and the average
CUI value of South America only includes Mexico City International
airport. However, on the airport level, peripheral conditions play a major
role: for instance, London Heathrow and New York LaGuardia have CUI
values of 0.84 and 0.76. These are large airports close to large agglomera-
tions with typically high noise and pollution emissions. Frankfurt airport
opened the fourth runway in late 2011 and traffic has slightly decreased
since then, so that the capacity situation has eased. In 2012, Frankfurt
airport had a CUI value of 0.77.
Overall, the relatively weak traffic growth of recent years in Europe
during the Great Recession, but in particular in North America, has helped
to prevent an even tighter capacity situation at many large airports. On
the other hand, emerging regions such as the Middle East and Asia have
overall traffic levels that are still lower than North America and Europe
and airport expansion plans are easier to implement in case of capacity
bottlenecks (e.g. Beijing and Dubai airports). However, the capacity situ-
ation at some airports may worsen again if traffic growth rates increase in
the long term. Then again, hubbing helps to utilize overall airport capacity
more efficiently.
CONCLUSIONS
In this chapter we have analysed the traffic growth at airports globally

between 2000 and 2014. Furthermore, we have analysed the traffic growth
by seven world regions, focusing on traffic growth and traffic distribu-
tion between airports. Overall, traffic growth has decreased considerably,

M4371-FINGER_9781786431851_t.indd 280 14/11/2017 11:14

especially since 2007. This ties in with weak economic development glob-
ally, but especially in highly developed regions such as North America and
Europe. As a result, traffic growth has also differed markedly between
regions: while highly developed world regions grow only slowly or are
even decreasing, emerging regions such as Asia and the Middle East are
growing strongly. However, these are regions whose economies have also
developed quite dynamically in recent years. Nevertheless, the largest
market remained North America in 2014, but Asia replaced Europe as the
second-largest market in terms of number of take-offs in 2014.
Traffic concentration at airports remains at a very high level globally:
the value of the Gini-coefficient for 2014 is 0.812 and it only increased
marginally from 2000. There are some differences in traffic concentration
between different world regions; however, these differences are relatively
small: Gini-values ranged from 0.734 for South America to 0.847 for North
America. While concentration in highly developed regions remained
roughly the same between 2000 and 2014, concentration tended to increase
for emerging regions such as the Middle East and Asia. Furthermore,
airline structure tends to influence traffic concentration: the higher the
share of airline alliance traffic, the higher the traffic concentration of a
region. On the other hand, LCC and regional airlines tend to dampen con-
centration, since they are more focused on point-to-point traffic. During
the period from 2000 to 2014, airline alliances and LCCs achieved massive
gains in traffic share, while independent FSNCs, regional and charter
carriers either stagnated or decreased their traffic shares. Therefore, liber-
alization has fostered two very different airline business models: FSNCs
with their hub-and-spoke system on the one hand, and LCCs with mainly
point-to-point traffic on the other hand. Airline alliances accounted for
more than half of the global traffic in 2014, while LCCs had a share of
about 20 per cent of global take-offs. Furthermore, LCCs increasingly
try to arrange some degree of hubbing (‘self-help hubbing’), and they are
serving a greater number of larger and hub airports. This trend is fos-
tered by FSNCs establishing their own largely independent LCC subsidi-
ary, such as Germanwings/Eurowings (Lufthansa), Vueling (International
Airlines Group) and Transavia (Air France/KLM).
Hub airports are prone to congestion and capacity constraints, but this
is not a particular hub airport problem, as large airports with attractive
catchment areas close to dense agglomerations are generally more affected
than smaller airports that are remotely located. In these cases, airport
expansion plans are difficult to implement and take a long time, since
the surrounding population is typically opposed to such plans because of
further noise and pollution emissions, especially in economically highly
developed countries with a high level of participation. Furthermore, it

M4371-FINGER_9781786431851_t.indd 281 14/11/2017 11:14

is unlikely that a shift away from hubbing will resolve airport capacity
problems in the long term, since a pure point-to-point network needs more
flights to connect a given network.
REFERENCES
Berster, P., M.C. Gelhausen and D. Wilken (2012), Demand and supply develop-
ment patterns of Low Cost Carriers in Africa, America, Europe, Australia and
Asia, paper presented at the Air Transport Research Society World Conference
2012, Tainan.
Berster, P., M.C. Gelhausen and D. Wilken (2015), Is increasing aircraft size
common practice at congested airports?, Journal of Air Transport Management,
46, 40–48.
Burghouwt, G. and J. de Wit (2005), Temporal configurations of European airline
networks, Journal of Air Transport Management, 11, 185–98.
Button, K. (2002), Debunking some common myths about airport hubs, Journal of
Air Transport Management, 8, 177–88.
Button, K., K. Haynes and R. Stough (1998), Flying into the Future: Air Transport
Policy in the European Union, Cheltenham, UK and Northampton, MA, USA:
Edward Elgar Publishing.
Daft, J. and S. Albers (2012), A profitability analysis of low-cost long-haul flight
operations, Journal of Air Transport Management, 19, 49–54.
Dobruszkes, F. (2006), An analysis of European low-cost airlines and their net-
works, Journal of Transport Geography, 14, 249–64.
Ehmer, H., P. Berster, J. Basedow and C. Jung (2000), Liberalisierung im Luftverkehr
Deutschlands – Analyse und Wettbewerbspolitische Empfehlungen, Cologne:
German Aerospace Center.
Gelhausen, M.C. (2013), A model of airport capacity expansion delays, paper pre-
sented at the Air Transport Research Society World Conference 2013, Bergamo.
Gelhausen, M.C., P. Berster and D. Wilken (2013), Do airport capacity constraints
have a serious impact on the future development of air traffic?, Journal of Air
Goetz, A.R. and T.M. Vowles (2009), The good, the bad, and the ugly: 30 years of
US airline regulation, Journal of Transport Geography, 17, 251–63.
Grimme, W. (2011), The evolution of the low cost carrier business model –
connections, hubbing and interlining, presentation at the 5th Annual Airneth
Conference, The Hague.
Hooper, P. (2005), The environment for Southeast Asia’s new and evolving airlines,
IATA (2008), IATA Economics Briefing No 9: Air Travel Demand, Montreal: IATA.
Independent (2016), Frankfurt airport opens up to no-frills flights as Ryanair
moves in, 2 November, available at http://www.independent.co.uk/travel/news-
and-advice/cheap-flights-frankfurt-ryanair-budget-airlines-germany-holidays-
spain-brexit-a7393231.html.
Malighetti, P., S. Paleari and R. Redondi (2008), Connectivity of the European
airport network: ‘Self-help hubbing’ and business implications, Journal of Air

M4371-FINGER_9781786431851_t.indd 282 14/11/2017 11:14

Moreira, M.E., J.F. O’Connell and G. Williams (2011), The viability of long-haul,
low cost business models, Journal of Air Transport Studies, 2, 69–91.
Morrell, P. (2008), Can long-haul low-cost airlines be successful?, Research in
Transport Economics, 24, 61–7.
O’Connell, J.F. and G. Williams (2005), Passengers’ perceptions of low cost airlines
and full service carriers: a case study involving Ryanair, Aer Lingus, Air Asia and
Malaysia Airlines, Journal of Air Transport Management, 11, 259–72.
Official Airline Guide (OAG) (2015), Market Analysis, Dunstable: Reed Travel
Group.
Pels, E. (2008), Airline network competition: full-service airlines, low-cost airlines
and long-haul markets, Research in Transportation Economics, 24, 68–74.
Redondi, R., P. Malighetti and S. Paleari (2011), Hub competition and travel times
in the world-wide airport network, Journal of Transport Geography, 19, 1260–71.
Redondi, R., P. Malighetti and S. Paleari (2012), De-hubbing of airports and their
recovery patterns, Journal of Air Transport Management, 18, 1–4.
Reichmuth, J., P. Berster and M.C. Gelhausen (2011), Airport capacity constraints:
future avenues for growth of global traffic, CEAS Aeronautical Journal, 2, 21–34.
Reynolds-Feighan, A. (2001), Traffic distribution in low-cost and full-service
carrier networks, Journal of Air Transport Management, 7, 265–75.
Sabre Airport Data Intelligence (ADI) (2015), Data based on Market Information
Intelligence Tapes (MIDT), Southlake: Sabre Airline Solutions.
Wensveen, J.G. and R. Leick (2009), The long-haul low-cost carrier: a unique busi-
ness model, Journal of Air Transport Management, 15, 127–33.
Wilken, D., P. Berster and M.C. Gelhausen (2011), New empirical evidence on
airport capacity utilisation: relationships between hourly and annual traffic
volumes, Research in Transport Business & Management (Airport Management),
1, 118–27.
Wilken, D., P. Berster and M.C. Gelhausen (2016), Analysis of demand structures
on intercontinental routes to and from Europe with a view to identifying poten-
tial for new low-cost services, Journal of Air Transport Management, 56, 79–90.
Williams, G. (2001), Will Europe’s charter carriers be replaced by ‘no-frills’ sched-
uled airlines?, Journal of Air Transport Management, 7, 277–86.
The World Bank (2016), World Development Indicators, Washington, DC: The
World Bank.

M4371-FINGER_9781786431851_t.indd 283 14/11/2017 11:14

15. Market instability
Kenneth Button
Regulation has gone astray. . . Either because they have become captives of
regulated industries or captains of outmoded administrative agencies, regula-
tors all too often encourage or approve unreasonably high prices, inadequate
service, and anticompetitive behavior. The cost of this regulation is always
passed on to the consumer. And that cost is astronomical.
Senator Edward Kennedy, opening remarks to the Subcommittee on
Administrative Practice and Procedure, 6 February 1975.
INTRODUCTION
The stereotype markets of perfect competition and monopoly associated,

for example, with Alfred Marshall, and with little else between them, have
been the bread and butter of elementary economic courses for a century
and a quarter or more. In practice, while serving a useful purpose as
benchmarks, they provide few insights into the detailed workings of most
markets or much guidance when assessing the implications of changes to
institutional structures. This is as true for the various markets that, in com-
bination, serve air transportation as for markets more generally.
The main problem with the nineteenth-century models of market struc-
ture is that their assumptions rarely apply in – amongst other things – a
world of imperfect information, concentrations of both monopoly and
monopsony power, manifest indivisibilities in inputs that make marginal
analysis unrealistic and a reluctance of governments to refrain from
intervening in markets that they seldom fully understand. One can easily
argue that these imperfections existed even when neoclassical economics
emerged, and that the assumptions involved in simply ignoring them, or at
best recognizing them in disclaiming caveats, were simply a device to make
tractable the study of complex problems. But the one thing of importance
that they inevitably did do was to distract analysis away from the dynamics
of many situations, especially the path from one equilibrium to another,
and related to this, to ignore markets that are, from even the simplest
observation, far from stable.
In practice, all markets suffer from a degree of instability, and the indus-
284

M4371-FINGER_9781786431851_t.indd 284 14/11/2017 11:14

Market instability 285
tries that comprise the air transportation sector – airlines, airport and air
navigation systems – are no exception (Button, 2005b).1 The various forms
of instability in air transportation tend to ripple through time and across
markets in an interactive way that affects its component industries. In the
context of this chapter, these ripples are made more complicated because
of the nature of market liberalization that has taken place. Unlike many
other sectors characterized by several component industries, airlines, air-
ports and air navigation systems have historically largely been supplied in
individual markets and regulated separately. Furthermore, the liberaliza-
tions of air transportation markets that have occurred since the 1970s have
come at different times and in different ways. Given the network nature of
air transportation, the diversity of timings and natures of these liberaliza-
tions have, as we see below, added to the inherent instability of the sector.
This chapter provides some guidance as to the implications that liber-
alization of air transportation markets has had for both the suppliers and
their customers, particularly focusing on the challenges associated with
varied types of instability that have emerged. In doing this it draws both
upon empirical evidence and theoretical arguments to highlight the nature
of the problems, their scale and their underlying causes. It also looks at the
variety of ways the suppliers of air transportation have acted to mitigate
adverse effects of market instabilities, and the responses of governments
that often see instability of any kind as an adverse element of their desired
social welfare function.
THE LIBERALIZATION OF AIR TRANSPORTATION

MARKETS
The writing examining the liberalization of the world’s various air trans-
portation markets is gargantuan. Some of the more important work can
be found in Levine (1987), Kahn (1988), Bailey (2010) and Morrison
and Winston (1995). Here we limit the description to a few paragraphs
as a brief aide-memoire, or as a simple introduction as the case may be,
for readers. It is far from comprehensive, but further analysis is readily
available.
Since its inception, the airline industry has been subject to economic
regulation, the first being the patent right afforded to the Wright broth-
ers for their method of steering a heavier-than-air flying machine.
Following that, regulations grew at the national and, later, international
levels. Governments sought ways to make use of civilian air transporta-
tion for mail services initially, and subsidies became common, and then
to provide the infrastructure for aviation with direct involvement in

M4371-FINGER_9781786431851_t.indd 285 14/11/2017 11:14

airport and air traffic control supply for local development and safety
reasons. The 1944 Chicago Convention drew up the framework for
agreements on international air transportation supply by creating a basis
of bilateral international negotiations and establishing an institutional
structure to oversee it (the International Civil Aviation Organization).
This, while it was not the intent, led to a mass of restrictive bilateral
national agreements that limited capacity on each route, the fares
charged, the carriers that could operate and often the ways that revenues
were split. The institutional regime had the potential for creating wider,
multilateral structures, but national self-interest and an inclination for
protecting national carriers prevented a more global perspective being
taken by individual states.
National domestic regimes varied in the regulatory structures they
imposed, with most giving monopolies to generally subsidized state-owned
carriers whilst at the same time controlling fares and routes. The US was a
notable exception to this in terms of ownership. Fares were inevitably regu-
lated. There was some flexibility for charter carriers offering unscheduled
services, but these had to conform to stringent criteria, for example, to be a
component of a package vacation, with the travel company de facto hiring
the entire aircraft.
While the rationale for regulations varied between countries and over
time, there were some common features. One was quasi-military in that a
country wished to have adequate air-lift at times of national emergency;
another was that countries often saw their international carriers, even if
not fully efficient, as a source of foreign exchange earnings (Button, 2010).
In larger countries, such as the US and Canada, air transportation was
viewed as an instrument of national political integration and, as such,
serving a key public interest.
Liberalization was initially introduced in 1977 for US inter-state freight
aviation and in the following year for passenger airlines as a way of reduc-
ing costs in the sector and inducing greater innovation in technology and
management. It was gradually taken up at the international level, nudged
along by the US’s Open Skies policy and the Three Packages of reforms
in the European Union in the 1990s that were part of a much wider Single
European Market initiative, as well as domestically by individual states.
There were also changes in the ownership of airlines in many countries,
as privatization of former state-owned carriers occurred; with this, there
has been a shift away from state subsidization, except for strategic reasons
(as with the US Civil Reserve Air Fleet) or in the case of social policy, and
even in the latter case competitive tendering has become the norm to make
more efficient use of public funds.
The liberalization of aviation infrastructure has been much slower, which

M4371-FINGER_9781786431851_t.indd 286 14/11/2017 11:14

in part can be explained by the more complicated legal process normally

required. In many countries airports have been made more commercial
with – to use the Canadian term – corporatization, essentially becoming
either public or private not-for-profit companies. In other cases, such as
in the UK, they have been privatized but under an umbrella of price cap
regulation (Button, 2007). A similar picture has emerged in the case of
air navigation service providers, with many countries following the New
Zealand example of setting up a not-for-profit government corporation,
although some have been private corporations. The UK is something of
an exception, with a for-profit, price-capped concession structure (Button
and McDougall, 2006).
It has generally been agreed that the outcome of these reforms has been
largely beneficial, and in economic terms would pass a Hicks–Kaldor com-
pensation test, with beneficiaries being able to compensate those who lost
out. Fares in real terms have fallen in most markets, and customers have
a far wider portfolio of fares/service combinations to choose from than
before. The advent of the low-cost carrier has added to the range of ser-
vices being offered, along with pulling down the average cost of flying for
the general public. The number of services has expanded, and with greater
flexibility of network design available to carriers; passengers generally now
have a much greater choice of routings and timing of flights. And this has
been achieved with no evidence of any decline in safety standards, a major
concern of many at the time of reform.
MARKET INSTABILITY DUE TO TRANSITION

PROCESSES
While it is broadly accepted that market liberalization has, at least in terms

of comparative statics, led to overall welfare gains, the dynamics of the
situation have often been bumpy, with markets in the short-term appearing
unstable. This type of instability, which has not been uncommon in other
sectors as liberalization has taken place, is an almost inevitable outcome of
a range of market imperfections and can vary in its intensity according to
the nature of the liberalization process (Meyer and Tye, 1981). The post-
liberalization path of any sector takes a J-curve form, with an immediate
adverse effect on the industry as it incurs costs of change as transmogrifi-
cation takes place, but then sees a subsequent upturn as the benefits of the
new situation materialize.
There are inevitable fixed costs in any transportation system that differ
with the role of the particular element under review and the technology
involved; for example, a larger part of an airport’s costs are fixed compared

M4371-FINGER_9781786431851_t.indd 287 14/11/2017 11:14

to an airline’s, where labour and fuel are the main cost considerations. This
inevitably means that it takes longer to adjust to a new institutional struc-
ture when a large part of a business’s costs are fixed. This is one reason
why airline markets were liberalized before airports and air navigation
service providers. But there is another element to this: some costs are sunk
and have to be left ‘stranded’ if an undertaking changes its activities; for
instance, an airport runway has few alternative uses if airlines take their
operations elsewhere (Baumol and Willig, 1981).
The challenge with airline liberalization is that there was little prior
experience of how a market-oriented air transportation system would
operate, or how long it would take for adjustments to take place. Indeed,
the end of the Civil Aeronautics Board in 1985 after carrying through the
reforms of the 1978 Airlines Deregulation Act was the first major regula-
tory body of any form to be disestablished in the US. While there were
some general ideas of what would occur, there were also surprises (Kahn,
1988). For example, the greater use of hub-and-spoke networks was antici-
pated once airlines were free to choose the routes and airports they would
serve, but the extent and the rapidity of their adoption was not known,
leaving companies like Braniff International Airways in the US with an
inappropriate fleet and network, resulting in bankruptcy. Other companies
experimented with route, fleet and service combinations as they sought to
fill the information gap and this was costly both to them and to their pas-
sengers as the market adjusted.
The approaches to handling short-term problems of adjustment insta-
bility stemming from institutional changes have differed across markets.
The US approach with its liberalization of air cargo transportation in 1977
and passengers in 1978 was to initiate the change rapidly and with very
little phasing in, expecting large short-term disruption followed by a fairly-
rapid move to a significantly higher equilibrium level of welfare: the ‘big-
bang approach’ (Button and Johnson, 1998). In contrast, although it was
probably more out of political expediency than economic efficiency, the
European Union adopted a phased approach to liberalization, bringing
in three packages of reforms in 1987, 1990 and 1992, with some phasing
in of each. The idea of this is to give the European carriers time to adjust
their operations and thus to minimize their stranded costs. However, the
challenge of this gradual approach is the possibility of the carriers captur-
ing the system and stymying the effectiveness of the change by slowing
or modifying it at each stage to their advantage, the extended use of
European restructuring aid being an example of this.

M4371-FINGER_9781786431851_t.indd 288 14/11/2017 11:14

MARKET SHOCKS AND INSTABILITY
Instabilities can occur if there are intermittent external shocks to the

market. In the case of air transportation these shocks initially impact the
airline industry and are then passed down to airports and air navigation
providers. The initial shocks may also be fairly systematic, such as those
tied to broader business cycles, or they can be unique events such as that
triggered by the attacks on the World Trade Center in 2001, or the SARS
pandemic of 2002/2003.
The demand for air transportation – be it by passengers or freight
consigners – is derived from the demand for some ultimate act of con-
sumption; as with most forms of transportation, air transportation is
seldom demanded for its own consumption. Put simply, people fly to
leisure resorts to enjoy a vacation, and to business districts to pursue their
employment, while freight consigners want to move inputs into the pro-
duction process or to their final customer. But in doing this their demand
levels and patterns change over time; manifestly this is so seasonally for
leisure travel or in the movement of Christmas gifts, but there are also
longer business cycles, and most notably the 7- to 11-year Juglar Cycle,
which affect the viability of the air transportation industries.
We see this effect, for example, in the operating margins of airlines,
although they can equally be detected in passenger numbers, load factors
of aircraft and airline bankruptcies that systematically move up and down
across time in pretty-regular patterns. Figure 15.1 shows not only the
extremely low average operating margins of airlines since the late 1980s,
but also the systematic shifts from low positive margins to negative ones.
While the data is not always strictly compatible because the composition
of airlines included in each year varies as some leave and new ones enter,
and because not all commercial airline categories – especially low-cost and
regional carriers – are always included, differences in regional markets are
apparent. In particular, there are greater extremes seen in the US than in
Europe.
Market stability is also affected by unique or very rare events. In the
case of the airlines, for example, the downturn in their margins in the early
2000s coincided not only with a cyclical macroeconomic decline in most
national economies, but also with the effects of the September 11th attacks
in New York and Arlington, and of the SARS pandemic. Because of the
use of commercial aircraft as weapons, these US attacks had a magnify-
ing effect on the industry, which was probably more pronounced than any
other shock in peacetime and generated major ripple effects that seriously
affected other elements of the air transportation supply chain.
In Figure 15.2, for example, we see the charges levied by a variety of

M4371-FINGER_9781786431851_t.indd 289 14/11/2017 11:14

10
Europe
United States
8
Global
M4371-FINGER_9781786431851_t.indd 290
4
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
–2
290
–4
–6
–8
–10
Notes:
1. A lack of a bar indicates a missing observation and not a zero operating margin.
2. Memberships of the various reporting bodies vary over time and thus the reported margins reflect the associated carriers at the time of
reporting.

Sources: Boeing Commercial Airplanes, Association of European Airlines, Airlines for America, International Air Transport Association.
Figure 15.1 Operating margins of airlines (1988–2014)
14/11/2017 11:14
SkyGuide NATS U.K. Airservices Australia

Irish Aviation Airways NZ DSNA
DFS LVNL
140
130
120
First year of data = 100
110
100
90
80
70
60
50
1997 1998 1999 2000 2001 2002 2003 2004
Note: All data indexed to 100 based on 1997 except for NATS UK 20015100.
Source: Button and McDougall (2006).
Figure 15.2 En route unit rates levied by air service navigation providers
for en route and terminal services in constant 2004 prices
air navigation service providers for the period around 2001. The pro-
viders vary in their corporate structure: some (such as NATS UK, the
British supplier) were purely commercial undertakings, while others
(such as Airways New Zealand and Airservices Australia) were govern-
ment corporations or government agencies (such as LVNL, the Dutch
provider), but they all exhibit an upturn in the rates they levy for a period
after 2001.
The reasons for the differences are in part due to the institutional nature
of the undertaking involved, the reaction of governments (a matter we
return to later) and the degree to which their local markets were affected by
the events in New York and Arlington. In particular, many of the privat-
ized or corporatized systems had the remit of cost recovery often over a
period as short as one year, basically implying a zero rate-of-return regula-
tion. In consequence, they did not have sufficient reserves to cover their
extensive fixed costs as airline traffic fell after 2001 and they had to raise
their fees. This is a situation running somewhat counter to normal market

M4371-FINGER_9781786431851_t.indd 291 14/11/2017 11:14

logic, whereby one would have lowered fees to attract more traffic after the
September 11th attacks.
We could further extend this descriptive analysis by looking at cycles
in airline bankruptcies and mergers, the geographical pattern and size of
airport hubs, air service withdrawals, market entry, and so on that accom-
pany economic recessions or sudden, unsystematic shocks to the aviation
market on either the demand or the supply side. But these sorts of events
are features that have counterparts across many sectors, and with some
minor peculiar characteristics fall within the generic umbrella of industrial
economics rather than being air-transportation-specific.
THEORIES OF INHERENT MARKET INSTABILITY
There are also – and of more interest – market structures that can be inher-
ently unstable simply because of their intrinsic characteristics, a point
made over 125 years ago but often overlooked by more recent economists
(Edgeworth, 1881).2 These markets are essentially competitive in nature,
but many of the assumptions of Marshall’s perfect competition do not
apply. There may be an ‘empty core’ with no equilibrium when there are
relatively large fixed costs, avoidable (set-up) costs, indivisibility, network
effects or severe fluctuations in demand. An unsustainable market may
also exist when some suppliers enjoy a degree of institutional or financial
protection and when there are significant variations in the costs of sup-
pliers. In practice, many public utilities, transportation industries and
some manufacturing industries meet these criteria. A general survey of
the theory is found in Telser (1978), with Button (2003; 2005b) linking its
relevance specifically to airline markets.
While developed in the context of a time when manufacturing concerns
dominated thinking about market structures, the idea of empty cores
is germane to the scheduled airline industry, not because there are high
physical fixed costs – although there may be – but because of the nature of
the service provided. An airline has to establish its schedule many months
in advance and, in doing so, commits to costs of having a plane, fuel, crew,
take-off slots, airport gates, check-in and baggage handling capacity, at
the very least, available at the time of take-off. These are the de facto fixed
costs of airline services.3 This is true even if the carrier is a virtual entity
that leases planes and crew; it has to have contracts in place. Because all
the competitors have similar fixed costs, there is a tendency to compete for
business; in effect, they compete to fill their seats by discriminate pricing
down to a marginal cost that is very low once a commitment to a schedule
has been made. Any revenues above this marginal cost can contribute to

M4371-FINGER_9781786431851_t.indd 292 14/11/2017 11:14

the fixed costs of the service. But with competition, the tendency is for
these extra marginal cost revenues to be small.
This can be illustrated – albeit with some rather strong assumptions
for pedagogic reasons – with a simple example based on Telser (1987).
Suppose that an economy has three individuals (B1, B2 and B3) who wish
to fly between two points. B1 is prepared to pay a fare of $100, B2 of $80
and B3 of $75 for the trip. There is one aircraft owned by A1 that can
carry up to two passengers. The plane can make the trip at a cost of $140
whether it carries zero, one or two passengers. All of the sustainable out-
comes have the property that B1 and B2 travel on the plane, whilst B3 fails
to make a contract. The outcomes can differ in the fare that B1 and B2 pay
for the journey. For example, the allocation in which B1 and B2 both have
a flight and pay a fare of $75 is in the ‘core’. In all the outcomes in the
core, airline A1 makes positive profits of at least $10. If the carrier lowers
its fare further, below $75, the third individual (B3) would also wish to fly,
potentially disturbing the coalition of A1, B1 and B2. There would be an
excess demand at this rate and hence the core would be empty. If another
identical aircraft, owned by A2, also with a maximum capacity of two
and the same costs as A1 enters the market, there would also be no core.
Because the two aircraft are identical and because equilibrium requires
their owners to receive the same profit, there is not enough demand to
sustain both planes. They would compete prices down until one is driven
from the market. But the remaining carrier can again make supernormal
profits attracting a new entrant, and so on.
One may suppose that businesspeople are not entirely myopic, and
potential entrants during the times of high profits would recognize the risk
they take in entering the market, but as Knight (1921) pointed out nearly a
century ago – and which has certainly not been disproved by the specula-
tive behaviour that continues to abound – where there are conditions of
risk there are winners and losers and there is always a tendency for some
optimists to be around.
The fact that this economic conundrum applies to the scheduled airline
industry has been appreciated for some time, but it has been largely ignored
in policy formulation. The complexity of the underlying economic model
has hindered the communication of the issue to decision-makers. The situ-
ation also runs counter to an often ingrained element of traditional, often
ideological views of competition policy that hold that there can ‘never be
too much competition’. The idea that there can be ‘too much competition’
is something of an anathema in such circumstances.

M4371-FINGER_9781786431851_t.indd 293 14/11/2017 11:14

EVIDENCE OF AN EMPTY CORE IN AIRLINE

MARKETS
Technically, it is impossible to prove that an empty core exists, and the

empirical problem extends beyond the wider ones of hypothesis testing.
To demonstrate the existence of an empty core and the degree to which it
leads to economic inefficiency, it is necessary to provide the counterfactual
of what the world looks like with a core, and such hypothesizing is fraught
with problems. Put another way, if there is an empty core then there is no
equilibrium market-clearing price, and thus there is no way of predict-
ing the optimal level of output. Consequentially, there is no way to test
whether actual output is below this level.
What one can do is look for indicators in the behaviour of airlines that
are suggestive of an empty core problem, although even here there are
challenges in markets that, as we have seen, also fluctuate for other reasons
and inevitably do so on occasions in sync with any core-based forces. For
example, Button (1996) and Antoniou (1998) both examined changes in
the European Union air transportation market to look for conditions
where there is the potential for an empty core to exist. Following the earlier
reasoning of Sjöstrom (1989) in the context of shipping markets, the 1990
cross-section data used indicated that the signs associated with variables
representing legal restrictions and market stability were consistent with
empty-core conditions. However, it should be emphasized that the findings
are indicative and not in any way definitive.
Another approach is to look at the pricing behaviour of airlines under
different market conditions. One of the features of liberalization has been
the extensive dynamic price discrimination used by carriers, often called
‘yield management’ in this context. This takes many dimensions, but adjust-
ing fares offered upwards as the date of a scheduled flight’s departure
approaches is an almost universal feature of liberalized airline markets.
Basically, when initially put on, the lowest fares are set at about marginal
cost, but then as departure approaches successive ‘buckets’ of the lowest-
fare seats are taken from the market. Thus, the temporal fares offered curve
turns up towards take-off time. The logic is simply that leisure travellers
book early because they have a known itinerary and are price sensitive, while
business travellers who often arrange travel at the last minute and require
flexibility are effectively a captive market that will support higher fares.
The evidence from a significant number of studies across a range of
airline markets is that when there are monopoly airlines serving a route this
pattern emerges very clearly, but when there are several competitive suppli-
ers the temporal fares offered curve remains pretty flat and at the marginal
cost level (Button and Vega, 2006; Button et al., 2007; Pitfield, 2005).

M4371-FINGER_9781786431851_t.indd 294 14/11/2017 11:14

STRATEGIES EMPLOYED BY AIRLINES TO REDUCE

MARKET INSTABILITY
Air carriers are not the only ones that have trouble recovering costs due
to what is often called excessive competition, nor are they the only case
where this leads to market instability. Ronald Coase (1946) argued that the
conditions for an empty core are the norm rather than the exception, and
that multipart pricing in various forms is the most practical way of cost
recovery, one that is widely adopted in industry. He goes on to say that
there is no neat solution to the problem, but what happens is that manage-
ment becomes inventive in finding ways to give their company an edge.
Many airline managers are certainly not unimaginative within this context;
indeed, in many ways they have been very creative in devising ways to try
to recover their full costs. However, the fact is that in gaining a temporary
edge they only add to the instability in the long term as others copy the
first mover.
Generally, industries suffering from an empty core problem often employ
a wide variety of techniques to recover their full costs, with the adoption
of a particular practice or measure depending upon a combination of the
technical characteristics of the industry and the institutional environment
in which it operates (Button, 2005a). These actions largely revolve around
efforts at creating a degree of monopoly power to allow full cost recovery.
Many airlines function for a considerable time without recovering their
full costs of capital. In some cases, this may be possible because capital
debts are written down through institutional means. The US Chapter 11
bankruptcy laws, for example, effectively allow an airline to restructure and
write down its capital without being broken up. Virtually all US carriers
have used this mechanism (some, such as US Airways, more than once).
From a wider perspective, competing firms often view this as unfair com-
petition, because the costs of their rival have been reduced through legal
means rather than through strict efficiency improvements. But Chapter 11
can be expensive not only in terms of legal fees, but also because while in
bankruptcy – and often for a period after – the costs of purchasing inputs
(fuel, aircraft leases, etc.) tend to be higher and can require immediate
payment.
Stability can be increased by de facto coordination down the air trans-
portation supply chain. As we have seen, airlines are at the end of a value
chain and, as such, other industries such as airports, air navigation service
providers, airframe and aero-engine manufacturers and global distribu-
tion systems rely on them to generate the revenues on which they in turn
rely. Because some of these industries are more successful – often because
they have monopoly power – in recovering their costs of capital, they may

M4371-FINGER_9781786431851_t.indd 295 14/11/2017 11:14

‘invest’ further down the value chain in airlines to ensure their solvency; it
is in their own interest to do so.
Airlines have also attempted to reduce competitive pressures by encour-
aging customer loyalty though frequent-flyer programmes that reward
regular passengers with free flights and other bonuses such as upgrades to
higher classes of service and access to airport lounges. The ‘miles’ earned
on carriers within alliances are normally interchangeable, albeit not per-
fectly, providing passengers with an extensive range of services for redemp-
tion. It is now also possible to collect miles through non-airline purchases
such as credit card use, car rentals and dining. The airlines effectively sell
their miles to other industries that then give them as rewards to their own
customers: in 2010 Delta Air Lines earned $1.6 billion from such sales.
Since liberalization, the greater flexibility afforded to airlines has led
to new carriers entering the market that provide lower fares and more
basic services. This has been the strategy of airlines like Southwest in the
US and Ryanair and easyJet in Europe, and many more. In some ways,
this business model represents the ‘blue ocean’ approach to market pen-
etration by seeking out new categories of customers as well as picking
up business from the edges of existing markets (Kim and Mauborgne,
2005). It generally entails considerable standardization in its operations
(e.g. the use of a common family of aircraft and a homogeneous, usually
short-haul, network of services), maximizing the use of its labour, serving
less-congested, smaller airports, providing a ‘no-frills’ service on the plane
and at the airport, limiting methods of booking, charging for non-core
services such as refreshments and checked bags and offering only one class
of service. Such measures can reduce costs by 30 per cent or so compared
to those of traditional airlines.
Liberalization has also allowed airlines flexibility in their price setting,
and the advent of computer booking systems has allowed carriers to fine-
tune this with prices being set according to customers’ willingness to pay.
This price discrimination allows an airline to offer seats at various prices,
and to continue to vary these offers as seats are purchased. The conditions
pertaining to a ‘seat’ can also differ; for example, the ticket may be refund-
able, it may be upgradeable to a higher level of service if combined with
frequent-flyer miles or it may be at a location on a plane (for instance, a
seat in an emergency exit row), and prices are adjusted according to these
quality factors. Yield management, being related to the willingness of cus-
tomers to pay, results in passengers who are less sensitive to price paying
more and contributing to the capital cost of the service, while those who
are less willing are charged lower fares that at least cover their marginal
costs. While it can be used to generate large profits – and this has been
done in many industries – its main purpose in air transport is to generate

M4371-FINGER_9781786431851_t.indd 296 14/11/2017 11:14

sufficient revenue to earn an acceptable return after all costs, including the
cost of capital, have been covered.
Instead of focusing on reducing costs, some airlines seek greater stabil-
ity by making revenue flows more certain. Charter airlines effectively do
this. They sell capacity well in advance to tour operators and therefore
know months ahead how much capacity needs to be delivered to custom-
ers, thus removing some risk from their operations. Scheduled airlines
cannot follow this pattern because they guarantee a service ahead of time,
and then effectively become common carriers of the traffic willing to pay
for flights. However, in some US cities, groups of businesspeople have
tried to ensure regular air services with guarantees of adequate patronage
for an initial period: ‘travel banks’. In Wichita, Kansas, for example, 400
businesses raised $7.2 million for this purpose, and similarly in Pensacola,
Florida, 319 businesses raised $2.1 million, while companies and indi-
viduals in Stockton, California bought $800 000 of prepaid tickets from
American West (Nolan et al., 2005).
The problem with all these measures is that while they can afford first-
mover advantages to those who initiate them and thus reduce the competi-
tive pressures confronting these carriers, the ability to rapidly emulate in
a liberalized market means that the short-term stability enjoyed by first
movers is soon evaporated. Just consider discriminatory pricing. This
afforded early adopters a degree of monopoly power because these carriers
sold many of their tickets through their own retail outlets, and then their
own computer reservation systems control the flow of fare information
to potential passengers. It was time-consuming for potential customers
to search for the cheapest ticket. The emergence of global distribution
systems has largely removed the asymmetric information advantage that
the airlines enjoyed: customers can easily get details of fares from a range
of airlines and the associated services and restrictions that go with them.
This may involve looking at a few pages of information, but sites such as
Orbitz and Kayak keep this to a minimum. It has become much harder for
any airline to differentiate between customers and to extract the highest
possible fares from them.
When it comes to low-cost carriers, while the underlying model clearly
was advantageous and remains so for some carriers today, the most suc-
cessful carriers adopting this approach have tended to be the first in the
market, allowing them to enjoy a ‘first-mover advantage’. The list of
subsequent failed low-cost airlines is long. One problem is that as they
have expanded, they have gradually moved into increasingly thin and less
suitable markets for their style of operation, and diminishing returns have
set in. Additionally, as more carriers have emerged, competition between
low-cost airlines has grown, hitting all their bottom lines (Button et al.,

M4371-FINGER_9781786431851_t.indd 297 14/11/2017 11:14

2007). The traditional airlines have also become leaner and more skilled
at resisting the challenges associated with low-cost carriers trying to enter
their routes (Morrison, 2001; Pitfield, 2008). Equally, loyalty programmes
have been reduced in their potency as any currency does when its supply
expands excessively, with ever-increasing numbers of miles being required
to buy flights and the number of flights for sale shrinking.
THE PUBLIC POLICY REACTION
Public policy has always played a key role in the air transportations sector,
and still does, even though the intensity of market intervention has signifi-
cantly declined in most countries since the 1970s and the instruments used
are somewhat less blunt than in the past. The concerns of policy-makers
has also changed somewhat, with less focus on anti-trust issues as their
understanding of the nature of competition in air transportation evolved.
Light-handed economic regulation, where regulation continues, is more
the norm. Similarly, state ownership has declined in its traditional forms,
although public–private partnerships have grown, often in the form of con-
cessionary structures. All these trends have removed some of the former
cushions against instability, but have resulted in greater efficiency in sup-
plying aviation services and in matching those services with the demands
of customers.
Subsidies have long been used to recover capital costs. The argument
often used is that once an investment has been made, it becomes economi-
cally efficient to maximize its use, subject to the willingness of users to pay
the incremental costs of their decisions. The original argument couched
150 years or so ago related to a bridge: once built, if the demand had been
overestimated it would be wasteful of the asset not to let people use it,
provided they paid the maintenance and repair costs associated with their
actions. The current trend to unbundle attributes of a service – such as
charging for food and second checked bags by some airlines – attempts to
isolate the activities in which the fixed costs are concentrated and to charge
explicitly for the incremental costs. The fixed costs in this sense can then be
isolated, and the other attributes – the food and bag service – sold in the
market at competitive prices. This concept of unbundling is quite common
in the service sector and is being increasingly used by airlines, where a basic
service can be augmented with add-ons.
Direct subsidies in this context are designed to cover the fixed costs
that cannot be recovered from customers. In the case of airlines, however,
where the fixed cost is that of a commitment to a schedule, it is difficult to
isolate the fixed cost in the traditional sense. Further, there is the generic

M4371-FINGER_9781786431851_t.indd 298 14/11/2017 11:14

problem that subsidies reduce the incentive toward efficient production. If

the recipient knows that losses are going to be covered by external sources,
there is less incentive to restrain costs. Moreover, there is less incentive to
provide the goods and products that customers seek. This has been long
recognized, and although there have been small levels of subsidy offered
through the Essential Air Services programme in the United States and
support has been given indirectly through the Civil Reserve Air Force
(CRAF) and Fly America programmes, direct subsidies to US airlines have
almost entirely been abandoned.
This does not, of course, mean that liberalization has been ubiquitous,
and there are some markets – particularly in Africa and Asia – where state
involvement in the sector remains considerable. Whether this reduces insta-
bility in the sector or simply passes its costs from the supplying industries
to the taxpayer has not been well researched. It is difficult to isolate volatil-
ity in these markets from the normal growth paths of rapidly developing
markets: Chinese aviation, which is largely state regulated and owned,
for example, has consistently experienced double-digit annual percentage
increases in air traffic over the past 15 years.
CONCLUSIONS
Air transportation experiences market instability, as does virtually every

other economic sector. Many of the factors causing it are generic in nature,
but regarding aviation and its infrastructure there are specific features
that can pose particular challenges to both suppliers and users of air
services. In particular – and something that has been little examined in
depth – scheduled and fairly homogeneous services of any kind provided
in a competitive market seem subject to instabilities. This is simply because
competition, by nature, forces producers to sell their product at marginal
costs, and often does not allow recovery of the full average costs associ-
ated with committing resources to be available at the time the service is
scheduled.
Coordination of suppliers (airlines, in this case) ensuring that competi-
tion is contained either by coordination, as with airline alliances, or by dif-
ferentiation in the nature of the service being provided, offers mechanisms
ensuring that a stable equilibrium emerges. And we have seen airlines in
markets across the globe seek to do these things as markets have been lib-
eralized. Their success has varied according to the nature of the markets
they function in and the response of regulatory bodies that often see efforts
at establishing a viable core as akin to seeking monopoly power for rent-
seeking reasons.

M4371-FINGER_9781786431851_t.indd 299 14/11/2017 11:14

Whether empty cores are economically a problem is unclear, and is

probably highly contextual. To simply stymie competition to obtain sta-
bility – very much as many governments sought to do in the immediate
Post-World War II period – demonstrably leads to high costs for consum-
ers and inertia in technology and management practices. It also assumes
that market intervention in air transportation offers a better solution to an
essentially technical managerial problem that management across a wide
range of other sectors has to handle on an everyday basis. Additionally, it
assumes that market intervention is neutral in its intent and practice and
aimed at serving the public interest, whereas there is evidence that this is
often not the case, especially after a regulatory regime has been in place for
some time; regulatory institutions and policy instruments can be captured
by parties seeking to pursue their own interests rather than those of the
wider public. Instability has its costs, but so do the policies that may be
introduced to contain it.
NOTES
1. We have adopted the standard definition of economic stability here as meaning a system
that is stable when it dissipates market imbalances that arise endogenously, as a result of
significant adverse and unforeseeable events, or as an endogenous consequence of the
intrinsic nature of the market. A stable system absorbs shocks largely via self-corrective
mechanisms, preventing adverse events from disrupting its internal economic efficiency,
or spreading over to other economic systems.
2. There are other arguments put forward as to why scheduled airlines cannot cover
their full economic costs. For example, Borenstein (2011) puts it down to demand factors
and the dichotomous market, with high-cost carriers competing with low-cost airlines.
3. One could argue that an airline could change the gauge of equipment used for a flight
as the levels of booking become apparent, but many low-cost carriers only use a single
broad type of plane (for example, Southwest and Ryanair both only use Boeing 737 air-
craft to reap economies of fleet).
REFERENCES
Antoniou, A. (1998), ‘The status of the core in the airline industry: the case of the
European market’, Managerial and Decision Economics, 19, 43–54.
Bailey, E.E. (2010), ‘Air-transportation deregulation’, in J.J. Siegfried (ed.), Better
Living Through Economics, Cambridge, MA: Harvard University Press.
Baumol, W.J. and R.D. Willig (1981), ‘Fixed costs, sunk costs, entry barriers, and
sustainability of monopoly’, Quarterly Journal of Economics, 96, 405–31.
Borenstein, S. (2011), ‘Why can’t US airlines make money?’, American Economic
Review, Papers and Proceedings, 101, 233–7.
Button, K.J. (1996), ‘Liberalising European aviation: is there an empty core
problem?’, Journal of Transport Economics and Policy, 30, 275–92.

M4371-FINGER_9781786431851_t.indd 300 14/11/2017 11:14

Button, K.J. (2003), ‘Does the theory of the “core” explain why airlines fail to
cover their long-run costs of capital?’, Journal of Air Transport Management, 9,
5–14.
Button, K.J. (2005a), ‘The economics of cost recovery in transport’, Journal of
Transport Economics and Policy, 39, 241–57.
Button, K.J. (2005b), ‘How stable are scheduled air transport markets?’, Research
in Transportation Economics, 13, 27–49.
Button, K.J. (2007), ‘The implications of the commercialization of air transport
infrastructure’, in D. Lee (ed.), The Economics of Airline Institutions, Operations
and Marketing 2, Oxford: Elsevier.
Button, K.J. (2010), ‘Air transportation services: both a traded commodity and
a transactions cost in international trade’, Journal of International Commerce,
Economics and Policy, 1, 105–20.
Button, K.J. and K. Johnson (1998), ‘Incremental versus trend-break change in
airline regulation’, Transportation Journal, 37, 25–34.
Button, K.J. and G. McDougall (2006), ‘Institutional and structural changes
in air navigation service-providing organizations’, Journal of Air Transport
Button, K.J. and H. Vega (2006), ‘Airlines competing with themselves: a note on
the temporal pattern of fare setting prior to departure’, International Journal of
Button, K.J., A. Costa and C. Cruz (2007), ‘Ability to recover full costs through
price discrimination in deregulated scheduled air transport markets’, Transport
Reviews, 27, 213–30.
Coase, R.H. (1946), ‘The marginal cost controversy’, Economica, 13, 169–82.
Edgeworth, F.Y. (1881), Mathematical Physics, London: Kegan Paul.
Kahn, A.E. (1988), ‘Surprises of airline deregulation’, American Economic Review,
Papers and Proceedings, 78, 316–22.
Kim, C.W. and R. Mauborgne (2005), Blue Ocean Strategy: How to Create
Uncontested Market Space and Make the Competition Irrelevant, Cambridge,
MA: Harvard Business Review.
Knight, F.H. (1921), Risk, Uncertainty and Profit, Boston, MA: Hart, Schaffner
and Marx.
Levine, M.E. (1987), ‘Airline competition in deregulated markets: theory, firm
strategy, and public policy’, Yale Journal on Regulation, 4, 393–494.
Meyer, J.R. and W.B. Tye (1981), ‘The regulatory transition’, American Economic
Review, Papers and Proceedings, 71, 268–72.
Morrison, S.A. (2001), ‘Actual, adjacent and potential competition: estimating the
full effect of Southwest Airlines’, Journal of Transport Economics and Policy, 35,
239–56.
Morrison, S.A. and C. Winston (1995), The Evolution of the Airline Industry,
Washington, DC: Brookings Institution.
Nolan, J., P. Ritchie and J. Rowcroft (2005), ‘Small market air service and regional
policy’, Journal of Transport Economics and Policy, 39, 363–78.
Pitfield, D.E. (2005), ‘A time series analysis of the pricing behaviour of directly
competitive “low-cost” airlines’, International Journal of Transport Economics,
32, 15–38.
Pitfield, D.E. (2008), ‘The Southwest effect: a time-series analysis on passengers
carried by selected routes and a market share comparison’, Journal of Air

M4371-FINGER_9781786431851_t.indd 301 14/11/2017 11:14

Sjöstrom, W. (1989), ‘Collusion in ocean shipping: a test of monopoly and empty

core models,’ Journal of Political Economy, 97, 1160–79.
Telser, L.G. (1978), Economic Theory and the Core, Chicago, IL: University of
Chicago Press.
Telser, L.G. (1987), ‘The usefulness of core theory in economics’, Journal of
Economic Perspectives, 8, 151–64.

M4371-FINGER_9781786431851_t.indd 302 14/11/2017 11:14

PART III
Future challenges

M4371-FINGER_9781786431851_t.indd 303 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 304 14/11/2017 11:14

16. Economic perspectives on aviation
security
INTRODUCTION
The aviation security problem is essentially an economic one: how best to

allocate scarce resources to reduce the probability of a successful attack
against civil aviation to an acceptable level. The economic concept of scar-
city has two important meanings here. First, resources devoted to defence
activities of any kind (including aviation security) do not directly increase
economic welfare (rather, such activities serve to prevent potential reduc-
tions in welfare). When people are forced to expend resources to protect
themselves, they are reducing the resources available for direct investment
in capital goods and technology and for production and consumption of
goods and services. Secondly, given a finite budget allocated to the general
activity of national defence, the resources devoted to aviation security
represent a reduction in resources available to protect non-aviation targets.
Moreover, the resource allocation problem is complicated by the fact that
allocation decisions are strategic; aviation security risks are not the same as
natural disaster risks. For example, if we decide to allocate more resources
to ensure buildings are earthquake-proof, this does not change the prob-
ability of an earthquake occurring. However, if we allocate relatively more
resources to one aviation security measure, and relatively fewer to another,
we change the expected pay-offs to terrorists and thus potentially change
the probabilities and modes of attack.
BACKGROUND
In the last 45 years trade, technology and economic growth have created
an age of globalization in which the welfare of people, firms and nations
has become ever more interconnected. During this period, civil avia-
tion has evolved from a heavily regulated system of national airlines and
government-operated airports to a much larger and more competitive
305

M4371-FINGER_9781786431851_t.indd 305 14/11/2017 11:14

100
1970s
90 1980s
73%
80 1990s
Total number of attacks
70 2000s
67%
60
50
40 84%
30
19%
20 17% 14%
43%
10 5% 11% 24% 24% 3%
2% 5% 0% 1% 2% 1% 0% 5%
0
Hijacks Bomb Cargo Bomb Hijack-Crash Other
Source: Rand Database of Worldwide Terror Incidents.
Figure 16.1 Global attacks inside aircraft by type of attack, 1970–2009
global industry in which private airlines and airports compete along with
publicly-owned counterparts and hybrid organizations under diverse regu-
latory regimes. There has been remarkable growth in the number of global
air passengers over the last 55 years, a long-term trend which has been
largely impervious to the negative shocks of macroeconomic recessions,
health crisis, military conflict and acts of terror. In 2016, IATA World
Air Transport Statistics show there were approximately 3.7 billion air
passengers.
However, civil aviation has been a visible target for acts of violence
and terrorism throughout this period. Figure 16.1 shows the number
of attacks inside planes worldwide by attack type over the four decades
between 1970 and 2009, and indicates the percentage of attacks in each
decade accounted for by each attack type. In the 1970s, attacks on aircraft
were heavily skewed towards hijackings. However, this mode of attack has
declined both in number and in relative importance over time. The figure
also shows a sharp rise in the number of bomb attacks that occurred in the
1980s, falling again in the subsequent decades. Thus, the data captures the
evolutionary nature of aviation security. As authorities implement security
measures to nullify a given mode of attack, terrorists adapt their strategies
and the preferred mode of attack evolves.
Overall, the number of attacks has declined significantly over time, with
111 attacks in the 1970s but just 21 attacks between 2000 and 2009. In

M4371-FINGER_9781786431851_t.indd 306 14/11/2017 11:14

Economic perspectives on aviation security 307
terms of fatalities, there were 557 deaths because of attacks inside planes
in the 1970s. This number rose to 1115 in the 1980s, mainly as a result
of a small number of attacks that inflicted a large number of casualties,
including Air India flight 182 in 1985 (329 fatalities) and Pan Am flight
103 in 1988 (270 fatalities). In the 1980s the objectives for attacking air-
craft had evolved from attention-seeking through prolonged live media
coverage of a hijacking to the shock and terror generated by the sudden
and unexpected mass killing of innocent civilians. By the 1990s, aviation
security had responded and was evolving into a complex (and expensive)
system combining intelligence agencies, security personnel at airports and
investments in scanning equipment to detect bombs, weapons and other
prohibited items. In the 1990s the number of fatalities from all terror
attacks inside planes declined to 160, but the following decade will forever
be defined by the 2938 deaths resulting from the attacks in New York and
Washington on September 11th, 2001. Excluding the 9/11 attacks, globally
there were just 94 fatalities as a result of terrorist attacks inside aircraft
from 2000 to 2009.
The events of September 11th, 2001 represent by far the biggest and
most shocking realization to date of the ever-evolving threat of terror
attacks against aviation. The attacks demonstrated how civilian aircraft
could be used as weapons to kill large numbers of civilians and destroy
assets on the ground. The attack created mass panic over the vulnerabili-
ties of the civil aviation system and led to sweeping and significant changes
in the design, provision and financing of aviation security throughout the
world. Since 2001, governments have created new organizations to imple-
ment airport security systems and there have been massive investments
in both technology and the hiring and training of security personnel.
Through these changes and increases in security costs, airports and airlines
have faced new challenges in managing passenger throughput, minimizing
delays and negative passenger experiences resulting from elevated levels
of security effort. The general public and the travelling public have borne
both the direct and indirect economic costs of these investments.
Despite the elevated efforts in the first decade of the 21st century, avia-
tion has remained a target for terror attacks. Most recently, in 2016, attacks
in Brussels and Istanbul involving suicide bombers and shootings resulted
in 77 fatalities. In Brussels suicide bombers detonated explosives in a
departure hall, while in Istanbul attackers opened fire at the international
terminal of Atatürk airport, detonating suicide bombs in the car park and
in the departure and arrivals halls. These attacks demonstrated that a rela-
tive focus in preventing potential attackers or weapons from entering the
airside of an airport left vulnerabilities on the landside (prior to screening
checkpoints) both inside and outside airport terminal buildings.

M4371-FINGER_9781786431851_t.indd 307 14/11/2017 11:14

AVIATION SECURITY IN CONTEXT
Aviation security operates within the general context of national security.

The objectives of both are to protect the citizens and assets of a jurisdic-
tion from attack. Intelligence gathering is a general input into national
security with the objective of assessing threats and foreclosing the oppor-
tunity for any threats to materialize in the form of an attack. Within the
national security system, aviation security can be defined as consisting of
two activities:
1. Pre-emption through the screening and monitoring of people and

cargo at – and at all points between – an origin and destination airport.
2. Defence against attacks on people and assets that materialize at – and
at all points between – an origin and destination airport.
This chapter is largely focused on the first of these activities, which

includes preparedness through investments in capital equipment, screening
technology, personnel training and in overall system design. The overall
security system is a set of institutions, the design of which requires govern-
ment policy, legislation, regulations and standards, oversight and monitor-
ing and information feedback mechanisms to enable ongoing evaluation
and evolution of the system.
Government policy maps a set of stated objectives onto a set of legisla-
tive rules, employing a set of institutions to ensure implementation and
enforcement. The stated objectives of any national aviation security policy
must lead to a level of security that is consistent with or exceeding interna-
tional standards (as defined by ICAO’s Annex 17). Yet despite the attempts
to implement harmonized global standards such as Annex 17, aviation
security policies can be and are interpreted and implemented in different
ways around the globe.
AVIATION SECURITY AS A PUBLIC FRANCHISE
Following Gillen and Morrison (2005) one can view a civil aviation
network as a ‘public franchise’. Similar to private franchise organizations,
the overall value created by a public franchise is the result of the com-
bined efforts of individual organizations in the franchise. The glue that
binds public franchises together is the vertical rules and regulations that
define the relationship between each organization and the government. In
this regard, the government (as franchisor) attempts to create incentive-
compatible rules that guide each member organization to voluntarily

M4371-FINGER_9781786431851_t.indd 308 14/11/2017 11:14

provide effort levels (in whatever dimensions are applicable) that will maxi-
mize the value of the public franchise. However, each organization in the
network also has a horizontal relationship with other member organiza-
tions and mal-incentives can exist because each organization draws value
from the overall success and value created by the public franchise, which
is due to the combined efforts of all the members. The benefit of a public
franchise network is that if implemented correctly, the overall objectives
of the franchise can be met while taking account of differences in the
circumstances of individual members. In other words, given heterogeneity
in member characteristics and circumstances, a public franchise approach
potentially avoids the problems of one-size-fits-all policy rules. However,
asymmetric information, measurement problems and monitoring costs
make the design of incentive-compatible franchise rules difficult and
potentially complex. The analytics of the franchising problem in a public
network setting (such as aviation) focuses our attention on the extent to
which the government as ‘franchisor’ can design a governance structure
that mitigates the horizontal externality problems and maximizes the
public welfare. Thus, a network of independently operated airports within
a nation or jurisdiction can be characterized as a public franchise network
in which the government – as franchisor – sets rules, regulations and a gov-
ernance structure to maximize the public value of the airport system. More
specifically, aviation security can be viewed in the same way, whereby the
government seeks to create an institutional structure that recognizes differ-
ences in the circumstances of individual airports in the system, while at the
same time ensuring that each airport contributes a level of effort in the pro-
vision of aviation security that achieves national (and international) objec-
tives. Canada’s National Civil Aviation Security Program seems to embody
this approach by stating a preference for ‘outcome-based’ approaches to
security and less reliance on prescriptive regulatory rules: ‘Prescriptive
regulations and “one-size fits all” solutions may not always account for the
wide variability in industry operations and skills, and can place unneces-
sary burdens on stakeholders resulting in ineffective security outcomes.
Therefore, the most effective way to address security gaps is often through
outcome-based rules.’ (Government of Canada, 2013).
Viewed in this way, global aviation security becomes a network of nation-
ally controlled public franchise networks. To visualize this, consider an
international environment with just two countries (A and B). International
laws and agreements create a common international standard for the
provision of minimum security levels at all airports. Each nation is then
responsible for the implementation of these standards. If it is assumed that
each country has two airports: (a1, a2) and (b1, b2), that there are domestic
flights between a1 and a2 and between b1 and b2, and international flights

M4371-FINGER_9781786431851_t.indd 309 14/11/2017 11:14

a1
V H
A a2
V
International
Standards H H
b1
V H
B
V b2
Figure 16.2 Vertical and horizontal relationships domestically and

internationally
between a1 and b1 and between a2 and b2. Even with this simple model,
there is a complex three-tier set of vertical and horizontal relationships, as
illustrated in Figure 16.2. At the top tier there is a horizontal relationship
between countries A and B. Formally, the governments of these countries
have agreed to the same international standards; however, in their vertical
relationships with their own airports, A and B can vary the level of enforce-
ment which does not guarantee that international standards are met, since
lower enforcement levels are less costly. These choices form a strategic
interaction between A and B that can be summarized as the first stage of
a two-stage game, as illustrated in Figures 16.3a and 16.3b. In Stage 1 of
the game governments decide on their level of enforcement, then in Stage
2 each airport makes decisions concerning its level of security effort. The
strategic interaction between airports occurs domestically between a1 and
a2 and between b1 and b2, but there is also strategic interaction internation-
ally between a1 and b1 and between a2 and b2.
In these strategic encounters, it is assumed each airport knows the level
of enforcement implemented by its domestic government. Figure 16.3b
shows the strategic interaction between international airports ai and bi,
and here there is a potential for a prisoner’s dilemma pay-off structure
(Coughlin et al., 2002). If it is assumed that citizens’ willingness to fly is a
function of the average perceived security at airports, then if one airport

M4371-FINGER_9781786431851_t.indd 310 14/11/2017 11:14

Strong B Weak
enforcement enforcement
Strong Weak
Strong enforcement enforcement
1 , 1 2, 2
enforcement
A Strong
1 , 1 2, 2
enforcement
Weak
3, 3 4, 4
A enforcement
a Weak
3, 3 4, 4
enforcement
Figurea16.3a Stage 1 bi-matrix game between jurisdictions
bi
High Security bi Low Security

Level Level
High Security Low Security
Level Level
High Security
ai1, bi1 ai2, bi2
Level
High Security
ai ai1, bi1 ai2, bi2
Level
Low Security
ai ai3, bi3 ai4, bi4
Level
Low Security
b ai3, bi3 ai4, bi4
Level
Figure 16.3b Stage 2 bi-matrix game between international airports
implements a higher level of security and bears the full costs of that action,
other airports will benefit from an increase in the average level of security
in the system.
As the number of airports in the system gets large, the average increase
in overall security resulting from the actions of one airport becomes small
relative to the cost it bears to implement higher security standards. In this
way, the implementation of higher security is discouraged. Theoretically,
this incentive problem can be avoided if there is strong enforcement of
international standards in the vertical relationship between a nation’s
government and its airports. Thus, a nation that strongly enforces security
standards changes the pay-offs in the Stage 2 games such that choosing
high security becomes a dominant strategy for its airports. However, the
prisoner’s dilemma can still prevail if the strategic interaction between

M4371-FINGER_9781786431851_t.indd 311 14/11/2017 11:14

nations exhibits a similar horizontal incentive problem. A single nation

may not gain much relative to the costs of implementing higher security
standards, given the average level of security provided by the combined
efforts of other nations. Note that a national decision for weak enforce-
ment need not be the result of free-riding born out of narrow self-interest.
Weak enforcement can arise from a choice of governance structures and
regulatory requirements (made with the best of intentions) that are rela-
tively ineffective in the vertical relationship between governments and air-
ports in the national public franchise.
To summarize, a nation’s aviation system can be represented as a public
franchise network with vertical interactions between individual airports
and the government and horizontal links between airports. Finding a
system of governance that balances the unique characteristics of each
airport with the objectives and value generated by the whole system
requires attention to be paid to the external impacts and incentives created
by the actions of individual airports and governments. The creation of
international standards for aviation security may be a necessary condition
for a net increase in social welfare, but may not be sufficient to guarantee
improvements if mal-incentives give rise to a lack of enforcement at the
national level.
THE PRODUCTION OF AVIATION SECURITY:

OUTPUT AND COSTS
The economic theory of production can be utilized to position thinking

about trade-offs between inputs in the production of aviation security
services (Coughlin et al., 2002). A production function describes the
technology of production by relating how various inputs are combined to
produce a definable output. This approach highlights the fact that technol-
ogy often enables the same level of output to be produced using different
combinations of inputs. For example, a given level of aviation security
could be produced in either a labour-intensive or a capital-intensive way
or via a more balanced combination of labour and capital. The optimal
means of production will depend upon the productivity and cost of each
input. Furthermore, changes in technology can alter not only the absolute
productivity of inputs but also relative productivities, thereby changing the
optimal mix of inputs. To take advantage of such changes in technology,
the institutions governing the provision of aviation security need to be flex-
ible enough not only to adopt new technologies, but also to make necessary
changes in the mix of inputs. Presently, many new technologies (biom-
etrics, for example) offer the potential to change the way that aviation

M4371-FINGER_9781786431851_t.indd 312 14/11/2017 11:14

security is provided; however, the adoption of new technologies is also

problematic because of the limitations in our ability to measure output. To
the extent that a change in technology implies a different combination or
level of inputs than was used in the past, policy-makers and those oversee-
ing the security system face a difficult task in assessing its impact. The lack
of counterfactual evidence means that a new technology could be blamed
for creating vulnerabilities (if a successful attack occurs) after its imple-
mentation, if the new technology replaces elements of the old system. Such
measurement problems create an incentive to make changes to the system
only by increasing resources rather than reallocating resources across the
system. The result is then a system that evolves as a series of layers that are
added onto the original system, which will not only cause ever-increasing
costs but may also create inefficiencies and suppress the intended benefits
of the new technologies.
The standard economic approach to such questions is benefit–cost
analysis, which – given measurement problems in aviation security – is
challenging but not impossible. Applying benefit–cost analysis to aviation
security, the benefits of a given security measure in year t can be defined
as:
Bt5(pt,−s − pt,s)kt (16.1)
where pt,−s represents the probability of attacks in year t without the imple-
mentation of the security measure; pt,s represents the (reduced) probability
of attacks in year t after the implementation of the security measure and
k represents the value of lives and assets expected to be destroyed in suc-
cessful attacks annually. These probabilities must be conjectured from
historical evidence and intelligence. Nevertheless, despite these difficulties,
it is possible for benefit–cost analysis to provide insights in support of
improved resource allocation decisions, providing that informed assump-
tions can be made about the risk reductions associated with any given
security measure. Stewart and Mueller (2008) used such an approach to
assess specific aviation security measures in the US in the aftermath of the
September 11th attacks. A key element is the use of sensitivity analysis to
provide the ranges of probabilities required for a given security measure
to generate net benefits. This step is essential as a robustness check on
the reliability of the estimates. In evaluating the PreCheck (trusted trav-
eller) programme in the US, Stewart and Mueller calculate the benefits
of the current aviation security system with and without the PreCheck
programme. Under conservative assumptions concerning the ability of a
trusted traveller system to detect terrorists and the overall probability that
any citizen is a terrorist, they showed that the net effect of the PreCheck

M4371-FINGER_9781786431851_t.indd 313 14/11/2017 11:14

programme is to generate benefits and efficiencies while probably improv-

ing the overall level of security.
MEASURING OUTPUT IN AVIATION SECURITY
What exactly has been produced from all the efforts to provide avia-
tion security? One way of answering is to ask ‘what is the probability
of someone being killed as a result of an aviation security incident?’
Measuring this requires a numerator, which is the fraction of people killed
in incidents, and the denominator, which measures the ‘opportunity’ for
such incidents – the number of flights in a given period in a given juris-
diction. In the case of aviation safety, a reasonable measure would be the
proportion of people who do not survive a flight in comparison to the
number of flights.1 The ‘reasonableness’ of this measure can be improved
by considering differences in the severity of an incident: a large proportion
of people dying from an incident is different from when a small proportion
of people die. In addition, we wish to account for the majority of aviation
incidents that take place during landing or take-off.
One metric of aviation security associated with passenger security ser-
vices at an airport could be ‘1 minus the ratio of the proportion of fatali-
ties (or injuries) over the number of domestic plus international scheduled
commercial flights in the jurisdiction being considered’. This metric could
be calculated for a region or country.2 The metric could include a broader
number to measure the impact in the numerator; for example the number
of people killed or the number of people killed or injured, or the number
m 5 acalculations
of non-airline passengers killed mor the number of air and non-air pas-
CRThese
sengers killed (or injured). si could take account of the
different potential for a threat byi51considering, for example, where flights
originate. Are they from high-risk jurisdictions? Would there be greater
risk with a flight of a large aircraft than a small aircraft? For example, are
than5ona
terrorists more likely to exert greater
N effort to inflict damage on a long-
haul, wide-body aircraft HH s2i 3 100
a shorter-haul, narrow-body aircraft?
To formalize an output measure i51
of aviation security, consider the
following:
a i51 f (i) t,k

n
j
SL j
t,k 512 (16.2)
nt,k
On the left-hand side of Equation 16.2, SL is the ex post measured level

T^ t, in jurisdiction k for injury/fatality of type j.
of aviation security, in time
T^
0C/0T^
M4371-FINGER_9781786431851_t.indd 314 14/11/2017 11:14
On the right-hand side of the equation f(i) is the proportion of people on

flight i who suffer a type j consequence (who are killed or injured) in period
t, in jurisdiction k; and n is the number of flights in period t in jurisdic-
tion k. As the right-hand side of the equation goes up, the measured level
of security as a result of a security incident goes down. As an example,
suppose one wanted to measure the level of aviation security in Canada
from 2000 to 2010, where my interest is in the number of people killed as a
result of air crashes that resulted from action taken by terrorists who have
breached security at an airport.3 Therefore, type j would mean ‘people who
died’, period t would be 2000–10, jurisdiction k is Canada, n would be the
number of flights between 2000 and 2010, and f(i) would be the proportion
of people killed on a flight due to a security incident and the summation
(S) means we are adding all f(i)’s up over all flights.
What are the weaknesses of this metric? First, the metric is based on
historical data, and such data does not reflect terrorists adapting their
behaviour over time with each incident. Second, the metric does not link
the security incident to the damages (lives lost). The actual sequence of
events is as follows. First there is a security event (E) which will have a
given probability in affecting the number of lives lost. For example, an
event in which a knife is smuggled onto an aircraft is quite different from
an event where a bomb is smuggled on board. The point is that as speci-
fied, the security metric is an outcome-based metric and does not inform
security personnel on how to improve security. However, creating such a
metric is certainly possible. Following Jackson et al. (2012), it is possible
to alter the numerator of Equation 16.2 by disaggregating f(i). Recall that
f(i) is the proportion of people who die from an event that resulted from
a security incident. The variable f(i) can be unbundled by taking account
of the probability that security incident M will be associated with airport
security screening passengers, the probability of a security incident of type
M will occur and the impact a security incident of type M has on bringing
about a crash. Therefore, f(i) can be written as:
f (i) 5 S (a)· a (M)· f(i (M)) (16.3)
where S(a) is the likelihood that the security incident will be at an airport
involving passenger screening, a(M) is the likelihood the incident will be of
type M and f(i(M)) is the consequence an incident of type M has on the
likelihood a plane will crash. For example, Jackson et al. (2012) reported
the security incident locations and the type of attack on the airplane. As
Figure 16.1 illustrated, in absolute terms hijackings have dominated air-
craft terrorism attacks, but the pattern of attacks has changed significantly
over time. From 1970 through to 2008, hijackings went from over 75 per

M4371-FINGER_9781786431851_t.indd 315 14/11/2017 11:14

cent of attacks to approximately 40 per cent, while bombings went from

less than 7 per cent to more than 25 per cent. However, the number of
attacks also changed: between 1970 and 1979 there were 111 attacks, while
from 2000 to 2008 there were 19 attacks. These numbers help to inform
about S (a) and a (M). There is also data to inform f(i (M)); for example,
over 65 per cent of the attacks on aircraft had no injury or fatality associ-
ated with them, and 75 per cent had no fatalities.
MEASURING AND COMPARING THE COSTS OF

AVIATION SECURITY ACROSS JURISDICTIONS
Current understanding of the cost function for aviation security is

extremely limited. For example, we do not know the degree to which there
are economies of scale in the provision of security services. However, it is
difficult to obtain data on the costs of providing aviation security services,
partly because different levels of government fund and recover costs of
aviation security in different ways and collect, maintain or publish data in
different ways.
Figures 16.4 and 16.5 show comparisons of the costs of providing avia-
tion security. The data was collected at the country level to be consistent
over time and comparable, to some degree, across jurisdictions. Even these
numbers, although somewhat comparable, represent different activities and
investments, and thus must be viewed with a degree of caution. Figure 16.4
$12.00
Canada
U.S.
$10.00
$8.00
$6.00
$4.00
$2.00
$0.00
2003 2004 2005 2006 2007 2008 2009 2010 2011
Figure 16.4 Cost per passenger; operating 1 capital ($CAD)

M4371-FINGER_9781786431851_t.indd 316 14/11/2017 11:14

Baggage and Screening Costs/Pax-CAN$

$10.00
$9.00
$8.00
$7.00
$6.00
$5.00
$4.00
$3.00
Canada
$2.00 U.S.
$1.00 New Zealand
$0.00
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Figure 16.5 Baggage and screening costs per passenger ($CAN)
provides a comparison of cost per passenger where costs are operating and
capital costs. From this data, costs in the US exceed those in Canada in
every year, and the spread is widening. These cost differences can reflect
several factors, such as the larger US market (number of airports) and
differences in governance and public finance between the two countries.4
Figure 16.5 displays the cost per passenger for checking baggage and
personal screening. Again, data limitations only allow for a comparison of
three countries: the US, Canada and New Zealand. Canada has the lowest
costs, which are approaching those of the US in 2010. New Zealand’s costs
are considerably higher than the other two countries. All three countries’
costs level off in 2010 and have tended to have the same spread.
To investigate further, two types of costs are examined: operating plus
capital, and passenger screening and bag check costs (expressed in real
terms). In the former, data from Canada, US and Australia is employed
and in the latter data from Canada, US and New Zealand expressed in real
Canadian dollars. In terms of total cost, no significant difference in costs
emerge between Canada, US and Australia that would be explained by
some other variables or that are inherent to those respective countries, and
the incremental cost of providing security services to a passenger is found
to be $11.79 CAD.5
The role of scale in the production of security could be an important
factor in determining the costs of providing aviation security. There
are significant variations in the size of airports both within and across
national borders, and it is possible that larger airports (with higher annual

M4371-FINGER_9781786431851_t.indd 317 14/11/2017 11:14

passenger volumes) will experience lower average costs of aviation security

compared to smaller airports. A large airport may be better able to accom-
modate large security capital equipment within its physical structure, and
such equipment may enjoy higher utilization rates at larger airports. The
existence of economies of scale would mean that a single security standard
enforced at all airports would have uneven cost impacts, with dispropor-
tionately larger costs in smaller airports or smaller jurisdictions. Our data
from Australia, Canada and the US yields an estimate of the elasticity of
total cost with respect to changes in the number of screened passengers of
0.96, which implies only slight cost economies.
In the case of passenger boarding and screening costs, once again no sig-
nificant difference is found between the countries, with passenger screen-
ing costs increasing in real terms over time. The incremental cost of serving
a passenger is estimated to be $9.56 CAD, and the elasticity of passenger
and screening costs is estimated to be 0.536. The latter indicates significant
cost economies, pointing to the ability of a screening team with scanning
equipment in a single lane to process additional passengers prior to operat-
ing at maximum capacity.
SECURITY COSTS AT EUROPEAN AIRPORTS
Europe has a different model to fund and provide security wherein the
national government sets security standards and each airport in a country
(or member state) provides the security services (either through producing
them themselves or contracting out) levying a charge on airlines and/or
passengers.6 Utilizing data from 2011 we were able to construct measures
of security costs for a sample of airports, as shown in Figure 16.6. Average
security cost per passenger varies from $6.28 at Salzburg Airport, Austria,
to $0.73 at Sabiha Gokcen Airport in Istanbul, with an average over all
airports of $2.88 per passenger. These numbers could be somewhat dis-
torted if transfer passengers are not screened when they connect and to
the extent that an airport handles a significant amount of cargo; this will
also upward bias the averages. To correct for these potential biases, a cost
model is estimated that includes variables for cargo volumes, the percent-
age of international passengers, whether the airport was a large hub and/
or gateway and dummy variables to examine differences across countries.
The results indicate that marginal cost per passenger is $2.59, which is
substantially less than our estimates using data for the US, Canada and
Australia. However, the elasticity of cost with respect to passengers is
0.94, which is like the US/Canada/Australia estimate. Security costs are
higher for large hubs, while cargo adds $36.33 per tonne to security costs

M4371-FINGER_9781786431851_t.indd 318 14/11/2017 11:14

14/11/2017 11:14 M4371-FINGER_9781786431851_t.indd 319
319
$0
$1.00
$2.00
$3.00
$4.00
$5.00
$6.00
$7.00
SZG
CGN
FRA
BUD
MUC
VIE
ADP
ORY
CDG
TLV
DAA
DUB
STR
HAJ
LHR
ZRH
LUX
AMS
DUS
GVA
Finavia
HEL
LIU
OSL
Avinor
HAM
TAV
MAG
WAW
MXP
SEA Milano
LIN
BTS
NCE
Source: Based on RDC Aviation; ATRS airport benchmarking data (2011).

Swedavia
ARN
Figure 16.6 Security cost per passenger: European airports, 2011

BSL
BLQ
LGW
TRN
AENA
SOF
STN
CPH
LTN
Security Cost per Passenger-2011 $US
MAN
ATH
Berlin
TXL
GLA
ADR
CIA
FCO
BGY
MAD
BHX
MAP
AGP
EDI
BCN
VCE
ANA
LIS
ALC
IST
LPA
BRS
PMI
SAW
Average
with elasticity of 0.09. The proportion of international passengers does

not appear to affect overall security costs in the EU, which is surpris-
ing given the large connecting hubs of Frankfurt, Munich, Charles de
Gaulle, Amsterdam Schiphol and London Heathrow. However, all traffic
between Schengen member states is considered domestic traffic. Spain,
Portugal and Turkey have lower security costs on average and Austria has
higher; otherwise, there are no significant differences across EU member
states.
SECURITY EFFICIENCY: ASSESSING TIME COSTS

OF PASSENGER SCREENING
Productive efficiency is an important aspect of costs and output for any

organization, and aviation security is no different in this regard; time and
disutility costs of passenger screening are an important element in the
assessment of indirect costs of aviation security. Economically efficient
service times, in principle, are determined when the incremental costs are
just balanced by the incremental benefits of the reduced service time.
This marginal balance is for a given level of security as measured by the
probability of being harmed on a flight or in the secure area of an airport.
The difficulty is that the level of service can be changed if a change in
threat level occurs. That is, if a potential danger is identified, security
screening levels can be increased, which increases the screening time per
passenger.
An ‘efficient service level’ can also be thought of as the economically
efficient time spent in the process, from arriving at a screening queue to
the time the screening is complete. There are two ways to reduce screening
time. Additional resources (that is, screening machines and/or screening
agents) can be added to the screening process or screening requirements
can be reduced (for example, if shoes, belts and other items need not be
removed).
Queues consume considerable resources, but queues that experience
delays and variation in service rates can be costly. Direct costs of more time
spent are incurred by passengers or baggage items in the queue; but delays
are variable, so uncertainty exists regarding the expected duration of a
delay. If delays are lengthy, passengers in the queue may miss an activity
(a flight) and/or the activity may be affected (delayed) so more resources
are used up. Looked upon as an inventory problem (the number of people
being held in a queue), the issue is how big the inventory of passengers
should be in the queue. Alternatively, the question could be asked: ‘If the
number of agents in the queue is X, what is the economic value of reduc-

M4371-FINGER_9781786431851_t.indd 320 14/11/2017 11:14

ing this number to X − t?’ Several forces will influence this value. First, the
(average) flow rate and the (average) flow time will influence the inventory.
The flow rate is the average number of passengers screened per unit time,
and this is governed by the amount of resources used to produce screening
services or the capacity of the system. An additional factor is the arrival
rate of the passengers: sometimes it may be high and at other times it may
be low. The flow time will be determined by the level of screening, and
how long it takes to complete a screening process. Rigorous screening will
increase flow time, while less rigorous screening reduces it. This time will
include waiting time in the queue. Thus, flow time is measured from the
time the passenger joins the queue to the time the passenger completes the
screening process and exits to the terminal. However, the flow time will also
depend on the ‘nature’ of the passenger: people who travel less often, the
elderly, foreign individuals, families and people with carry-on luggage all
take more time for a given level of screening.
Another factor is the size of the list of prohibited items. As the list grows
longer there is more inspection time as well as increased demands on pas-
sengers (shoes off, belts off, computers out of cases, etc.). The inventory
will also be affected by the processing capacity of the system. Combining
machines and humans produces security screening: the machines can
operate at a given processing speed, but the human agents create variability
in the time for processing. Different agents may have different experiences,
differing levels of risk tolerance or different training. The combination of
agents may affect processing speed.7
Developing an efficient screening service time is a matter of identifying
cost and benefit components and linking changes in those components to
changes in service times. The measure of costs would be the resource cost
in total costs to the airport, or to the security agency in charge. The same
is true on the benefit side; it is the full economic benefit to all parties that
needs to be measured. The benefit measure raises a question: who are the
beneficiaries of security screening? Some jurisdictions take the position
that it is air passengers, while for others it is air passengers and the general
public.8 This could raise a problem of how to measure benefits to society
from changes to security screening times, although this question is more to
do with screening levels than with screening times.
The economic cost of providing security screening can be represented
as:
C(K, L,r,T )(16.4)
where K is number of screening machines (servers); L is the amount of

labour (screening personnel); r is other resources such as floor space and

M4371-FINGER_9781786431851_t.indd 321 14/11/2017 11:14

CR m 5 a si
m
CR m 5 a si
i51
m
CR m 5 a si
i51
m
HH 5 a s2i 3 100
N
i51
of a
i51
N
tables in the screening area; and T is amount HH 5 times 3
2
itaken
100for the screen-
HH 5 a s2i 3 100 a i51 f (i) t,k

ing. Note T is provided by passengers.
N
i51
n
j
a i51 f (i) t,k

T 5 ϕ(ts, ts)(16.5) i51 s SL t,k 5 1 2
j
n
j nt,k
a i51 f (rigour
where tS is the amount of time required toSL n complete
t,k 5 1 2
j
the screening and tS 5
i) jt,k (shoes nt,k
f(S′) where S′ is the defined level of screening on or off, belt
t,k 5 1
SLetc.).
on or off, laptops in or out, j s is the variability of time and
ts 2 T^ is a func-
nt,k
tion of N/K, which is the ratio of passengers (N) to machines (K) (a volume
capacity ratio). T^
Equation 16.4 can be rewritten to represent private costs for a defined T
which is the service time T^ : T^
C(K, L,r)T^ (16.6)
where C9 can be defined as T^ the marginal cost of service time, 0C/0T^ .

The benefit of changes in T stems from reductions in tS and reductions
in tss , and from reductions in costs imposed 0C/0T ^ other agents in the supply
on
chain such as airlines who face real costs from delay or passengers 0B missing
0B 0B
0C/0T^ at security screening. Thus, the ,
flights due to excessive queuing 0, , 0,and s , 0
marginal
0T̂ 0ts 0t0
benefit is:
0B 0B 0B
, 0, , 0,and s , 0
0T̂ 0ts 0t0
0B 0B 0B0B 0B0B
, 0, , ,
, 0, 0, 0,and
and s s,,0 0(16.7)
0T̂ 0T̂ 0ts0ts 0ts0t0
which states that marginal benefits are decreasing with increases in service
time, queuing time and variance in queuing time.
To put the problem in context, consider a passenger taking a flight
from point A to point B who must complete several transactions: getting
from home/work to the airport, checking in and obtaining a board-
ing pass and checking any luggage, passing through security screening,
queuing at the boarding gate and finally boarding the aircraft. There
can be delays in any of these processes. It is assumed all other processes
do not change if the screening service time changes, meaning if security
invests resources to reduce waiting times in screening queues, airlines will
not reduce check-in resources to capture some of the added time now
available. It is also assumed that passengers do not change their behav-
iour, except that any time savings can be used to produce income for busi-
ness passengers or leisure. Further it is assumed that the screening service
level does not change; that is, the level of security does not change. This
allows investigation of how service times, and hence costs, can change as

M4371-FINGER_9781786431851_t.indd 322 14/11/2017 11:14

the screening process changes, such as excluding more items and requir-
ing passengers to remove shoes, belts, laptops from cases and liquids
from carry-on bags. Costs and benefits are assumed to accrue to air pas-
sengers and not to the general public. Security screening for baggage or
freight is not considered.
Delays and variations in delays in completing the security screening
process will impose costs on two economic agents: airlines and passengers.
The added cost of delay is the opportunity cost of a passenger’s time plus
any disutility associated with waiting in line longer than is needed to com-
plete the screening process. For example, if the agreed service standard is
‘screening should take no more than five minutes’, the flow time should
be no more than five minutes; it is the time in excess of five minutes that
is the basis of cost measurement. In addition, if there is variability in the
flow time the system is not considered robust by passengers; therefore,
they will be forced to self-insure and arrive at the airport sooner than they
would have otherwise, using up valuable time. Thus, the time measure is the
expected flow time composed of the processing time including the average
time in the queue plus the insurance time, which is the variance in process-
ing time. These time quantities can be monetized by using value of time
and value of reliability measures.
The data requirements for this type of study would be:
●● the average queue length;

●● the average processing time;
●● the variance in processing time;
●● the arrival rate of passengers.
This data would be delineated by screening machine, by domestic, trans-

border and international passengers and by airport, for representative
days, including allowance for time of year, by 10–15-minute periods. One
would also require flight schedules by airline for domestic, transborder
and international destinations, and average load factor per flight for each
airline for representative days. One would need data on average departure
delays for each flight by each airline on representative days.
Turning to airline costs, these are costs associated with delayed depar-
tures or handling passengers and baggage that did not board their sched-
uled flight due to delays in the security screening process. The cost of
delays has been studied in the academic literature, examining delays at
airports due to congestion. These would be the source of this category of
airline cost due to screening delays. Handling missed passenger loading
is troublesome, since it requires measures of resource costs to the airline
of rescheduling a passenger or, if it is a connecting flight, handling new

M4371-FINGER_9781786431851_t.indd 323 14/11/2017 11:14

connections as well. If the airline was not able to fill the seat, one might
argue there is an opportunity cost to the airline of an unused seat.
For airports, there are costs of space used in processing passengers and
the incremental costs of space if additional screening machines and per-
sonnel are added. One must take account of any costs to an airport related
to contracting security services and how these costs vary by number of
screeners and machines, plus any other direct and indirect costs associated
with security screening.
FINANCING AVIATION SECURITY: SECURITY-

RELATED CHARGES
Most countries around the world recover at least a portion of the costs of
aviation security through charges levied on airlines, which are passed on to
passengers through ticket prices. We uncover, in a comparable way, avia-
tion security charges for a sample of 60 airports (Table 16.1). To do this, a
proprietary data set on airport charges, compiled by RDC Aviation Data
Insight is accessed.9 This data set is principally used by airlines for budget-
ing purposes, but it allows calculation of stated aviation security charges
while controlling for equipment used, load factor, dwell time, date (charges
were all calculated for the same date in May 2014) and exchange rates.
The airports in our sample can be compared by country, annual passenger
volume and governance structure. The data in Table 16.1 shows the high
variation in published aviation security charges between countries, with
Canada having by far the highest charges, at over $25 per international
passenger.
China, Spain and the US have roughly uniform charges, while variation
in charges between airports is evident in Canada and Germany. Figure 16.7
shows the relationship between stated airport security charges and annual
passenger volumes. The figure indicates no discernible pattern linking
the airport size and aviation security charges. For example, San Diego
and Atlanta levy the same security charge but differ widely in passenger
volumes. Frankfurt and Vienna also have comparable security charges but
also differ significantly in passenger volumes.
An alternative explanatory variable for security charges is governance
structure, that is, airports with the same ownership structure may set
similar security charges. Figure 16.8 compares security charges by govern-
ance structure and it is evident that there is still variation across different
ownership classes. Here two types of variation emerge: within-country
(Canada’s airport authorities) and between-country variations (US public

M4371-FINGER_9781786431851_t.indd 324 14/11/2017 11:14

Table 16.1 Airport security charges and other passenger charges by airport (2014 $)
Airport Country Annual Pax Governance Stated Pax Stated Govt Total Stated Unspecified
Vol 2013 Security Security Security Pax Charges
Charge Charge Charges
SYDNEY, AU (SYD) Australia 38 254 039 Private $ – $ 52.66
M4371-FINGER_9781786431851_t.indd 325
BRISBANE (BNE) Australia 21 791 997 Private $ –
VIENNA, AT (VIE) Austria 21 999 820 Priv/Pub $ 10.76 $ 10.76 $ 72.50
BRUSSELS (BRU) Belgium 19 133 222 Priv/Pub $ 8.99 $ 8.99 $ 28.15
SAO PAULO (GRU) Brazil 36 460 923 Priv/Pub $ – $ 12.22
TORONTO ON, CA Canada 36 037 962 Airport Authority $ 25.91 $ 25.91
(YYZ)
VANCOUVER BC, Canada 17 971 883 Airport Authority $ 1.48 $ 25.91 $ 27.39
CA (YVR)
325
CALGARY (YYC) Canada 14 316 074 Airport Authority $ 25.91 $ 25.91
MONTREAL (YUL) Canada 14 095 272 Airport Authority $ 25.91 $ 25.91
EDMONTON (YEG) Canada 6 983 229 Airport Authority $ 2.80 $ 25.91 $ 28.71
OTTAWA (YOW) Canada 4 578 591 Airport Authority $ 2.63 $ 25.91 $ 28.54
HALIFAX (YHZ) Canada 3 585 864 Airport Authority $ 3.18 $ 25.91 $ 29.09
WINNIPEG (YWG) Canada 3 483 898 Airport Authority $ 25.91 $ 25.91 $ 12.96
SANTIAGO (SCQ) Chile 2 100 000 Public $ 5.06 $ 0.78 $ 5.84 $ 11.24
BEIJING (PEK) China 83 712 355 Public $ 2.03 $ 2.03 $ 11.86
HONG KONG, China 59 609 414 Public $ 4.44 $ 4.44 $ 16.14
HK (HKG)
SHANGHAI (PVG) China 47 189 849 Public $ 2.03 $ 2.03 $ 11.86
XIAMEN, CN (XMN) China 19 753 016 Public $ 2.03 $ 2.03 $ 11.86
COPENHAGEN (CPH) Denmark 24 067 030 Priv/Public $ 8.08 $ 1.12 $ 9.20 $ 8.21
CAIRO (CAI) Egypt 15 741 000 Public $ – $ 28.18

PARIS CDG (CDG) France 62 052 917 Priv/Public $ – $ 36.68
14/11/2017 11:14
Table 16.1 (continued)
Airport Country Annual Pax Governance Stated Pax Stated Govt Total Stated Unspecified
Vol 2013 Security Security Security Pax Charges
M4371-FINGER_9781786431851_t.indd 326
Charge Charge Charges
HAMBURG (HAM) Germany 13 484 628 Crown Corp $ 1.48 $ 4.99 $ 6.47 $ 13.80
FRANKFURT, DE Germany 58 036 948 Crown Corp $ 1.69 $ 9.18 $ 10.87 $ 34.16
(FRA)
BERLIN, DE (TXL) Germany 19 591 849 Public $ 0.92 $ 7.47 $ 8.39
DUSSELDORF (NRN) Germany 21 228 226 Priv/Pub $ 5.08 $ 5.08 $ 8.04
STUTTGART (STR) Germany 9 577 551 Priv/Pub $ 1.93 $ 6.58 $ 8.51 $ 15.86
MUNICH (MUC) Germany 38 672 644 Public $ 1.09 $ 7.44 $ 8.53 $ 25.19
ATHENS (ATH) Greece 12 944 041 Crown Corp $ 6.70 $ 6.70 $ 16.29
326
JAKARTA (CGK) Indonesia 59 701 543 Public $ – $ 13.97
TEL AVIV (TLV) Israel 14 227 612 Public $ – $ 32.11
ROME (FCO) Italy 36 165 762 Private $ 3.56 $ 3.56 $ 43.23
MILAN (LIN) Italy 9 229 890 Private $ 3.52 $ 3.52 $ 26.39
TOKYO (NRT) Japan 35 341 341 Crown Corp $ 5.56 $ 5.56 $ 22.36
KUALA LUMPUR (KUL) Malaysia 37 704 510 Crown Corp $ 1.82 $ 1.82 $ 17.94
SCHIPOL (AMS) Netherlands 52 569 250 Crown Corp $ 17.91 $ 17.91 $ 21.07
AUCKLAND, NZ (AKL) New Zealand 14 862 485 Crown Corp $ 9.89 $ 9.39 $ 25.42
OSLO (OSL) Norway 22 956 540 Crown Corp $ 5.17 $ 5.17
MOSCOW (DME) Russia 30 760 000 Private $ 6.30 $ 6.30
RIYADH (RUH) Saudi Arabia 18 300 000 Public $ 1.10 $ 1.10 $ 13.34
SINGAPORE (SIN) Singapore 35 840 000 Public $ 6.26 $ 6.26 $ 15.58
CAPETOWN (CPT) South Africa 8 434 799 Crown Corp $ 1.57 $ 1.57 $ 48.37

SEOUL INCHEON South Korea 40 273 068 Public $ – $ 15.52
14/11/2017 11:14
MADRID (MAD) Spain 39 710 903 Public $ 0.81 $ 5.28 $ 6.09 $ 31.27
PALMA MALLORCA Spain 22 768 082 Public $ 5.28 $ 0.81 $ 6.09 $ 13.82
M4371-FINGER_9781786431851_t.indd 327
(PMI)
BARCELONA (BCN) Spain 35 210 735 Public $ 5.06 $ 0.78 $ 5.84 $ 19.71
BANGKOK (BKK) Thailand 51 363 451 Crown Corp $ – $ 22.06
ISTANBUL (IST) Turkey 37 406 025 Private $ – $ 15.00
DUBAI (DXB) UAE 50 977 960 Public $ 1.42 $ 1.42 $ 21.30
LONDON, GB (LHR) UK 72 368 030 Private $ – $ 62.30
MANCHESTER (MAN) UK 20 751 581 Public $ 7.27 $ 7.27
LONDON, GB (LGW) UK 34 200 000 Private $ – $ 19.26
GLASGOW (GLA) UK 7 363 764 Private $ 0.03 $ 0.03 $ 18.42
327
BIRMINGHAM (BHX) UK 9 120 201 Private $ 0.55 $ 0.55 $ 17.87
ATLANTA (ATL) USA 94 431 224 Public $ 2.61 $ 2.61 $ 35.88
LOS ANGELES (LAX) USA 66 667 619 Public $ 2.61 $ 2.61 $ 35.88
DALLAS (DFW) USA 60 436 739 Public $ – $ 35.88
NEW YORK (JFK) USA 50 423 765 Public $ 2.61 $ 2.61 $ 35.88
SAN FRANCISCO (SFO) USA 45 011 764 Public $ 2.61 $ 2.61 $ 35.88
MIAMI FL, US (MIA) USA 40 563 071 Public $ 0.64 $ 2.61 $ 3.25 $ 35.88
SAN DIEGO CA, US USA 17 710 241 Public $ 2.61 $ 2.61 $ 35.88
(SAN)
Source: Based on RDC Aviation.

14/11/2017 11:14
$35.00
Canadian airports
$30.00
Total delineated charges (2014 $US)
$25.00
$20.00
Schiphol
$15.00
Vienna
San Diego Frankfurt
$10.00
$5.00 Atlanta
$–
0 20 000 000 40 000 000 60 000 000 80 000 000 100 000 000
Annual passenger volume
Figure 16.7 Airport security charges by annual passenger volume, 2014
$35.00
security charge (2014 $US)
Total delineated airport
$30.00
$25.00
$20.00
$15.00
$10.00
$5.00
$0
Airport Authority
Airport Authority
Airport Authority
Airport Authority
Crown Corp
Crown Corp
Crown Corp
Crown Corp
Crown Corp
Priv/Pub
Priv/Pub
Private
Private
Private
Public
Public
Public
Public
Public
Public
Public
Public
Public
Public
Governance Structure
Figure 16.8 Delineated airport security charges by governance structure,

2014

M4371-FINGER_9781786431851_t.indd 328 14/11/2017 11:14

airports all levy essentially the same charge, but this differs from the
charges levied by public airports in other countries).
The figures provide some indication of the variance in stated passen-
ger charges for airport security across airports, and they appear to show
Canada’s airports as charging significantly more than airports elsewhere.
However, many airports do not explicitly state their security charges,
and even those that do have explicit security charges also have ‘passen-
ger service fees’ that could include additional security-related fees. For
example, London Heathrow Airport (like almost all UK airports) does not
specify a passenger security charge; however, the airport does levy a pas-
senger service charge of $62.30. We have been unable to determine which,
if any, security-related charges are included in this fee. Similarly, Brussels
International Airport delineates a security charge of $8.99, but also has
unspecified passenger service charges of $28.15.
Canada’s airports are the only ones that do not have any unspecified
passenger service charges. Therefore, it seems likely that at least part of the
seemingly high charges for airport security in Canada is overstated relative
to charges in other countries. The lack of transparency in data on airport
charges for security worldwide is a real problem that prevents any proper
analysis and comparison.
GOVERNANCE, OVERSIGHT AND PUBLIC VERSUS

PRIVATE MARKET SOLUTIONS
For centuries, economists and philosophers have debated the benefits of

free-market institutions versus government control in the economy. Such
debates have left their mark on aviation – and by extension on aviation
security. Internationally, the predominant trend in civil aviation over the
last three decades has been towards market-based solutions and associated
governance structures. This has not only meant the lifting of legislative and
regulatory barriers to private market competition (liberalization and ‘open
skies’ agreements), but also in many cases it has given rise to the creation of
new regulatory rules and agencies to insure against private-market failures.
The goal of liberalization and privatization has been to reap the benefits
of private markets whilst ensuring against the dangers of free markets
brought about by externalities and the abuse of market power by industry
participants. Interestingly, rather than an evolution towards a common
model or approach, we observe a variety of governance structures and
regulatory approaches around the world.
Many airports around the world have been devolved from direct gov-
ernment control and operation to be replaced by a number of different

M4371-FINGER_9781786431851_t.indd 329 14/11/2017 11:14

g overnance structures. In some jurisdictions (e.g. Australia and UK) air-

ports have been fully privatized, while in others (e.g. Germany and France),
they have been partially privatized. In Canada, the management and
operation of airports was devolved from direct federal government control
and operation to be operated by not-for-profit ‘local airport authorities’.
The airline industry has also undergone large-scale deregulation, and along
the way many former state-run airlines have become private firms. Airline
business models have evolved significantly, the emergence of ‘low-cost car-
riers’ having fundamentally changed the business strategies and operations
of all airlines.
Lastly, international law governing aviation has trended towards liberal-
ization, with ‘open skies’ agreements replacing bilateral rules that limited
routes and airline participants. Nevertheless, state-owned and operated
airlines and airports have not been eradicated – in 2010, 78 per cent of
Europe’s airports were publicly owned – and civil aviation displays many
different governance structures across national and regional boundaries.10
The same is true in aviation security, with different combinations of gov-
ernment departments, quasi-government agencies, private for-profit firms
and non-profit enterprises involved. This raises the question: ‘Is there a
uniquely optimal way to provide aviation security, or does it depend upon
jurisdiction-specific variables?’ For example, the fundamental laws that
govern individual rights and freedoms differ between countries and there
are different structural, economic and cultural realities across countries
and regions. Such differences are likely to create barriers to having a
common model for delivering aviation security. This implies that beyond
the need for international standards governing the minimum desired level
of aviation security, international laws and/or guidelines must also define
equivalency in the delivery of aviation security. That is, international
standards should be output-based rather than input-based. This is a chal-
lenging task because of the difficulties in measuring output in aviation
security.
The Economic Justification for Public Financing of Aviation Security
The bombing of Pan Am flight 103 over Scotland in 1989 resulted in 11

non-passenger fatalities in the town of Lockerbie. Although the main
targets of the attack were not civilians on the ground, the non-passenger
casualties could have been much worse, as Lockerbie is just 75 miles from
the far more densely populated city of Glasgow. However, the events of
September 11th 2001 demonstrated the large-scale devastation and loss
of life possible when civilian aircraft are deliberately used as weapons
to target densely populated targets on the ground. Clearly, the potential

M4371-FINGER_9781786431851_t.indd 330 14/11/2017 11:14

Table 16.2 Direct and indirect costs measured in estimate of 9/11 impact
Direct Costs Indirect Costs

Human lives Lost employee income and business
Property loss profits
* Buildings * Days business closed or services cut
* Technology and fixtures back because office infrastructure
* Subway stations damage or destroyed
* Phone and power utilites * Other firms that depend on those
Response to the emergency that are closed or cut back
* Emergency management (including Reduced tax revenues
loss of equipment) Delays to travellers and commuters
* Debris removal
* Building stabilization
Health effects, injuries, and emotional
distress
Temporary living assistance
Source: Cordes et al. (2006).
destruction from attacks on aviation is not limited to the passengers, crew

and aircraft or those using or working in airport facilities. Table 16.2
highlights the (non-aviation) direct costs and indirect costs associated with
the 9/11 attacks as calculated by Cordes et al. (2006), who estimated that
these costs totalled $83 billion. This illustrates the significant benefits to
the broader (non-air traveller) community of preventing such an attack.11
Consequently, IATA and other organizations representing the civil avia-
tion sector have argued for many years that aviation security is the respon-
sibility of national governments and that the current situation has an
inequitable and damaging impact on the air transport industry and, more
generally, on the economy.
In economics, situations in which third parties either benefit or suffer as
a result of the actions of others are called ‘externalities’. Activities which
create externalities are characterized as having either social benefits or costs
over and above the private benefits and costs experienced by those engaged
in the activity. Such activities are said to have ‘public good’ (or ‘public
bad’) characteristics. Specifically, activities are public goods (bads) when
the benefits (costs) they create are ‘non-excludable’ and ‘non-rival’. The
term ‘non-excludable’ means that one individual or group of individuals
cannot exclude others from also realizing the benefits (costs) resulting from
the activity. The term ‘non-rival’ means that one person’s consumption

M4371-FINGER_9781786431851_t.indd 331 14/11/2017 11:14

does not preclude consumption by others. As a standard example of

a ‘public bad’ we understand that the private activities of firms whose
production activities create pollution, if left unchecked, will not take into
account the negative effect of their activities on the larger environment.
Education is a classic example of a good which confers private benefits on
those who consume it, but also confers broad benefits to society that would
not be taken into account by the private market. Although an advocate of
free markets, economist Milton Freedman put it this way:
A stable and democratic society is impossible without widespread acceptance

of some common set of values and without a minimum degree of literacy and
knowledge on the part of most citizens. Education contributes to both. In con-
sequence, the gain from the education of a child accrues not only to the child
or to his parents but to other members of the society; the education of my child
contributes to other people’s welfare by promoting a stable and democratic
society. (Friedman, 1955)
Thus, there is an economic justification for governments to use tax rev-

enues to help finance education so that the socially desirable (socially effi-
cient) level of education is produced and consumed by our citizens.
Many goods may confer benefits on a wider community beyond the
private benefits enjoyed by those who consume them directly. However,
this is not a sufficient condition for government intervention. For example,
an indoor shopping mall is a private market venture involving a com-
mercial property developer and several privately-owned stores. However,
indoor shopping malls are often climate-controlled to provide heat (air
conditioning) during cold (hot) weather. This climate control is what we
would term a ‘local public good’: every shopper in the mall benefits from
the central heating. The enjoyment of the climate by one shopper does not
prevent others from also enjoying it (non-rival), and no one can exclude
some shoppers from enjoying it (non-excludable). Yet in this case there
is no need for the government to intervene for the optimal provision of
climate control in a shopping mall. Optimal provision occurs because it is
in the mall’s best interest to have contented shoppers, and the costs of the
climate control are recovered through the lease payments charged to each
store. Finally, the stores factor these costs into the prices they charge their
customers. An important difference between this example and the provi-
sion of aviation security is that there are no significant broader benefits to
society resulting from climate control in a shopping mall. Consequently,
the private market is capable of allocating resources efficiently and extract-
ing the value that is created through private market prices.
Thus, to the extent that aviation security has public good characteristics,
there is an economic case for the government to intervene and ensure that

M4371-FINGER_9781786431851_t.indd 332 14/11/2017 11:14

The Optimal Quantity of Aviation Security

P
Sp
Ssub
P1
Subsidy
P2
Ds
Dp
Qp Qs Q
Source: Coughlin et al. (2002).
Figure 16.9 Aviation security as a public good
the socially efficient amount of aviation security is provided. Coughlin et

al. (2002) illustrated the basic problem as in Figure 16.9.
In the figure, Sp represents the private market supply of aviation security.
It is upward sloping because providing additional security will increase
costs, requiring a higher price to induce that level of output. The demand
curve Dp represents the private market demand for aviation security (by air
travellers). The demand curve is downward sloping because the marginal
private benefits of security fall as the level of security increases. The inter-
section of Sp and Dp thus shows the quantity of aviation security (Qp) that
would result if everything were left to the private market. However, since
aviation security also benefits society, the social value of any given level
of security exceeds its private value, which gives rise to the social demand
curve Ds (which lies everywhere above private demand Dp). The addition
of social demand curve Ds illustrates that the private market will provide
a level of security that is lower than the socially efficient amount (Qs).
Finally, the diagram shows supply curve Ssub indicating that the govern-
ment can induce the private market to provide the socially optimal level of
security by providing a subsidy to the private providers of security equal
to (P1 − P2) per unit of security. This has the effect of shifting down the
private supply curve.12
Despite the public good arguments presented above, the predominant
trend appears to be towards recovering most, if not all, of the costs of
aviation security by levying taxes exclusively on air travellers (the user-pays

M4371-FINGER_9781786431851_t.indd 333 14/11/2017 11:14

principle). As noted in Button (2016), the driving force behind changes

in the financing of aviation security appears to have been expediency in
responding to new security concerns and notions of equity in terms of who
is able to pay (a perception, perhaps, which in large part is erroneous, that
all air travel demand is price inelastic). In Canada, the ‘air travelers secu-
rity charge’ (ATSC) is one of the highest airline passenger security taxes
in the world. The Canadian National Civil Aviation Security Program, in
outlining the financial framework for aviation security, justifies funding
aviation security entirely from taxes on air travellers by emphasizing that
‘The ATSC is payable by air travellers, who principally and directly benefit
from the Canadian air travel system.’ (Government of Canada, 2013).
The economic problem with such statements is that they could be equally
applied to the education sector, and yet it is unlikely that the Canadian
government would ever argue that students should pay all the costs of their
education since they are ones who principally and directly benefit from the
education they receive.
CONCLUSIONS
We have argued that aviation security is not immune from fundamental

resource allocation problems. This means that decisions must be made
about how to allocate scarce resources efficiently and how to generate
the maximum net social welfare. Making such decisions requires data,
and despite our best attempts to find and analyse relevant data, we have
been struck by the lack of available data – and where data does exist, by
the lack of consistency and transparency. Future research into problems
related to aviation security could make significant contributions if inter-
national standards of publishing, reporting and delineating data relating
to aviation security were developed both at the international, national and
individual airport level of analysis. We see the potential for more research
into measuring output in aviation security (how safe are we?) and for more
extensive use of benefit–cost analysis to determine the performance of
security measures or programmes (groups of measures). But again, data
is required. With more and better quality data, analysis can help us better
understand cost relationships and the economic consequences of different
financing mechanisms.
In terms of the efficiency of aviation security, it should be noted that
efficiency is different from ‘effort’, and efficiency matters because a more
efficient aviation security system can reduce the indirect economic costs to
society while improving the benefits and the overall safety of the travelling
and general public. With access to data, more research can provide better

M4371-FINGER_9781786431851_t.indd 334 14/11/2017 11:14

estimates of how screening service levels relate to technology investments,

human capital investments and the time-costs imposed on passengers, air-
lines and airports.
On the financing side of aviation security, the trend towards increased
reliance on security charges levied on air travellers is troubling and unjus-
tified on economic grounds. The rationale appears to have been ‘we can
finance aviation security with a user fee because air travellers can and
will pay’; however, this approach disregards the external economic costs
imposed on aviation and the economy more generally.
There seems to be little doubt that the current standard implementa-
tion of aviation security at airports around the world will have to change,
given projected growth in air travel and the growing number of passengers
passing through airports. Specifically, we cannot continue to work under
the assumption that every passenger represents an equal threat to aviation.
A move towards risk-based systems is an inevitable consequence of eco-
nomic pressures compounded by current institutions which have evolved
from hastily taken decisions in the wake of the 9/11 attacks.
International industry associations (notably IATA and ACI) have been
developing and promoting a coordinated move to ‘next-generation’ avia-
tion security. Working independently until recently, IATA named its vision
for aviation security ‘checkpoint of the future’, while ACI developed a
concept it called ‘better security’. This year at ICAO meetings in Montreal,
IATA and ACI announced a memorandum of understanding to harmo-
nize their work and to jointly promote the next generation of aviation secu-
rity, with a focus on ‘airline–airport interface, airport throughput capacity
and efficiency’.13
Annexes to the agreement are now being finalized to focus on the
following improvement areas:
●● passenger flow at border crossings based on the Automated Border

Crossing project;
●● passenger screening processes at targeted security checkpoints to
maximize efficiency and productivity, as well as minimizing passen-
ger dissatisfaction;
●● gaining support from airports and national regulators to build on
the achievements of the Checkpoint of the Future project;
●● airline–airport cooperation on Common Use Self Service;
●● common technical specifications for data exchange standards at the
airport;
●● best practices in ground handling to drive improvements in safety,
productivity and reduce overall risks;

M4371-FINGER_9781786431851_t.indd 335 14/11/2017 11:14

●● reducing mishandled bags and offering new products to passengers,

including permanent bag tags and home-printed bag tags.
The challenge is to engineer a coordinated international evolution

to risk-based aviation security. Many of the problems faced at the
individual jurisdiction level are compounded with coordination prob-
lems of moving to an international standard model for risk manage-
ment and adoption of new technologies. Such coordination creates
prisoner’s-dilemma-type incentives at the international level. There is an
incentive for each nation or jurisdiction to delay implementation of the
new measures to wait and see the outcome of other countries’ adoption
of the new model. If all countries follow this incentive, then delay is the
outcome.
A central problem with plans as envisaged by IATA and ACI is the lack
of data and analysis with which to measure expected outcomes and net
benefits, along with costs and net efficiency changes. IATA and ACI have
begun the process of pilot studies, but there are no published results at
the time of writing. Another problem relates back to the issue of adding
new layers of security to existing ones. Future risk-based systems will be a
hybrid of new security layers and existing security layers, which implies a
reallocation of resources as new layers replace some existing layers, while
other existing layers remain or are augmented. The degree of complemen-
tarity and substitutability between layers will play an important role in
determining the effectiveness of security, the efficiency of throughput and
the quality of traveller experience.
Differences in the way that individual nations choose to design, organize
and implement the security systems of the future will give rise to differ-
ences in effectiveness, efficiency and cost. Thus, we need more data and
analysis to understand better exactly how different layers of security relate
to one another and how adding, removing, increasing or decreasing layers
would affect the whole system. As argued, the danger is that in the absence
of compelling analysis and data, political incentives will push us towards
the addition of new security layers with minimal adjustments to and no
removal of existing layers.
ACKNOWLEDGEMENTS
We wish to thank Kathleen Rodenburg and Jennifer Ng for their excel-

lent research assistance. Funding for this research was provided by the
Kanishka Project Contribution Program, administered by the Government
of Canada’s Department of Public Safety. The Kanishka Project was

M4371-FINGER_9781786431851_t.indd 336 14/11/2017 11:14

initiated through the efforts of victim families of the Air India tragedy of
1985, in which 329 innocent people lost their lives.
NOTES
1. Using something like ASMs (available seat miles) as a measure of the opportunity for
incidents would bias (upward) the measure of safety.
2. It is unlikely that the metric could be calculated for an airport, since data is too thin.
3. A flight would not be counted that may be damaged or blown up due to an air-to-
ground rocket attack.
4. In the US, the Transportation Security Administration is part of the Federal Department
of Homeland Security, whereas the Canadian Air Transport Security Agency is an inde-
pendent agency established by, but at arm’s length from, the federal government. In the
US most security screening agents at airports are federal government employees, with
some airports using contractors under the Screening Partnership Program, whereas
in Canada they are employees of private security companies contracted by individual
airports.
5. Details of the econometric results can be found in Gillen and Morrison (2015).
6. Aviation Security in the European Union is set out in EC Regulation No. 300/2008,
which came into full effect on 29 April 2010.
7. A good example of differences in variability in screening processes is screening people
versus screening baggage; baggage moves at a constant and higher speed.
8. It is often pointed out that many more people died on the ground than on aircraft as a
result of the 9/11 attacks.
9. Source: http://www.rdcaviation.com/.
10. Airports Council International Europe report that publicly owned airports handled 52
per cent of Europe’s passenger traffic. This in part indicates how private investment has
been drawn to larger airports in Europe.
11. In addition to the indirect costs listed in Table 16.2 we can add negative economy-wide
effects such as stock market reaction, insurance rates and potential suppression of
foreign direct investment through changes to perceived risk.
12. The diagram does not proscribe governance issues, that is, there is no argument pro-
vided here to support government provision (separate from government financing) of
aviation security.
13. IATA (October 2013). Available at: https://www.iata.org/publications/ceo-brief/oct-
2013/Pages/aci-iata-mou.aspx.
REFERENCES
Button, K.J. (2016) The Economics and Political Economy of Transportation

Security, Cheltenham, UK and Northampton, MA, USA: Edward Elgar
Publishing.
Cordes, J., A. Yezer, G. Young, M. Foreman and C. Kirschner (2006) Estimating
Economic Impacts of Homeland Security Measures, George Washington
Institute of Public Policy, Working Paper No. 022.
Coughlin, C., J. Cohen and S. Khan (2002) Aviation Security and Terrorism: A
Review of the Economic Issues, Federal Reserve Bank of St. Louis Review, 84,
pp. 9–25.

M4371-FINGER_9781786431851_t.indd 337 14/11/2017 11:14

Friedman, M. (1955) The Role of Government in Education, in R.A. Solo (ed.),

Economics and the Public Interest, New Brunswick, NJ: Rutgers University Press.
Gillen, D. and W. Morrison (2005) The Economics of Franchise Contracts and
Airport Policy, Journal of Air Transport Management, 11, 43–8.
Gillen, D. and W. Morrison (2015) Aviation Security: Costing, Pricing, Finance
and Performance, Journal of Air Transport Management, 48, 1–12.
Government of Canada (2013) Canada’s National Civil Aviation Security Program.
Available at http://www.tc.gc.ca/media/documents/security/NCASP_FINAL_
ENGLISH_(2).pdf.
Jackson, B., T. LaTourrette, E.W. Chan, R. Lundberg, A.R. Morral and D.R.
Frelinger (2012) Efficient Aviation Security: Strengthening the Analytic
Foundation for Making Air Transportation Security Decisions, Santa Monica,
CA: Rand Corporation, Homeland Security and Defense Center.
Stewart, M. and J. Mueller (2008) A Risk and Cost–Benefit Assessment of United
States Aviation Security Measures, Journal of Transportation Security, 1, 143–59.

M4371-FINGER_9781786431851_t.indd 338 14/11/2017 11:14

17. The need to evolve air traffic
management: Europe as a
laboratory
Matthias Finger, Marc Baumgartner and
Engin Zeki
Air navigation services provision and its main components, notably air
traffic management (ATM), is one of the three elements of civil (and mili-
tary) aviation, along with air transport and airports. Together, these three
elements form the complex and dynamic socio-technical aviation system.
In this chapter, we will focus on the core function of air navigation ser-
vices provision, namely ATM. ATM is typically performed exclusively at a
national level by a so-called air navigation services provider (ANSP) that,
with very few exceptions, is a national entity with a level of autonomy that
varies from country to country. However, our focus is not on the different
ANSPs and their reforms and transformations. Instead, we look at the core
function of air navigation services provision, namely ATM, and examine
how ATM evolves, could evolve, or would need to evolve, both in parallel
to and in support of rapidly growing air transport, transforming airport
operations, and the changing entire socio-technical aviation system.
There is a need for ATM to evolve. Solely from an economic point of
view, it can easily be seen that almost all elements of the aviation value
chain have been optimized, ranging from aircraft producers (airplanes are
now standardized products) to highly efficient air transport operations,
to optimized aircraft maintenance, to highly successful yield management
techniques, to increasingly efficient airport operations, and so on. ATM is
probably one of the last, if not the last element of the aviation value chain
waiting to be optimized, that is, being looked at from an economic perspec-
tive rather than just from safety and national sovereignty points of view.
This argument has been made by many authors who have subsequently
tended to advocate for various forms of liberalization of ATM and air
navigation services provision, yet generally end up in simplistic and, in our
view, ideological solutions when calling for the privatization of ANSPs or
some other free market constructs (see ATM Policy Institute, 2016).
339

M4371-FINGER_9781786431851_t.indd 339 14/11/2017 11:14

Here we do not take this point of view and instead take a (socio-
technical) systems approach. Such an approach not only helps us see that
air transport (civil and military), airports, and air navigation services form
a single system, but, more importantly, recognizes the increasingly inte-
grated global or at least regional (for example, North America, Europe,
South East Asia) nature of such a system, to the point that current,
national air navigation services provision is not only anachronistic, but
also inefficient and sub-optimal. Accordingly, we focus on Europe as an
integrated aviation region and on the efforts of the European Commission
(EC) since 2000 to create a so-called ‘Single European Sky’ (SES), mainly
by way of rules and regulations, as the EC typically does. Such efforts have
been a direct response to the recognition of Europe’s overly fragmented
airspace in light of increasingly integrated (European) air transport and
subsequent traffic growth. In this sense, Europe – particularly its SES
initiative – clearly constitutes the only laboratory of evolving ATM.
However, as we will also see, the EC’s SES initiative has ended up in grid-
lock (Baumgartner and Finger, 2014), leaving the problem unsolved, but
also, because of the pressing need to evolve ATM, opening up new and
more technologically oriented avenues for adapting ATM to the needs of
the European and global aviation systems.
Therefore, this chapter aims to offer a comprehensive and systemic
understanding of the challenges facing ATM today. We look at the case
of Europe, where the differing interests of the main actors, namely the EC
and some member states, have led to a trial-and-error approach, making
Europe an interesting laboratory of evolving ATM thanks to an elaborate
system of European-wide rules and regulations. The remainder of the
chapter is structured as follows. In the first section, we define the issue and
explain the main functions of ATM. In a second section, we will briefly
outline the conceptual framework underlying our approach to evolving
ATM. In the third section, we describe how ATM evolved up to the year
2000, which marked the beginning of the Single European Sky project. The
fourth section presents the EC’s SES initiative, which we call the ‘labora-
tory of ATM’, along with its gridlock today. In the fifth section we identify
the emerging, more technology-focused attempts as a way out of this grid-
lock. The final section concludes.
WHAT IS ATM?
From the moment the doors of a commercial flight close at the departure
gate until they open again at the arrival gate, the crew is controlled by
an air traffic controller who guides the aircraft through the airspace. A

M4371-FINGER_9781786431851_t.indd 340 14/11/2017 11:14

The need to evolve air traffic management 341
standardized approach to all phases of a flight was defined in Article 28

of the Chicago Convention (1944), thus precisely defining ATM services
(see below). Accordingly, every civil flight must deliver a flight plan indi-
cating the intention of the flight, taking into account certain constraints.
Consequently, ATM offers three types of services: air space manage-
ment; air traffic services (including air traffic control); and air traffic flow
management.
●● Air space management (ASM) pertains to the management of

the airspace between the different users, namely civil and military
aviation but also vehicles such as hang-gliders and, in the future,
certain types of drones. The core function here is the definition of
the ‘airways’ or ‘highways in the air’, and this happens in two steps.
First, there is a strategic and political definition of the usage of the
airspace, identifying the portion of the airspace that is reserved
for the military. Military airspace typically takes up vast areas of
national airspace. In the second step, the concrete usage of the air-
space is adjusted, typically on a daily basis, to the needs of the differ-
ent airspace users. In the future, one can imagine even more flexible
definitions of the airspace, that is, adjustments within even shorter
timeframes.
●● Air traffic services (ATS), and in particular air traffic control (ATC),
provide safe, efficient and orderly flows of traffic in their flight infor-
mation region (FIR), which, in most cases, is identical to the airspace
above the national borders. For each flight, a so-called flight plan is
submitted (to the ANSP). This plan states the intention to fly from
A to B and thus contains the departure and the destination, the call
sign, the type of aircraft, the altitude and the speed of the aircraft,
and the planned route. As the flight progresses, this flight plan is
passed along the different so-called ‘control sectors’, which are then
grouped into so-called ‘ATC units’. Indeed, each FIR is divided into
different control sectors, which are basically the units within which
ATC is performed by air traffic controllers. Europe is currently
divided into more than 500 control sectors within 63 area control
centers and about 278 approaches, whereas the current number of
control sectors in the United States is 20 area control centers and 160
approaches (Eurocontrol, 2016). An air traffic controller, who typi-
cally handles between 10 (approach) to 30 flights (overflight) simulta-
neously, will then corroborate the information contained in the flight
plan with the up-to-date information that he or she derives from the
surveillance means at his/her working station. Where radar or other
forms of visual surveillance exist, this corroboration is done several

M4371-FINGER_9781786431851_t.indd 341 14/11/2017 11:14

times a minute. The core function here, as performed by the air traffic
controller, is to separate air traffic, that is, to separate the different
airplanes from one another within a given control sector and, by
doing so, to guide them safely from one control sector to the next
one. Over deserted regions and oceans, radar equipment is not avail-
able and other more procedural means of separation will be applied.
●● Air traffic flow management (ATFM) has been put into place
because each control sector or ATC unit – and each air traffic
controller for that matter – can only handle a limited number of
airplanes (typically 10–30) simultaneously. This leads to the defini-
tion of so-called ‘air traffic control slots’, of which there is a limited
number within each control sector. Furthermore, each airport also
has (physical) limits in terms of the number of airplanes that can
either take off or land, thus defining what is called ‘airport slots’.
This is in addition to the fact that airplanes can only remain in the
air for a limited amount of time for safety reasons and should fly
as directly as possible from A to B for both economic and environ-
mental reasons. Consequently, the capacity of both airspace (air
traffic control slots) and airports (airport slots) must be balanced
against the demands of aircraft wishing to fly. Such balancing is pre-
cisely the function of ATFM, which is performed simultaneously at
various levels. At the highest level, this is currently the responsibility
in Europe of the so-called ‘network manager’, that is, it is handled
operationally by an organization called Eurocontrol. However, bal-
ancing between capacity and demand also takes place at the FIR
level by the ANSP and at the control center level by the ATC unit.
While air traffic has increased exponentially, these three functions – defin-
ing the airways, separating aircraft, and balancing capacity and demand
– have remained constant. So too has the conceptual approach to, the prac-
tice of, and the technology for ATM. What has changed is the amount of
air traffic, which will continue to increase with the forecasted rapid growth
of unmanned air vehicles (UAVs). As ATC units (can) handle more or
less the same number of aircraft given the current (radar) technology, the
size of the control sectors has decreased – causing the number of control
sectors to increase steadily – to be able to handle this growing traffic. This
has led to exponentially growing interfaces between ATC units, diminish-
ing the overall performance of ATM per flight. Therefore, we will inevita-
bly reach a point where a qualitative change in the way ATM works will
become necessary. The EC had understood this during the late 1990s, but
has mainly looked at changing the rules governing ATM thanks to its so-
called Single European Sky (SES) project (see EC, 1996 and EC, 1999),

M4371-FINGER_9781786431851_t.indd 342 14/11/2017 11:14

which we will discuss below. However, changes in technology may also

need to be envisaged, because the SES has led to gridlock, as we will see
later, and because a change of rules may no longer be sufficient to address
growing air traffic. Before looking at these issues, we will briefly outline the
conceptual framework that enables us to better understand evolving ATM.
CONCEPTUAL FRAMEWORK
This section will briefly outline our conceptual framework. This framework
is based on the idea of a co-evolution between institutions (rules) and tech-
nology (Crettenand and Finger, 2013; Finger et al., 2005), as well as on the
assumption that there are certain critical technical functions that must be
fulfilled for this socio-technical system to work (Finger and Künneke, 2007).
In any socio-technical system, rules evolve – in principle, at least – in par-
allel to technology so as to make the system perform. Performance, in turn,
depends upon a series of critical technical functions that must be fulfilled.
For each of these critical technical functions there is a need for a certain
degree of alignment or coherence between the technology used and the rules
applied. The degree of coherence or alignment among these critical technical
functions will determine the overall performance of the system. This system
is also dynamic and its dynamism results from the (strategic) behavior of
the various actors involved, whether they be market actors, institutional
actors (such as regulators or policy-makers) or technological actors (such as
technology developers). Overall, institutions evolve from government (top-
down rules) to governance (negotiated rules among the relevant involved
actors). Similarly, and in parallel, technology evolves from more centralized
technologies to ever more distributed technologies, notably also thanks to
the pervasive nature of the information and communication technologies.
Figure 17.1 summarizes this broad conceptual framework.
We apply this conceptual framework to ATM. As we have seen in the
first section, there are three critical technical functions that must be ful-
filled for ATM to perform: defining the airways; separating aircraft; and
balancing capacity and demand. Each function is fulfilled via a combi-
nation of rules and technology. Both the rules and the technologies that
have been applied so far will be discussed in the next section. Here, we will
simply summarize the main drivers of the dynamics of this ATM socio-
technical system (see Figure 17.2):
●● Growing air traffic leading to the inability of the current system to

satisfy demand; in Europe, for example, such growing traffic is the
direct result of prior air transport liberalization (push).

M4371-FINGER_9781786431851_t.indd 343 14/11/2017 11:14

Institutions
(rules)
Institutional
Actors
Market
Coherence
Actors
Performance
Technology
Actors
Technology
Figure 17.1 The underlying conceptual framework
Institutional Institutional changes:

Actors Single European Sky
Balancing capacity and demand
Separating aircraft
Defining airways
Push: ATM
Market Market &
- Air traffic growth performance
Actors demand pull
- Efficiency concerns
Technological changes:
Technology SESAR
Actors
Figure 17.2 Summary of the conceptual framework as applied to ATM
●● Dissatisfaction with the performance of the current ATM system,

resulting either from traffic growth (above) or from inefficiencies in
the way ATM is provided (by ANSPs) or from both combined.
●● Pressure from policy-makers, basically aimed at changing the rules
(push), in the case of Europe; and pressure from the EC through its
Single European Sky project.
●● Pressure from some of the actors seeing new business opportunities,
basically as a result of new technological developments (pull).

M4371-FINGER_9781786431851_t.indd 344 14/11/2017 11:14

THE EVOLUTION OF ATM UP TO THE YEAR 2000
In this section, we will present the evolution of ATM up to the years

1999/2000, which represent the beginning of the Single European Sky
project (see Baumgartner, 2007). Indeed, soon after World War I ended,
nation-states made available their national military and telecommunica-
tion authorities (which employed professional radio operators) for the
relay and exchange of information among themselves. Subsequently, as
more and more aircraft took to the skies, the need to separate those air-
craft became paramount and national military and telecommunications
authorities were charged with this task.
Although the numbers of aircraft were not yet large, governments and
administrators realized that aviation needed to be regulated and that some
standardization should be applied. This was particularly important in
Europe with its multiple national boundaries and languages. The Versailles
Peace Treaty that ended the First World War was instrumental in launch-
ing the International Convention for Air Navigation (ICAN), which was
then signed by 19 nations and led the way to the development of so-called
‘General Rules for Air Traffic’.
Apart from ICAN rules and regulations, ATM probably has its earli-
est roots in the growth of London–Amsterdam/Brussels/Paris traffic.
Also, following the world’s first commercial mid-air collision on 7 April
1922 over France, measures were taken to ensure this would not happen
again. Consequently, in July 1922 the Croydon observation hut was built
with a glass upper story (incidentally, this was also the first time that the
term ‘control tower’ was used). Compulsory use of the wireless service at
Croydon by commercial operators, the introduction of the ‘Q Code’,1 and
the first use of Standard Departure Routes (SID) started in 1927. After a
demonstration of ‘instrument’ flying in 1929, the USA settled upon stand-
ards and in 1933 laid down Instrument Flight Rules (IFR). With this pro-
gress, the need to separate aircraft flying at night and in bad weather was
recognized. Formal rules by which ‘airway traffic control centers’ could
handle the traffic were subsequently developed, leading to such centers
being established at Newark and Chicago in 1935.
Later, blackboards and map tables gave way to paper flight progress
strips in movable holders on flight progress boards. In 1938, when the
US Civil Aeronautics Act established Civil Air Regulations, pilots were
required to comply with ATC instructions. By 1942, around 500 control-
lers manned IFR centers around the United States, covering more than
40 000 miles of air routes. The final major step in this progression came
in the early 1950s, when adequate air–ground communications facili-
ties enabled direct controller–pilot communications, thus eliminating the

M4371-FINGER_9781786431851_t.indd 345 14/11/2017 11:14

laborious and time-consuming practice of relaying all instruction via the

airlines dispatchers.
Whilst raw display of radar for terminal area use was first introduced
simultaneously in Sydney and Melbourne in 1959 (with the first en-route
radar center opening in Sydney in 1965), it was the Netherlands that was
at the forefront of the introduction of modern techniques to handle the
burgeoning traffic in Europe. The increasing speeds and higher cruising
levels of aircraft soon highlighted the limitations of early radar. New
surveillance radar2 was developed that provided coverage far beyond the
lateral boundaries of the new airways and to a height sufficient for the
emerging jet airliners. With the raising of the upper limit of their con-
trolled airspace from 15 000 feet to 25 000 feet and the earlier replacement
of Non-Directional Beacons (NDBs) with VHF Omnidirectional Ranges
(VORs), the Dutch controllers now had the tools they needed to satisfy the
fundamental demands of air traffic control.
The main aim of such tools is to assist the controller in determining
the past, current and future position of the aircraft. To reduce the uncer-
tainty (requiring large separations), the interaction between the Air Traffic
Control Officer (ATCO) and the pilot tries to achieve a greater predict-
ability of the current and, in particular, the future position of the aircraft.
For this, the flight plan is the central element that unites all the actors in the
operational environment. As mentioned, the flight plan is the intention of
a company or the pilot to fly from A to B and serves as reference along the
flight path. To establish the position of an aircraft, surveillance means (such
as radar over continental airspace or ADS-B3 over oceans or vast inhabited
land) are used. When surveillance means are not available or not practica-
ble, any form of communication is used to establish a pilot position report.
This basic approach has not changed over the years. What has changed
is the level of precision achieved in all the communications, both in naviga-
tion and in surveillance, allowing for all-weather operation and reduced
separation minima over the years; this, in turn, has increased the capacity
offered to the airspace users. Of course, all of this requires harmonization
of rules and standards.
The main actor of such harmonization is the so-called Convention on
International Civil Aviation (also known as the Chicago Convention),
which was signed on 7 December 1944 by 52 states. In 1947, the
International Civil Aviation Organization (ICAO) was officially created as
a specialized agency of the United Nations with the mandate to ‘develop
international civil aviation in a safe and orderly manner’. The Convention
is regularly adapted to evolving needs as a result of member states’ nego-
tiations and is currently in its ninth edition (dated 2006). Its annexes are
adapted to these changes individually.

M4371-FINGER_9781786431851_t.indd 346 14/11/2017 11:14

As for ATM, Article 28 of the Chicago Convention appears to be

most relevant, as it requires member states to provide air navigation
facilities over their territory, which comply with ICAO’s Standards and
Recommended Practices. More precisely, the purpose of Article 28 is to
secure a comprehensive, seamless and continuous network of air naviga-
tion services around the globe and to support the safety, regularity and
efficiency of international air transportation. Article 28 formalizes a com-
mitment by each contracting state towards the other parties to the Chicago
Convention. The responsibilities involved are essentially regulatory and
supervisory in nature. These responsibilities remain those of the contract-
ing states, even when they have mandated an autonomous or foreign entity
to perform the task of providing ATM.
However, the obligations for states under Article 28 are neither strictly
nor narrowly defined. The states retain considerable discretion and indi-
vidual appreciation regarding the means by which they intend to fulfill
their commitment, the level of infrastructure and service they wish to make
available to international aviation, and the degree of compliance of their
air navigation facilities with relevant ICAO regulations (Schubert, 2011).
Most importantly, under Article 28 of the Convention, a member state
has the obligation to provide, as far as practicable, an infrastructure that
meets the needs of an interoperable and seamless sky. The same state is also
responsible to allocate licenses to operate to airspace users.
Consequently, the ICAO has globally harmonized its approach to all
the domains of aviation including ATM. Of course, ATM has relied since
its inception on simultaneous and synchronized standards and technol-
ogy. However, technology and standards are being developed by different
actors. While standards are global in nature and should be harmonized (by
ICAO), technology is being developed by industry (manufacturers), typi-
cally for the different national air traffic service providers.
It is important to list the core elements that need to be standardized
across different technologies (and technology providers), both in terms
of functions and categories, to allow for ASM (air space management),
ATS (air traffic services) and ATFM (air traffic flow management),
the three types of services that make up ATM. All these elements have
remained unchanged since the late 1980s (ICAO, 2013) and can be grouped
around communication, navigation and surveillance (CNS) as the needed
infrastructure and ATM as the service provision based on this very
infrastructure.
As for communication, standards are needed for: (1) air–ground data
link communication; (2) ground–ground communication; and (3) air–
ground voice communication. As for surveillance, standards are needed
for: (4) surface; (5) ground-based; and (6) air-to-air surveillance. As for

M4371-FINGER_9781786431851_t.indd 347 14/11/2017 11:14

navigation, standards are needed for: (7) dedicated technology; and (8)
performance-based navigation. It is the presence or absence of such stand-
ards that determines the technological evolution in these eight areas, as
well as, more generally, the performance of ATM resulting from interoper-
ability and seamlessness at the global level (Sudarshan, 2003).
To achieve global interoperability and a globally seamless sky, it is very
important that all the signatory states to the ICAO follow its Standards
and Recommended Practices as closely as possible. With the advent
of the Single European Sky and its regulations and standards, it has
become crucial that ‘regional’ (European) standards match the global ones.
Therefore, the European Commission coordinates between ICAO, EASA,
the SESAR JU and Eurocontrol.4 If, under ICAO, a state has an opt-out
possibility, this is no longer possible under the SES. Therefore, it is essen-
tial that SES standards meet at least the ICAO standards and that, when
the SES standards go beyond the ICAO standards, they do not jeopardize
the global harmonization efforts of ICAO.
THE SINGLE EUROPEAN SKY, THE LABORATORY

OF ATM
In this section we will present the SES process, that is, the various steps
and measures by which the EC is trying to create a Single European Sky.
This is done in three steps: SES I, SES II, and the Performance Regime. We
will also discuss the most recent developments, specifically the European
Commission’s ambition to present a new so-called SES II1 package.5
Single European Sky I (2004)
As a result of the various driving forces outlined, the EC launched its

(first) Single European Sky Package in 2004, also called SES I. This
package took the form of four regulations: (1) EC Regulation 549/2004,
which laid down the framework for the creation of the Single European
Sky (the so-called Framework Regulation); (2) EC Regulation 550/2004
on the provision of air navigation services in the Single European Sky (the
so-called service provision regulation); (3) EC Regulation 551/2004 on the
organization and use of the airspace in the Single European Sky, including
the creation of so-called functional airspace blocks (FABs) (the so-called
airspace regulation); and (4) EC Regulation 552/2004 on the interoper-
ability of the European air traffic management network (the so-called
interoperability regulation).
SES I pursued a so-called ‘four-pillar approach’, which focused on four

M4371-FINGER_9781786431851_t.indd 348 14/11/2017 11:14

specific areas: (1) performance; (2) technology; (3) safety; and (4) capacity.
For technology a SESAR (Single European Sky ATM Research) master-
plan was drawn up, while EASA (the European Aviation Safety Agency) was
declared responsible for safety.6 As it does in all other sectors, the European
Commission periodically reviews the application of its actions (EC, 2007).
Consequently, it requested the newly created PRC (Performance Review
Commission) to assist in this evaluation. The PRC delivered its first report
in December 2006, recommending, among other things, the acceleration of
the FABs, corresponding technology development (SESAR), as well as the
empowering of Eurocontrol, especially in its pan-European functions and
ATM network design. As a result, in 2009 the Commission proposed a SES
II package. This package identified fragmentation as the major bottleneck
in improving the performance of the European aviation system.
Singe European Sky II (2009)
The SES II adopts a total system, gate-to-gate approach for air navigation
services via the performance scheme, a strengthening of FABs, as well as a
central network management function. It is based on closely interrelated
pillars that all converge towards the keystones of aviation performance and
sustainability. The safety pillar extends the competence of the EASA to
ATM and airports. The technological pillar, SESAR, aims to develop state-
of-the-art technology, supporting member states in their efforts to improve
performance, safety and sustainability. To that effect, a new organization
called SESAR-JU (joint undertaking) has been established. The airport
pillar, comprising the creation of the Airport Capacity Observatory, is
complemented by terminal ANS performance indicators as part of the
performance scheme. Finally, the human factor pillar is considered as
the ‘overarching SES pillar’, as it is acknowledged by all stakeholders that
the SES cannot be successful without the full buy-in and support of the staff
concerned.7
The Performance Dimension added (2010–)
The ‘Performance Report’, dated 2006,8 clearly stated that a Single

European Sky could only be achieved if a performance scheme was intro-
duced. Through EC Regulation 1070/2009, the foundation for an increased
performance in ATM was laid and Regulation 1108/2009 extended the remit
of EASA to ATM and CNS (Communication Navigation Surveillance). In
particular, Implementing Rule (EU) 691/2010 established a performance
scheme for air navigation services and network functions under the respon-
sibility of the Performance Review Body.

M4371-FINGER_9781786431851_t.indd 349 14/11/2017 11:14

The performance targets set out under the so-called ‘SES performance
scheme’ should reflect and foster improvements in all five SES pillars, as
well as from individual FABs, member states, ANSPs, airports and air-
space users. There are three specific targets: (1) the environmental target,
which will ensure carbon-neutrality of aviation growth insofar as en-route
ANS is concerned; (2) the capacity target, which is designed to avoid major
disruptions and indirect costs for airspace users and their passengers from
ANS delays; and (3) the cost-efficiency target, which will secure progres-
sive improvement in unit rates and maintain route charges nearly constant
over the period. A second reference period started in January 2015 and will
last until the end of 2019.9 The first year of RP2 (Reference Period 2) has
passed and the corresponding monitoring report was published during the
first quarter of 2017.10
Towards a SES II1
SES II1 was announced explicitly by Commissioner Siim Kallas in his

speech in Limassol in October 2012, entitled ‘10 years and still not deliv-
ering’.11 In it, Kallas noted that the targets, especially in matters of costs
and delays, have not been reached and that €5 billion is being wasted
annually because of inefficiencies. He also announced that new ambi-
tious and evidence-based performance targets will be set for the second
reference period. Finally, he threatened infringement proceedings against
FABs and member states, which will be held accountable for not meeting
their performance targets. Furthermore, ANSPs were explicitly labeled
as an impediment to progress, as was the confusion between regulatory
and oversight tasks and operational and service provision activities.
Ultimately, Kallas said, the SES ‘would lead to a single aviation authority
and an industry-led European infrastructure manager, with centralized
functions and services.’ Furthermore, ‘the European Aviation Safety
Agency should progressively cover all aspects of technical rules and over-
sight in ATM, not just safety.’ This, in turn, would allow Eurocontrol to
focus on running the network, take on infrastructure management and roll
out SESAR deployment.
On 11 June 2013 the European Commission published COM 408/2013
under the heading ‘Accelerating the implementation of the Single European
Sky’ or SES II1. The declared intention of the EC was to reaffirm the
objectives of the SES and work on the performance of air navigation
services, by enforcing and improving the existing rules. This was expected
to lead to enhancing the efficiency of the SES by orienting the focus of
the ANSPs on customer needs, particularly the delivery of performance,
increasing efficiency of the support services and strengthening of the

M4371-FINGER_9781786431851_t.indd 350 14/11/2017 11:14

National Supervisory Authorities (NSA). The proposed regulation (EC

COM 2013/410 – COD 213/186) further aimed to remove the fragmenta-
tion of the European ATM system by enabling industrial partnerships
(namely in FABs) and by reinforcing the role of the network manager. On
the institutional side, the Commission proposed building a more consist-
ent institutional framework by allocating a new role to EASA and focusing
Eurocontrol on the management and operations of the European ATM.
In short, while the EC has been quite successful at liberalizing air
transport, and somewhat successful at regulating access to airports, it is
currently in gridlock when it comes to realizing the Single European Sky,
which is a centerpiece in overall air transport liberalization and perfor-
mance. How can this gridlock be overcome to achieve the SES?
Indeed, and despite some improvements in overall performance (such as
delays and costs according to RP1 Monitoring, albeit at least in part due
to traffic downturn), the main bottleneck appears to be at the institutional
level, where progress appears increasingly unlikely. The main reason seems
to be that member states are not willing to give up national control over
their airspace and especially over their ANSPs. In other words, the SES
project has so far ended in (an institutional) gridlock. Consequently – and
in line with our conceptual framework – we turn to technology to explore
whether technological change does have the potential to overcome the SES
gridlock.
ASSESSING TECHNOLOGY’S POTENTIAL TO

OVERCOME THE SES GRIDLOCK
In this section we will identify the major technological developments cur-

rently under way in ATM. To recall, in light of the conceptual framework
as applied to ATM, technology can indeed become a driver for change in
European ATM. To tackle the problem of capacity, while increasing cost-
efficiency and performance, a series of potentially disruptive technological
innovations do offer some new opportunities.
In fact, disruptive technologies in European ATM are being
increasingly talked about by some ANSPs, Eurocontrol, and others.12
Although it may be premature to estimate which technologies will ulti-
mately disrupt European ATM (and prevail), this section highlights what
we believe are currently the three most prominent candidates, namely:
virtual air traffic control and cloud-based services; remote towers; and
centralized services.

M4371-FINGER_9781786431851_t.indd 351 14/11/2017 11:14

Virtual Air Traffic Control and Cloud-Based Services
Virtual air traffic control (VATC) and cloud-based services (CBS) are still
treated as two different technological developments. This may be mainly
because they are being developed by different technology providers, each
pushing for their own technological ends. In reality, VATC is a tool, spe-
cifically a tool for standardization, that constitutes the basis for services
offered via the cloud. Sooner or later, both may well merge into something
called a cloud-based ATM infrastructure.
More precisely, VATC is a method of providing air traffic control
services through common and standardized interfaces from a location-
independent virtualized environment. VATC is composed of two main
technologies: the common controller cockpit and a protected ATM infor-
mation cloud.13 Both enable functional and geographical decoupling of
ATM data service providers (ADSPs) from air traffic service units (ATSUs)
by introducing open and modular service architectures. As conceptualized
in the SESAR Master Plan, these services include surveillance, flight
data, trajectory, correlation and coordination.14 Currently, a consortium
of seven ANSPs and four industry partners are working on the maturity
and demonstration phases.15 Skyguide, the Swiss ANSP, has progressed to
realize VATC, in direct response to overcoming the SES gridlock, and is
now implementing it in between its two ATC units at Geneva and Zurich,
thus standardizing the technologies and operational procedures of the two
ATCs. Consequently, ATC can be provided independently of location and
the airspace user has only a single interface. In short, VATC is expected to
provide common operational requirements, cross-border service provision
and, of course, better performance.
CBS are air navigation services that use the principle of shared alloca-
tion of computing resources, such as processing power, storage and ser-
vices. This technology, originating from the IT industry, enables the service
provision of dynamic computing resources regardless of location. Future
ATM systems will indeed rely on integration and automation, using the
ATM information cloud as the backbone. Currently, large ANSPs are
migrating from legacy Flight Data Processing (FDP) systems to cloud-
based FDP services. These services also make standardization and con-
solidation processes possible among ANSPs (Kampichler and Eier, 2012).
A contemporary example of this technology is co-flight cloud services
(CCS).16 Co-flight e-FDP is a flight data processing system that enables
remotely accessible flight data.
VATC and CBS are both expected to bring benefits that are aligned
with SESAR’s objectives by integrating FDP systems among ANSPs and
strengthening the common vision of aircraft trajectories. Reducing the

M4371-FINGER_9781786431851_t.indd 352 14/11/2017 11:14

number of systems by way of consolidation will bring economies of scale

and increase performance. In effect, small ANSPs will be able to benefit
from modern high-performance architectures. This will, in turn, create
new business opportunities, whereby ANSPs will be able to consolidate
data processing and supply services between one another, using a common
information cloud. Eventually, competition for virtual service provision is
expected to bring benefits in terms of cost-effectiveness. Virtualization of
data provision will allow more efficient load sharing among ATC Units
and will enable more efficient resource allocation when traffic is low. It has
been estimated that, after a transition period of 15–20 years, VATC imple-
mented throughout Europe will bring annual savings of approximately
€1 billion.17 These savings are based on the partial closure of overnight/
off-peak centers, better organization of operations between ATC units,
and standardization. However, both VATC and CBS face challenges
related to sovereignty and security aspects of cross-border service provi-
sion. Also, the sharing of FDP data between multiple ANSPs will require
new regulations. Since the provision of cloud services can be supplied by
third-party manufacturers or ANSPs, consolidated databases make the
cloud structure an important target in terms of security. To comply with
the security requirements, centralized or system-based security operations
centers (SOC) and computer emergency response teams (CERT) will have
to be implemented.18
Remote Tower
Remote towers (RTWR) are a disruptive technology in which full video

coverage of an airport is displayed and managed remotely using high-
resolution panoramic displays. High-resolution video cameras are used to
capture real-time footage of an airport and this footage is sent in real-time
to a remote tower located somewhere other than the airport. Moreover,
the video is enhanced by augmented reality and infrared capabilities.
Typically, tower services are location-dependent and managed separately
for each airport. RTWR would also enable the relocation of tower services
for multiple airports into a central location.19 However, research has also
shown that air traffic controllers (ATCOs) sometimes confuse information
cues due to the multi-remote tower working position (Meyer and Hartmut,
2016). Therefore, extensive validation exercises and tests with ATCOs are
still required to show the benefits.
While the RTWR concept has been implemented at some small aero-
dromes, it is also expected to be implemented in international airports, such
as Shannon, Milan Linate and Budapest. This will bring cost-effectiveness
in infrastructure, deployment and maintenance for the ANSPs. To stand-

M4371-FINGER_9781786431851_t.indd 353 14/11/2017 11:14

ardize the technology, EASA and EUROCAE have been working on speci-
fications and regulation for RTWR.
Centralized Services
The EC’s quest for harmonization of air navigation services (ANSs),

dating back to 2012, created the concept of centralized services (CS).20 CS
are specific ANSs that are designed to be implemented at a Pan-European
and central network level. The main objective of CS, as promoted by
Eurocontrol, is to defragment and ideally centralize ANS services by
unbundling and creating a market for Pan-European service provision.
Initially, nine main services were proposed and further subdivided into
sub-services. These services include operational ANSs, such as flight plan,
trajectory, surveillance, performance monitoring, information manage-
ment, and communications. Rather than being implemented on a local
or national level, CS are designed to provide high-tech solutions to all
Eurocontrol Member States from a centralized system. Moreover, the
identification of CS relied on the fact that some ANSs would be performed
more efficiently at a centralized network level.
Following the disruptive business model of centralized services, ANSPs
that are accustomed to providing local solutions will adapt to compete for
the provision of Pan-European scale solutions. Cost–benefit analyses of
CS show that it is possible to save up to €1.6 billion over 10 years in the case
of centralizing ANSs.21 The savings are due to the unbundling of services
and the reduction in the number of fragmented solutions. Furthermore,
CS are also stated to be the key element for the member states’ meeting of
EC’s performance targets of ATM. CS are also designed to be compatible
with SESAR projects, such as the System Wide Information Management
System (SWIM). However, the implementation of CS requires cooperation
and management among the stakeholders, both at political and technical
levels.
Lately, the SESAR2020 Project No.15 called ‘Common Services’
(COSER) has focused on the harmonization of services such as delay
sharing, capacity balancing, trajectory, aeronautical data, and map services.
COSER are specific SESAR services that have the potential to be used by
several stakeholders and implemented at different types of network levels.
Moreover, COSER aim to provide a harmonized solution simultaneously at
the Pan-European, network, FAB, local, or other levels.22 Specifically, these
services will support ANS functionality and reliability in a harmonized
manner among stakeholders. SESAR2020 continues to develop common
services by examining the feasibility of new business models and their
impacts on the achievements of the current SES objectives.

M4371-FINGER_9781786431851_t.indd 354 14/11/2017 11:14

Disruptive business models might eventually emerge, as it has been

estimated that COSER can change the relationships between stakeholders
due to the configuration of new services. Harmonization typically brings
cost-effectiveness due to overall lifecycle costs, maintenance and training.
Eurocontrol, as the network manager, plays a key role in introducing and
validating these disruptive centralization solutions involving the different
European ATM stakeholders.
CONCLUSION: EUROPEAN ATM AS A

TECHNOLOGICAL LABORATORY?
Because the SES project has not delivered on its original promises,
European ATM is now passing through a challenging process of new tech-
nological developments, driven by a variety of actors. Advancements in the
three ATM technologies are opening up a new frontier for European and
probably also global ATM. All three of the technologies presented above
have the potential to defragment European ATM, but they will also lead to
consolidation of ANSs and ANSPs because of economies of scale (cost-
effectiveness) and broader performance improvements. Their adoption,
the pace of their implementation, and ultimately their disruptive nature
will depend largely upon current and future regulations, and especially
upon how the different stakeholders will react and behave.
The main actors (stakeholders) in European ATM all face these three
and even other different technological options and, consequently, face dis-
ruption. However, their reactions will vary depending on how they antici-
pate winning or losing in the process of such technology adoption. For
example, Eurocontrol is already trying to centralize air navigation support
services, whereas ANSPs are trying to consolidate some of these services
among themselves. Regulations play a key role in determining the outcome
of the European ATM laboratory. With proper regulations, the European
technological ATM experiment could be successful without consuming
too much time and too many resources. In this regard, in-depth analysis of
the consequences of such potentially disruptive technological innovations
should be carried out before moving on with separate experimentations by
the different actors. Moving towards a defragmented and technologically
intelligent Pan-European ATM is a gradual process, where the alignment
of interests of the actors with these technologies is required (Cook et al.,
2015).
Still, there is general agreement that, to obtain performance improve-
ments in European ATM, future ATM services should be more standard-
ized, more integrated, and probably more centralized, and that technology

M4371-FINGER_9781786431851_t.indd 355 14/11/2017 11:14

(one of the three technologies or a combination of them) will be key in

getting there. It is conceivable that if the current ATM stakeholders do not
adopt these technologies or move fast enough, ATM may well be disrupted
by outside actors. Companies such as Google, Microsoft, Amazon, tele
communications operators, as well as NASA23 are already experimenting
with autonomous solutions for unmanned air vehicles (such as drones).
Their solutions to standards and operational procedures may also have the
potential not only to transform ATM, but perhaps even to replace current
ATM. For example, Aireon24 is currently launching space-grade receiv-
ers on Iridium’s second-generation satellite constellation, called Iridium
NEXT, allowing for 100 percent global surveillance. Google’s Project
Loon25 is deploying an unmanned balloon network and Facebook’s Aquila
solar-powered unmanned flying wing is another example of tech giants
exploring26 connectivity platforms. In the future, these platforms could,
like Iridium-NEXT, also have the potential to provide ground-independent
ATM.
Contemporary technological developments such as those outlined in
the last section of this chapter all have the ability to disrupt or at least
transform current European (and global) ATM practices. The different
stakeholders will or will not adopt corresponding solutions, and experi-
mentation with these different technologies – if properly accompanied by
EU rules and regulations – may eventually lead to a way out of the SES
gridlock.
NOTES
1. The ‘Q’ code is a standard collection of three-letter codes that all start with the letter
‘Q’. It was initially developed for commercial radiotelegraph communication and later
adopted by other radio services, such as for maritime communications to promote radio
clarity, particularly for international speakers. The code set was adopted by a conven-
tion in London in 1912, in which codes in the range QAA–QNZ were reserved for
aeronautical use. Although their use is quite rare these days, a few terms persist, such as
‘QNH’ in aviation to promote unambiguous communications.
2. Not to be confused with Secondary Surveillance Radar (SSR), the first of which in
Europe was installed in France in 1962.
3. Automatic dependent surveillance–broadcast is a surveillance technology installed on
board equipped aircraft reporting in a ‘dependent’ (from the aircraft) way, different
parameters, like position and altitude. Compared to an independent ground surveil-
lance system, however, it is limited by its dependence on the aircraft’s technological
capabilities.
4. http://ec.europa.eu/transport/modes/air/single_european_sky/co-operation_icao_en.
5. This section is based on Baumgartner and Finger (2014).
6. EASA was created in 2002 and came into full functionality in 2008. The responsibili-
ties of EASA include to conduct, analyze, and research safety; provide authorization
for foreign operators; advise regarding the drafting of EU legislation; implement and

M4371-FINGER_9781786431851_t.indd 356 14/11/2017 11:14

monitor safety rules (including inspections in the member states); provide certification
of aircraft and components; and provide approval for organizations involved in the
design, manufacture and maintenance of aeronautical products.
7. PRB: Proposed regulatory approach for a revision of the SES Performance Scheme
addressing RP2 and beyond, Version 1, 1.3.2012.
8. Performance Review Commission and Eurocontrol: Evaluation of the Impact of the
Single European Sky Initiative on ATM Performance, 21 December 2006.
9. EC 390/2013 article 8.
10. https://ec.europa.eu/transport/modes/air/single_european_sky/ses-performance_en.
11. http://europa.eu/rapid/press-release_SPEECH-1 2–711_en.htm.
12. http://fsr.eui.eu/event/8th-florence-air-forum/.
13. Skyguide. The Virtual Centre Model. Skyguide, Geneva, 2014.
14. http://ec.europa.eu/transport/sites/transport/files/modes/air/sesar/doc/eu-atm-master-
plan-2015.pdf.
15. http://www.eurocontrol.int/news/demonstration-world-first-virtual-centre.
16. DSNA. DSNA Services Coflight as a Service. DSNA, 5 March 2015. Accessed 20
January 2017 at http://dsnaservices.com/portfolio/coflight-as-a-service/.
17. Skyguide. Blueprint: The 2017 Skyguide Technology Outlook. Skyguide, Geneva, 2017.
18. Directive (EU) 2016/1148 of the European Parliament and of the Council of 6 July 2016
concerning measures for a high common level of security of network and information
systems across the Union.
19. http://www.frequentis.com/fileadmin/content/Brochures/ATM/2016/RVT_whitepaper.
pdf.
20. Eurocontrol. Centralised Services CONOPS. Eurocontrol, Brussels, 2014.
21. Eurocontrol. The challenges and premise of implementing centralised services. Skyway,
Eurocontrol, Brussels, 2015.
22. http://ec.europa.eu/research/participants/portal/desktop/en/opportunities/h2020/topics/
sesar.ir-vld.wave 1–1 8–2015.html.
23. M. Baumgartner and A. Smoker, ‘Will unmanned vehicles modernise Air Traffic
Management?’, in Regulating Drones – Creating European Regulation that is Smart
and Proportionate, European Transport Regulation Observer 3/2015.
24. The era of space-based surveillance is here! Air Traffic Management, 14 January 2017.
25. ICAO Separation And Safety Panel (SASP) 1, 16 November, Information Paper 04.
26. Tech giants explore connectivity potential of unmanned platforms, IHS Jane’s Airport
Review, February 2017.
REFERENCES
ATM Policy Institute (2016). The case for liberalising air traffic control. Available at
http://atm01.drack-design.ch/wp-content/uploads/2016/12/The_ATM_Policy_
Institute_Report.pdf.
Baumgartner, M. (2007). The organisation and operation of European airspace. In
A. Cook (ed.), European Air Traffic Management. Farnham: Ashgate, pp. 1–34.
Baumgartner, M. and M. Finger (2014). The Single European Sky gridlock: a dif-
ficult 10-year reform process. Utilities Policy, 31, 289–301.
Cook, A., H.A.P. Blomb, F. Lillo, R.N. Mantegna, S. Miccichè, D. Rivas,
R. Vázquez, et al. (2015). Applying complexity science to air traffic management.
Crettenand, N. and M. Finger (2013). The coherence between institutions and
technologies in network industries. Competition and Regulation in Network
Industries, 14, 106–29.

M4371-FINGER_9781786431851_t.indd 357 14/11/2017 11:14

EC (1996). Freeing up European Airspace. White Book. EC COM 57/1996. Brussels:

European Commission.
EC (1999). The Creation of the Single European Sky. EC COM 614/1999. Brussels:
European Commission.
EC (2007). First Report on the Implementation of the Single Sky Legislation:
achievements and the way forward. Communication COM (2007) 845 final.
Brussels: European Commission.
Eurocontrol (2016). 2015 Comparison of Air Traffic Management Operational
Performance: U.S/Europe. Brussels: Eurocontrol.
Finger, M. and R. Künneke (2007). Technology matters: the cases of the liber-
alization of electricity and railways. Competition and Regulation in Network
Industries, 8(3), 301–33.
Finger, M., J. Groenewegen and R. Künneke (2005). The quest for coherence
between institutions and technologies in infrastructures. Journal of Network
Industries, 6, 227–60.
ICAO (2013). Global Air Navigation Plan 2013–2028. DOC 9750 – AN/963.
Montreal: ICAO.
Kampichler, W. and D. Eier (2012). Cloud based services in air traffic management.
Paper presented at Integrated Communications, Navigation and Surveillance
Conference, 24–26 April, Herndon.
Meyer, L. and F. Hartmut (2016). Investigating the safety-relevance of limited dis-
tinctive features on a multi remote tower-working position. SESAR Innovation
Days 2016, 9 November, Delft.
Schubert, F. (2011). The liability of air navigation services providers: some lessons
from the single European sky. In P. Mendes De Leon and D. Calleja Crespo
(eds), Achieving the Single European Sky: Goals and Challenges. Amsterdam:
Kluwer Law International.
Sudarshan, H.V. (2003). Seamless Sky. Aldershot: Ashgate.

M4371-FINGER_9781786431851_t.indd 358 14/11/2017 11:14

18. Canada and USA: a tale of two
ANSPs
Rui Neiva
INTRODUCTION
Since New Zealand first started experimenting with commercialization in

1987, the air navigation services (ANSs) industry around the world has
evolved tremendously. From being traditionally provided by the govern-
ment or the military, air navigation service providers (ANSPs) now come
in a variety of institutional arrangements, including government corpora-
tions, non-profit private entities and public–private partnerships (PPP).
This chapter will explore those changes and their impacts, with a focus on
two case studies in North America: the US and Canada. These are two
interesting contrasting ANSPs: one that underwent institutional reform
(Canada) and one that did not (US).
LIBERALIZATION INITIATIVES
Since New Zealand created, in 1987, a for-profit government corporation

(Airways New Zealand) to become its ANSP, many other countries have
followed a similar path towards more independence in governance and
funding. This section will provide an overview of issues regarding liberali-
zation and an analysis of its impacts.
Overview
Commercialization of ANSPs – that is, governance and funding reform of

a governmental organization to function like a commercial business – has
followed the liberalization of the airline and airport industries. ANSPs
have traditionally been operated by national governments under the aus-
pices of the Chicago Convention. As mentioned, that began to change in
New Zealand in 1987. Since then more than 50 countries have experienced
some sort of institutional reform. While most commercialized ANSPs
359

M4371-FINGER_9781786431851_t.indd 359 14/11/2017 11:14

became government corporations, two exceptions exist: Canada, where

NAV CANADA is a private non-profit user cooperative, and the UK’s
NATS, a PPP.
Reasons for commercialization varied. One was ideological, as some
political quadrants might see ANSs as a function that should not be in
direct control of a government. On the operational side, commercializa-
tion was seen as needed to liberate ANSPs from government bureaucracies
that were hindering the ANSP’s ability to invest in the system and serve
its customers instead of its political masters. Also, budget restrictions and
competition for public money between governmental agencies were stop-
ping capital projects seen as needed by the aviation industry. Finally, by
moving ANSs into a separate entity, governments were able to remove the
ANSP budget from the national accounts, with implications for the public
debt and governmental deficits.1
Rationales against commercialization have come from a few different
areas. A common argument is that safety would be jeopardized when
ANSs were not provided by a governmental agency. Additionally, it has
also been argued that the provision of ANSs is an inherently governmental
function, be it because of the military importance of airspace or to guar-
antee equal access to all airspace users.
New Zealand
New Zealand has become a well-studied case on the field of ANS liberali-
zation, as it was the first country to do so in 1987 with the creation of a
for-profit government corporation, Airways New Zealand. It was also the
first country to separate ANS operations from safety oversight, a principle
that both the ICAO and the EU now recommend. While a small system
in terms of traffic (around one million flights per year), it has a very large
continental and oceanic airspace of around 30 million km2.
The commercialization of ANSs in New Zealand came at a time when
the country was going through macroeconomic difficulties. The govern-
ment at the time decided to go through a number of reforms, which
included the creation of Airways New Zealand. The company is owned
by the Minister of Finance and the Minister of State-Owned Enterprises,
while the Minister of Transport performs safety oversight duties.
The company is a for-profit commercial enterprise, paying dividends
to the national government as a result. Its funding comes from user fees
charged to airspace users (including some charges for general aviation),
and it has access to private capital markets for financing needs. Economic
regulation is performed by the Commerce Commission of New Zealand.
In the almost three decades since it has been created, Airways New Zealand

M4371-FINGER_9781786431851_t.indd 360 14/11/2017 11:14

Canada and USA: a tale of two ANSPs 361
has always had an operating profit (NZD$15.1 million in 2015, with rev-
enues of NZD$186.3 million), and has paid around NZD$40 million
(nominal values) in dividends to the New Zealand government since 1987,
including NZD$4 million in 2015.
Australia
Airservices Australia, the Australian ANSP, was created in 1995 as a for-

profit government corporation, much like Airways New Zealand. The
move to commercialize ANSs came shortly after airlines were deregulated
(1990) and airports began to be privatized (1993), resulting in a significant
overhaul of the aviation industry in the country.
The system is governed by a board composed of members appointed
by the Minister of Transport, while the Civil Aviation Authority provides
safety oversight. Financing comes from user fees on all airspace users,
including general aviation. Economic regulation is done by the Minister
of Transport as well as the Australian Competition and Consumer
Commission.
The system faced a serious financial situation after it was created, and
had to undergo an extensive cost-cutting programme that led to 2001 being
the first year where it posted a profit. Airservices had revenues of AUD$1
billion in 2015, and reported a profit of AUD$5.9 million before tax. It
also paid a dividend of AUD$11 million to the government during the
exercise.
In 2016, major stakeholders in the Australian aviation community,
including general aviation, called for the system to be privatized into a non-
profit organization similar to NAV CANADA (Australian Aviation, 2016).
Rest of the World
While widespread in most of Europe (discussed in Chapter 17), Oceania

(discussed in this section) and Canada (discussed in the next section),
ANSP liberalization has yet to take hold in other areas of the world. While
in countries like Thailand and South Africa some form of commercializa-
tion has been present with the creation of government corporations, in
countries like Brazil the system is still run by the military, and in others like
China the military still impose heavy restrictions on the use of airspace,
limiting the efficiency of the system.
In sub-Saharan Africa, with the exception of South Africa, the lack of
navigation equipment still greatly limits the availability of ANSs in the
region. For those, space-based satellite navigation2 offers great potential,
as it will allow ANSPs in the region to allow full territorial coverage,

M4371-FINGER_9781786431851_t.indd 361 14/11/2017 11:14

without having to install expensive ground infrastructure. For example, the

Southern African Development Community (SADC), a regional organiza-
tion of 15 member states, is expected to start using space-based satellite
navigation in 2018 (Aireon, 2016).
However, there has been a trend to separate provision from safety over-
sight, which the ICAO has been calling for.3 Outside of Europe (where
this separation is now in European law), countries like Egypt, Argentina,
Mexico, Peru, South Africa, Nigeria and Thailand have all separated those
functions. Others, like Ghana, have also shown interest in pursuing func-
tional separation.
Measuring Liberalization Outcomes
While there are a few studies that deal with the issue of ANSP benchmark-
ing,4 the literature dealing specifically with the effects of ANSs’ liberaliza-
tion is somewhat scarcer. Additionally, most of the works in this field are
case study approaches, with little economic analysis. With those caveats
present, this section will provide an overview of the works in the area.
The case study approaches include, among others, Majumdar (1995),
Dempsey et al. (2005), United States Government Accountability Office
(2005), mbs Ottawa Inc. (2006), Oster and Strong (2007) and Neiva
(2015).5 The first of these works, Majumdar, was the most sceptical about
commercialization. The study noted that the pressure to reduce costs
would inevitably lead to safety issues, which was a common theme at the
time. As the later works reported, time would assuage this fear, and com-
mercialized ANSPs have followed the overall trend of the industry with
ANSs becoming safer regardless of governance status. Overall, these works
concluded that institutional changes were mostly positive, and at worst
neutral. Positive outcomes include improvements in customer service,
financial stability with great control of increasing costs and more capabil-
ity to finance and implement modernization projects. However, it has been
argued that some of these changes might not result directly from all aspects
of commercialization, but simply because of greater managerial and finan-
cial independence, which might happen without commercialization. Still,
taking political realities into consideration, commercialization appears to
be the better way to achieve such independence.
On the works that provide an econometric approach to the problem,
Button and Neiva (2014a, 2014b) and Neiva (2015) performed several ana
lyses of the European system using Data Envelopment Analysis (DEA) and
Stochastic Frontier Analysis (SFA). The results showed that, overall, more
governmental control was associated with better economic efficiency. The
authors offered three explanations for the results. One possibility is that

M4371-FINGER_9781786431851_t.indd 362 14/11/2017 11:14

since those commercialized entities are still subjected to full cost recovery
every year, they are subjected to rate-of-return regulation6 (with a return of
zero in this case, as they have full cost recovery requirements). This form of
regulation has been associated with the Averch–Johnson effect (Averch and
Johnson, 1962), which leads to lower efficiency through suboptimal high
capital to labour ratios. Another explanation considered by the authors
was that the ANSPs that were first commercialized were the ones under-
performing, and were still ‘catching up’ to their peers in terms of efficiency.
Additionally, Neiva (2015) considered that perhaps some of the corpora-
tized ANSPs operated de facto like governmental agencies, clouding the
distinction between commercialized and non-commercialized entities.
Conclusion
Liberalization initiatives in the ANS industry have generally been regarded

as positive, from both a safety and an operational standpoint. In terms
of safety, initial fears that commercialized ANSPs would have detrimen-
tal impacts on safety were unsubstantiated. The industry as a whole has
achieved a remarkable safety record, regardless of ownership. From an
operational perspective, the results are also positive, with commercialized
agencies reporting that they are able to operate more freely, focusing on
their customers instead of political masters, and prioritizing investments
based on safety, efficiency and return-on-investment.
In a time when ANSPs are faced with significant challenges to adapt to
a new technological environment, the financial and managerial independ-
ence that liberalization brought will make it easier, or at least less difficult,
for ANSPs to succeed.
CANADA: NAV CANADA
The Canadian ANSP, NAV CANADA, is a private non-share capital

corporation. Operating since 1996, it controls the second busiest airspace
(with 3.4 million flight hours controlled in 2014) after the US, and has been
touted as one of the best examples of the potentialities of commercializa-
tion. This section will explore the history behind the decision to privatize
and how the system is performing since it became a private corporation.
Privatization
Discussion on the privatization of Canadian ANSs came in the early 1990s.

At the time, ANSs were provided by Transport Canada, a governmental

M4371-FINGER_9781786431851_t.indd 363 14/11/2017 11:14

branch, and funded by taxation. Back then, many in the industry, includ-
ing airlines and labour unions, were displeased with the state of affairs; the
airlines felt that the system was not keeping pace with the growth in traffic
and labour unions were unhappy with salary freezes lasting for several
years.
After a consultation process that discussed several options (including
full privatization and contracting out), it was decided to create a non-profit
private entity to run the system. That company would be totally independ-
ent from the Canadian government, having full control of its manage-
ment and budget, including the ability to charge user fees and issue debt.
Governance would be provided by a number of aviation stakeholders,
including airlines, general aviation7, labour unions, the government and
independent members.
Following this consultation process NAV CANADA was created in
1996, with 6300 federal employees being transferred to the new company.
The government received a payment of CAD$1.5 billion in the privatiza-
tion process. User fees were established after 18 months, and since 1999
they have to provide full cost recovery to the company, a principle that still
exists today. To deal with downturns in traffic, NAV CANADA has a rate
stabilization fund that is replenished during periods of growth in traffic
and is used during downturns. This allows the ANSP to avoid a ‘death
spiral’ in which less traffic leads to higher fees, leading to less traffic. After
the September 11 attacks in the US in 2001, NAV CANADA used all of
the money available in the fund and user fees had to be raised six times over
a period of three years. In the aftermath of the 2008 financial crisis the
fund was used but never depleted, allowing user charges to be maintained
at the same levels. Since 2012 the fund has been maintained at 7.5 per cent
of revenues.
Given its board structure, it makes the company in essence a non-profit
user cooperative, a relatively common structure in North America for utili-
ties. The board structure, with several stakeholders in the governing board,
but none with a majority, also reduces the need for economic regulation.
With the mandate to fully recover costs every year, and with those stake-
holders governing, there is an incentive to keep costs and user fees in check.
Still, every change in user fees requires an extensive consultation process,
and there is an ex post right to appeal user fee changes to the government.
After its creation NAV CANADA had a few tumultuous years, with one-
sixth of the workforce being laid off, most of them being middle managers.
An extensive restructuring and consolidation plan was also put in place, as
well as a capital expenditure programme. Ultimately, that initial transition
period had the help of an expanding economy and aviation market, which
allowed NAV CANADA to fund these restructuring programmes and

M4371-FINGER_9781786431851_t.indd 364 14/11/2017 11:14

please its labour force with more attractive compensation packages. When
the first real test to the system came in 2001 with the September 11 attacks
in the US, the company was much better prepared to withstand the shock.
NAV CANADA at 20 Years Old
Having just celebrated 20 years in 2016, it is now possible to assess how

NAV CANADA has fared. And, by many measures, the company is doing
very well for itself, its 4600 employees (including 1900 air traffic control-
lers) and for the Canadian airspace.
In 2015 the company had CAD$1.3 billion in revenues, and operating
expenses of around CAD$1.1 billion. Depreciation and amortization
expenses accounted for around CAD$200 million. By the end of the fiscal
year, the rate stabilization account had CAD$98 million. The company’s
debt of around CAD$1.7 billion had its credit rating upgraded by Moody’s
in 2016 to Aa2.
In 2016 NAV CANADA announced that user fees would be reduced by
an average of 7.6 per cent in 2017. This reduction will result in savings of
around CAD$100 million for the airlines in 2017 (Air Traffic Management,
2016). Additionally, such a reduction will mean that in 2017 user fees will
only be 1.5 per cent higher in nominal terms than in 1999, the first year in
which full cost recovery was enacted. Between 1999 and 2016, inflation in
Canada accrued to around 38 per cent.
Operationally, the system has been steadily increasing its safety statistics.
The cases of losses of separation reached 0.75 per 100 000 aircraft move-
ments, roughly 30 per cent less than in 2001.8 In terms of technology, it
has also been at the forefront across the industry. It now offers electronic
communication via text message throughout the country;9 since 2009,
electronic paper strips have been implemented in all towers and centres;
ADS-B Out is available over 4 million km2 in the Hudson Bay and the
Northeast region of the country.
NAV CANADA also has a commercial arm, NAVCANatm, which
develops and sells many technologies, like electronic flight strips, to other
ANSPs. NAV CANADA is also a majority holder – with an investment
of US$150 million – in US-based Aireon LLC, a company that aims to
deliver space-based satellite navigation by 2018 using a constellation of
66 low-orbiting satellites.10 The use of space-based satellite navigation
will allow unprecedented coverage over remote regions, such as oceans
and the poles, where radar-based navigation was never a possibility and
navigation relies on approximate self-reported positioning.11 It will also
make it much more cost-effective to offer more precise navigation in areas
like Africa, which never had the possibility to install the expensive ground

M4371-FINGER_9781786431851_t.indd 365 14/11/2017 11:14

infrastructure necessary to offer radar coverage throughout extensive

swaths of territory.
Conclusion
NAV CANADA has been the poster child of commercialization – and

with good reason. Before privatization the system had to be subsidized
by the government to the tune of 25 cents for every dollar spent every
year. Capacity was not keeping up with the growth in traffic. Labour
was unhappy. Twenty years later, safety has improved significantly, the
system is in a strong financial position, user fees are practically at the
same nominal level as they were when first introduced in 1999, labour is
strongly involved in the success of the company, and some of the more
modern ANS technologies are not only implemented but also developed by
NAV CANADA. It is not easy to find many drawbacks in the system that
Canada has put in place.
UNITED STATES: FEDERAL AVIATION

ADMINISTRATION
In the US, ANSs are provided by a federal agency, the Federal Aviation
Administration (FAA). This agency controls the biggest airspace in the
world, with 10.4 million km2 of continental airspace and 65 million km2
of oceanic airspace. Besides being the biggest in area, US airspace is also
the biggest in terms of traffic, with 23.1 million flight hours controlled
in 2015. The agency employs around 46 000 people, from which less than
11 000 are fully certified air traffic controllers.12 In terms of governance
and funding, the system is a holdout among developed nations, as it is still
directly operated by the national government and is funded by taxation,
not user fees. This section will present an overview of the current system,
along with a discussion of the most important attempts that have been
made at reforming it.
History, Governance and Funding
The involvement of the US federal government in ANS provision dates

from the 1930s.13 By the beginning of the following decade, the federal
government, via the Civil Aeronautics Administration (CAA), would be
responsible for providing both en-route and terminal services.
The system that exists today started to be constructed in the 1950s, first,
with the introduction of radar after a mid-air collision over the Grand

M4371-FINGER_9781786431851_t.indd 366 14/11/2017 11:14

Canyon in 1956, then, in 1958, with the creation of a standalone agency

(the CAA was part of the Department of Commerce) to provide ANSs
and safety oversight: the Federal Aviation Agency. The standalone agency
was to last less than a decade, and was merged in 1967 into the newly
created Department of Transportation. The Federal Aviation Agency then
became the Federal Aviation Administration that we have today. Besides
providing ANSs, the FAA is responsible for safety oversight of the entire
aviation industry (including its own ANS operations), and also gives
out grants for airport capital investments as well as research initiatives.
Additionally, the FAA provides oversight and promotes the commercial
space transportation industry.14
The current funding system was established in the early 1970s. Until
then, funding for the FAA and its predecessors came from funds appropri-
ated directly by Congress. As such, funding was unstable and unpredict-
able, as spending priorities shifted. To try to ameliorate these issues, the
Airport and Airway Trust Fund (AATF) was created in 1970, with dedi-
cated fuel and passenger ticket taxes. While the initial goal was to fund the
entirety of the FAA with the AATF, that vision never came to be. Since
1990 the AATF only covers, on average, around 75 per cent of the agency’s
budget. As such, the FAA remains to this day subjected to the whims of
the legislature and, as it will be seen in a later section, this budget instabil-
ity has been used by critics to promote the commercialization of ANSs.
By 2017, taxes for the AATF came from a variety of sources:
●● tax on domestic passenger tickets: 7.5 per cent of ticket price;

●● tax on domestic flight segments (excluding flights to or from rural
airports): $4.10 per passenger per segment (indexed to inflation);
●● tax on international arrivals and departures: $17.50 (indexed to
inflation);
●● tax on flights between the continental United States and Alaska
or Hawaii (or between Alaska and Hawaii): $8.70 per passenger
(indexed to inflation);
●● tax on mileage awards: 7.5 per cent of value of miles;
●● tax on domestic cargo or mail: 6.25 per cent of amount paid for the
transportation of domestic cargo or mail;
●● aviation fuel for general aviation: aviation gasoline, $0.193/gallon; jet
fuel, $0.218/gallon; $0.141/gallon surcharge on fuel used in aircraft
that is part of a fractional ownership programme;
●● aviation fuel for commercial use: $0.043/gallon.
In fiscal year 2015, these taxes collected around $14.3 billion, rep-
resenting 92 per cent of FAA’s budget for that year (Federal Aviation

M4371-FINGER_9781786431851_t.indd 367 14/11/2017 11:14

$12.00
$10.00
$8.00
Billions
$6.00
$4.00
$2.00
$–
1990 1995 2000 2005 2010 2015
Source: US Department of Transportation (2016 and other years).
Figure 18.1 FAA operations budget (1990–2016) in billions of

2016-adjusted dollars
Administration, 2016a). Most of FAA’s budget is spent on ANSs; from

that spending, most goes to the operation of the system, with a similar
share going to capital expenditures.15 Figure 18.1 shows the budget of
FAA’s operations over time. Since 2000, ANS operational expenses have
grown 19 per cent in real terms (41 per cent since 1990). This is despite
a decline in total (commercial and general aviation) traffic of 25 per cent
between 2001 and 2015 at airports, and a reduction of 17 per cent in en-
route movements (Federal Aviation Administration, 2016b).
In terms of cost efficiency, and when compared to Canada, unpublished
results by economic consultants GRA, Inc. showed that if the user fees
that are charged in Canada were applied at the same exact levels to 2013
US traffic, they would only generate around two-thirds of the revenue
required to run the US system.16 This strongly suggests that the system in
Canada is much more cost-effective than the one in the US.17
As mentioned, funding from the agency comes from the AATF and
regular general funds. Funding is normally established on a multi-year
basis, with a ‘FAA reauthorization’. But when one of those reauthoriza-
tions is set to expire, Congress can enact short-term extensions, compli-
cating things further. For example, a 2008 bill set to expire by 2011 was
extended 23 times, until a new bill was passed. Every time those extensions
were nearing the end, the agency needed to prepare for an eventual lapse
in funding, which delays investments, and makes them more expensive.
Finally, in 2012 a new multi-year bill was passed. That bill expired in 2015,
and by early 2017 three extensions had already been enacted by Congress.

M4371-FINGER_9781786431851_t.indd 368 14/11/2017 11:14

NextGen
The FAA has had quite a few modernization programmes throughout

the decades. The current iteration is called the Next Generation Air
Transportation System, or NextGen. This programme was launched in
2003 and now encompasses a series of different projects, some of which
were already in progress and are now included under the NextGen banner.
Perhaps the main component of NextGen is the replacement of radar-
based with satellite-based navigation, with ADS-B Out. The system
consists of 634 ground stations, which have been operational since 2014.
However, not all FAA’s facilities are able to make use of these new capabili-
ties. From those that are, government overseers18 have noted that control-
lers have not had the training to use new satellite-based procedures, and
are still relying on old radar-based procedures (Office of the Inspector
General, 2014).
Other initiatives in the NextGen programme include:
●● Digital data communications between controllers and pilots; already

in use at airport towers, and to be deployed in en-route procedures
by 2019.
●● En Route Automation Modernization (ERAM), a new en-route
computer system, replacing the 1970s Host system. The replacement
for Host first started being discussed in the early 1980s. ERAM
finally became operational in 2015 with a $370 million budget
overrun. The air traffic controllers union has criticized the system
as being 1990s technology. For terminal control, the Terminal
Automation Modernization and Replacement (TAMR) system will
be deployed by 2019.
●● The System Wide Information Management (SWIM) system to
gather and distribute real-time information about the ANS system
(weather information, airspace restriction, airport runway configu-
ration, etc.) throughout the entire community, including eventually
aircraft in flight.
●● Electronic flight strips, with the Terminal Flight Data Manager
(TDFM) system to be deployed by 2028 across 89 FAA towers.
NextGen implementation has been problematic. Government bureau-

cracies have never been particularly good in implementing large capital
projects, such as the case of NextGen (initially it was a $20 billion invest-
ment over 20 years). In the case of NextGen, this is exacerbated by the
peculiarities of the FAA budget as well as by the procurement rules the
agency is subjected to. Realizing that the initial timeline was not feasible, in

M4371-FINGER_9781786431851_t.indd 369 14/11/2017 11:14

2015 the FAA scaled back its intentions significantly, focusing the agency’s
resources on some key projects that had the most potential to deliver ben-
efits in the short term. Adding to those implementation issues, many air-
space users have been delaying equipping their airplanes with the necessary
technologies for the FAA to take advantage of the NextGen technologies
already in place. Many of the benefits touted by the FAA will only materi-
alize once a critical mass of airspace users is equipped with those technolo-
gies. But until that critical mass comes, it will be hard to prove to the users
that equipage is advantageous.19
Attempts at Reform
There have been a multitude of attempts to reform the governance and

funding of ANS provision in the US. These varied from discussions
to privatize ANS provision in the late 1980s to the creation of a non-
profit organization à la NAV CANADA in 2016 and 2017. This section
will explore three of the more relevant reform attempts: the proposal to
create a government corporation in 1994, the internal reorganization of
the early 2000s; and the proposal to create a non-profit organization in
2016–2017.20
The first reform proposal to gather significant steam in the US was
the 1994 proposal to create a government corporation to operate ANSs,
the United States Air Traffic Services Organization (USATS). Part of a
broader initiative to ‘reinvent’ government,21 USATS would be a non-
profit federal government corporation. The corporation would be gov-
erned by a board of directors, with representatives from stakeholders such
as airlines and labour. For funding and finance, USATS would charge user
fees and issue debt.
While USATS had the support of the executive branch and the air
traffic controllers union, general aviation and some airlines opposed the
proposal. Most of Congress was opposed as well, because they feared they
would lose control over the system. As such, the proposal never came to
be. Given the experience with government corporations in the US, where
Congress has been known to meddle with the internal affairs of these cor-
porations,22 it is likely that the goals in the USATS proposal would not be
achievable, and USATS would function much like the FAA, with a high
degree of control from Congress in their operations and business decisions.
Following the USATS proposal, in 1996 Congress gave the FAA the
ability to establish its own procurement and personnel rules. The aim of
this reform was to give the agency more flexibility in hiring and keeping
personnel and to become more efficient and cost effective in its capital
investments. A few years later, government overseers noted that the reform

M4371-FINGER_9781786431851_t.indd 370 14/11/2017 11:14

had failed because the FAA created rules for itself that mimicked the rest
of the government, ending up in a situation similar to the starting point
(United States Government Accountability Office, 2003).
In 2000 came the most relevant reform to be enacted: the internal
spin-off of the ANS functions within the FAA into a new department,
the Air Traffic Organization (ATO). This reform followed a 1997 report
from the so-called Mineta Commission that recommended the creation
of a performance-based organization within the FAA to provide ANSs
(National Civil Aviation Review Commission, 1997).23 However, the
organization envisioned by the Mineta Commission never came to be. The
report recommended the creation of an entity separated from the federal
budget process, with access to capital markets and financed by user fees.
None of this was enacted. Instead, the ATO was created as a department
of the FAA, but remained part of the federal budget process and funded
by taxes on the aviation system as well as general funds, and ultimately very
little changed in ANS provision.
After the creation of the ATO, the focus has been on the implementation
of NextGen. But that changed in 2013 when political rows over the federal
budget led to a ‘sequester’; that is, a mandatory budget cut throughout the
government, and a government shutdown, in which non-essential func-
tions of government are stopped. These events had significant impacts
for the FAA, as it was forced to stop training new air traffic controllers.
Air traffic controllers were furloughed, leading to traffic delays, and many
capital projects were put on hold. Because of this, since 2013 an interest for
institutional reform resurged in the US, with many in the industry – includ-
ing the majority of the airlines and the air traffic controllers union – openly
supporting reform. Those efforts culminated in early 2016 with a proposal
from the House of Representatives, the ‘Aviation, Innovation, Reform and
Reauthorization Act’ (AIRR), to spin-off ANSs into a non-profit entity.
AIRR would create a federally chartered non-profit corporation, the
ATC Corporation, modelled after NAV CANADA and to start opera-
tions on October 2019. It would be funded by user fees on commercial
airlines (general and business aviation would be completely excluded from
payment) and totally independent of the government. Its governance
would be assured by a board of directors with members from airlines, the
government, labour, general aviation, business aviation and manufactures.
In contrast to what happened in Canada, there would be no payment for
the transference of assets to the new corporation.24
Despite the strong support of the air traffic controllers union and the
majority of the airlines,25 the bill failed in Congress. (In 2017, a similar
bill was once again introduced in the House of Representatives.) As with
other previous efforts, the proposal failed to gain support from the general

M4371-FINGER_9781786431851_t.indd 371 14/11/2017 11:14

aviation community. Given that the bill pre-empted their fear of having to
pay user fees by excluding them from that payment, their opposition was
based on concerns about access to airspace, and that a private provider
would eventually impose restrictions for general aviation operators to fly
as they pleased. Together with reluctance from some members of Congress
in giving up control of the system, the bill suffered the same fate of the
USATS proposal 22 years earlier.
Conclusion
Since the AATF was created in 1970, FAA governance and funding have
remained essentially the same. Despite a myriad of research produced in
the last four decades pointing to the problems with ANS provision in the
US, the system is still provided by a governmental agency funded by taxa-
tion on the aviation system and general funds, appropriated on a short-
term basis. The reforms that have been implemented (procurement reform
in 1996, ATO in 2000) ended up being mostly cosmetic, without significant
improvements.
Today, the focus of the agency is on the deployment of NextGen.
While there have been some successes in its implementation, namely since
the focus was shifted in 2015 to the projects with the most benefits in
the shorter term, serious challenges remain. Overall, the programme is
still underfunded. Even if everything went smoothly at the FAA’s end,
Congressional fights over budgets mean that every so often the agency is
faced with externally imposed delays on the implementation of NextGen.
Every time that happens, a few million dollars are added to the costs of
the ongoing projects, and future projects get delayed. While this status quo
remains, it is hard to envision a future where NextGen delivers at the level
that allows the system to accommodate the expected growth in traffic.
CONCLUSION
In a naturally slow-moving industry, ANSP liberalization has been taking

its time to percolate throughout the industry. Still, it has now been three
decades since the first liberalization initiatives were implemented. And
the experience in those three decades shows that liberalized systems, with
greater control over their management and finances, are able to focus their
efforts on efficiency and their customers.26 Meanwhile, governmental agen-
cies have many more interests to please and, in doing so, are hindered in
their ability to focus.
The comparisons between the two North American systems show the

M4371-FINGER_9781786431851_t.indd 372 14/11/2017 11:14

drawbacks and benefits on liberalized vs. non-liberalized ANSPs. While

one system (Canada) is at the forefront of technological innovation, the
other (US) is unable to deploy technology at an efficient pace and in a
cost-effective manner.27 This seems mostly due to self-sustaining funding
and financing mechanisms that exist in Canada, together with the system’s
independent management and a governance regime constituted by a set of
stakeholders with a stake in creating an efficient system.
One example of how this independence works is space-based satellite
navigation. While NAV CANADA has been at the forefront of the tech-
nology, with its investments in Aireon, the FAA, despite controlling a sig-
nificant portion of the world’s oceanic airspace in the Pacific and having
greatly to gain from its widespread use, has not indicated its willingness
to adopt the system.28 In mid-2016 the US Senate introduced provisions
in a bill that would require the FAA to adopt a form of space-based satel-
lite navigation by 2018, but the bill was not passed. If the FAA remains
in charge of ANSs in the US, it will probably take such a mandate from
Congress – with accompanying funding – for the agency to implement
space-based satellite navigation.
While liberalization might not be the solution to all the challenges that
ANSPs face in the upcoming years, there does not seem to be much doubt
that ANSPs with greater levels of independence appear to be in a better
position to face those challenges.
NOTES
1. In the case of Canada and the UK, commercialization also resulted in a payment to the
government when the respective commercialized ANSPs were created.
2. Space-based satellite navigation will be discussed in greater depth in the section con-
cerning Canada.
3. For a study on the separation of provision from safety oversight and its outcomes, see
Brown et al. (2014).
4. See, for example, the various works produced by the Eurocontrol Performance Review
Commission (available at http://www.eurocontrol.int/prb/publications) or the Civil Air
Navigation Services Organisation (available at https://www.canso.org/publications).
Additional works include NERA Economic Consulting (2005); Helios Economics and
Policy Services (2006); Button and Neiva (2013); Arnaldo et al. (2014) and Standfuß et
al. (2015).
5. For a more complete bibliography on the issue, see Neiva (2015).
6. For more on rate-of-return regulation, as well as other forms of regulation, see, for
example, Train (1991).
7. General aviation agreement with privatization came with a cost for the company:
general aviation pilots only have to pay between CAD$67.64 and CAD$225.84 annu-
ally for unlimited en-route and terminal access (some larger airports impose a $9.95
surcharge).
8. For an analysis of the relevance of the use of this metric, see Button and Drexler (2006).
9. However, NAV CANADA still reports low usage of this technology. Still, from fewer

M4371-FINGER_9781786431851_t.indd 373 14/11/2017 11:14

than 50 000 messages exchanged in the first quarter of fiscal year 2013, more than
300 000 messages were exchanged in the fourth quarter of fiscal year 2015.
10. Inmarsat is another company intending to offer space-based satellite navigation
commercially.
11. For example, Aireon estimates that on the North Atlantic routes the use of space-based
satellite navigation will allow lateral separation to be cut by half to 15 nautical miles.
12. This is the lowest number in over three decades, and is creating serious concern among
aviation stakeholders, who fear that the agency will be unable to increase those numbers
(they have failed their hiring targets multiple times and around 25 per cent of controllers
are already eligible for retirement and have mandatory retirement by age 56), leading
to delays in the future because of a controller shortage. In many installations across
the US, controllers already have mandatory six-day workweeks because there are not
enough controllers. In some instances, as with New York Terminal Radar Approach
Control (TRACON), those mandatory six-day workweeks have been in place since the
late 1990s.
13. The first efforts of what are now known as ANSs were by a group of airlines that came
together to form a non-profit user cooperative, the Aeronautical Radio, Incorporated
(ARINC), a company that still exists today. It started by providing air/ground commu-
nications, and eventually evolved into providing terminal control and en-route separa-
tion before the federal government took over that role. For more on the history of ANSs
in the US, see Nolan (2010) and Neiva (2015).
14. Until the mid-1990s the agency was also responsible for the promotion of the airline
industry, but by seeing the conflict of interest of having an agency regulating the safety
of an industry and having to promote it, Congress would eventually remove that obliga-
tion from the FAA charter.
15. The FAA, like other agencies in the US federal government, does not have a capital
budget per se; it simply has an account where expenses related to maintaining and
buying equipment on a given year are accounted for.
16. Of costs totalling USD$12.9 billion, fees in the US at the same level as NAV CANADA’s
fees would only raise USD$8.8 billion.
17. While it might be argued that the overall complexity of the US system is higher, par-
ticularly in some high traffic areas like New York City or Chicago, the system is also
much bigger, with seven times the traffic, thus allowing for bigger economies of scale.
Additionally, the Civil Air Navigation Services Organisation, in their benchmarking
report for 2014, also reports a much higher cost per continental flight hour controlled at
the FAA compared to NAV CANADA: USD$465 for the former and USD$321 for the
latter (Civil Air Navigation Services Organisation, 2015).
18. For an extensive list of reports by government overseers on the implementation of
NextGen, see Neiva (2015).
19. For general aviation users, the FAA created a rebate programme where the agency pays
up to $500 of the price of equipping an aircraft with ADS-B Out capabilities.
20. For a more complete discussion of US attempts at governance reform, see Neiva (2015).
21. This initiative, the National Performance Review, was spearheaded by US Vice
President Al Gore. Its goal was to streamline governmental functions and it would make
recommendations in many areas, from public housing to defence and use of informa-
tion technologies throughout the government. For more information on the issue, see
National Partnership for Reinventing Government (2001).
22. Two good examples of this congressional meddling are Amtrak, the medium- and long-
distance rail operator, and the Postal Service. The former has been known to try to end
service to underperforming locations, and the latter has tried to end the delivery of
regular mail on Saturdays. Both initiatives have been stopped by Congress multiple times.
23. The Mineta Commission was very critical about FAA’s operation and congressional
oversight. Among other things, it noted that the federal budget rules were not appropri-
ate for the provision of ANSs and there were ‘too many cooks’ trying to influence the
FAA in the different layers of government (from Congress to the government overseers).

M4371-FINGER_9781786431851_t.indd 374 14/11/2017 11:14

24. This was justified on the basis that those assets were mostly paid by the users of the
system via taxation, so if the ATC Corporation had to buy those assets from the govern-
ment, users would be paying twice.
25. The holdout was Delta Airlines, which opposed the intent to privatize the system. Delta
would eventually quit Airlines for America, the main trade association for the airline
industry, because of the association’s support of privatization.
26. Naturally, the number one focus on any ANSP, liberalized or not, is and should always
be safety.
27. This is despite FAA’s budget for maintenance and modernization being roughly 20 times
bigger than NAV CANADA’s capital budget, while the US system is only seven times as
large, in terms of traffic, as the Canadian one.
28. The FAA, and other ANSPs, would gain even more if they had adopted space-based
satellite navigation from the beginning for both their oceanic and continental airspaces,
instead of installing hundreds of ground stations to collect the ADS-B Out signals that
aircraft emit. But the FAA cannot be faulted for not using a solution that was not avail-
able when ADS-B ground stations started to be installed in 2007, and as such they are
now stuck in paying for a system potentially much more expensive to maintain.
REFERENCES
Air Traffic Management (2016) Nav Canada 2017 Fees to Save Airlines $100m,
accessed 1 August 2016 at http://www.airtrafficmanagement.net/2016/07/
nav-canada-2017-feestosave-airlines-100m/.
Aireon (2016) Aireon and ATNS Sign First Regional Commercialization
Agreement for African Air Traffic Surveillance Data, accessed 1 August 2016
at http://aireon.com/2016/05/03/aireon-and-atns-sign-first-regional-commercial
ization-agreement-for-african-air-traffic-surveillance-data/.
Arnaldo, R.M., V.F.G. Comendador, R. Barragan and L. Pérez (2014) European Air
Navigation Service Providers, Efficiency Evaluation Through Data Envelopment
Analysis (DEA), 29th Congress of the International Council of the Aeronautical
Sciences, St Petersburg.
Australian Aviation (2016) TAAAF Calls on Government to Privatise Airservices,
accessed 1 August 2016 at http://australianaviation.com.au/2016/04/
taaaf-calls-on-government-to-privatise-airservices/.
Averch, H. and L.L. Johnson (1962) Behavior of the Firm Under Regulatory
Constraint, The American Economic Review, 52, 1052–69.
Brown, D., T. Berry, S. Welman and E.J. Spear (2014), CAA International
Structures, McLean, VA: MITRE.
Button, K. and J. Drexler (2006) Are Measures of Air-Misses a Useful Guide
to Air Transport Safety Policy?, Journal of Air Transport Management, 12(4),
168–74.
Button, K. and R. Neiva (2013) Single European Sky and the Functional Airspace
Block: Will They Improve Economic Efficiency?, Journal of Air Transport
Button, K. and R. Neiva (2014a) Economic Efficiency of European Air Traffic
Control Systems, Journal of Transport Economics and Policy, 48, 65–80.
Button, K. and R. Neiva (2014b) European Air Navigation Services Industry
Regulatory Reform and Averch–Johnson Effects, International Journal of

M4371-FINGER_9781786431851_t.indd 375 14/11/2017 11:14

Civil Air Navigation Services Organisation (2015), Global Air Navigation Services
Performance Report 2015 – 2010–2014 ANSP Performance Results – The ANSP
View, Montreal, QC, Canada: Civil Air Navigation Services Organisation.
Dempsey, P.S., R. Janda, Y. Nyampong, J. Saba and J. Wilson (2005), The McGill
Report on Governance of Commercialized Air Navigation Services, Montreal,
McGill University Center for Research on Air & Space Law.
Federal Aviation Administration (2016a) Airport and Airway Trust Fund (AATF)
Fact Sheet, accessed 1 August 2016 at https://www.faa.gov/about/budget/aatf/
media/AATF_Fact_Sheet.pdf.
Federal Aviation Administration (2016b) FAA Aerospace Forecasts, accessed
1 August 2016 at https://www.faa.gov/data_research/aviation/aerospace_fore
casts/.
Helios Economics and Policy Services (2006) Cost Benchmarking of NATS Relative
to Selected Air Navigation Service Providers – Phase 2 Report, Farnborough and
London: Helios Economics and Policy Services and Civil Aviation Authority –
Economic Regulation Group.
Majumdar, A. (1995) Commercializing and Restructuring Air Traffic Control:
A Review of the Experience and Issues Involved, Journal of Air Transport
Management, 2(2), 111–22.
mbs Ottawa Inc. (2006) Air Traffic Control Commercialization Policy: Has it Been
Effective?, Ottawa, ON: mbs Ottawa Inc.
National Civil Aviation Review Commission (1997), Avoiding Aviation Gridlock
and Reducing the Accident Rate – A Consensus for Change, Washington, DC:
National Civil Aviation Review Commission.
National Partnership for Reinventing Government (2001) A Brief History of Vice
President Al Gore’s National Partnership for Reinventing Government During
the Administration of President Bill Clinton 1993–2001, accessed 1 August 2016
at http://govinfo.library.unt.edu/npr/whoweare/historyofnpr.html.
Neiva, R. (2015) Institutional Reform of Air Navigation Service Providers: A
Historical and Economic Perspective, Cheltenham, UK and Northampton, MA,
USA: Edward Elgar Publishing.
NERA Economic Consulting (2005) Cost Benchmarking of Air Navigation Service
Providers: A Stochastic Frontier Analysis, London: NERA Economic Consulting.
Nolan, M.S. (2010) Fundamentals of Air Traffic Control, 5th edn, Boston, MA:
Cengage Learning.
Office of the Inspector General (2014) ADS-B Benefits are Limited Due to a Lack
of Advanced Capabilities and Delays in User Equipage, Washington, DC: US
Department of Transportation.
Oster, C.V. and J.S. Strong (2007) Managing the Skies: Public Policy, Organization
and Financing of Air Traffic Management, Aldershot: Ashgate.
Standfuß, T., M. Schultz and M. Kreuz (2015) Multidimensional Benchmarking
of Air Navigation Service Providers, Air Transport and Operations Symposium
2015, Delft.
Train, K. (1991) Optimal Regulation: the Economic Theory of Natural Monopoly,
Cambridge, MA: MIT Press.
United States Government Accountability Office (2003) National Airspace System:
Reauthorizing FAA Provides Opportunities and Options to Address Challenges,
Washington, DC: United States Government Accountability Office.
United States Government Accountability Office (2005) Air Traffic Control:
Characteristics and Performance of Selected International Air Navigation Service

M4371-FINGER_9781786431851_t.indd 376 14/11/2017 11:14

Providers and Lessons Learned from their Commercialization, Washington, DC:

United States Government Accountability Office.
US Department of Transportation (2016) Budget Estimates – Fiscal Year 2017 –
Federal Aviation Administration, accessed 1 August 2016 at https://cms.dot.gov/
sites/dot.gov/files/docs/FAA-FY-2017-CJ.pdf.

M4371-FINGER_9781786431851_t.indd 377 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 378 14/11/2017 11:14

Index
Aer Lingus 69–70, 72–3, 75, 77–81, average hub premium 252
84–6, 89, 143, 152 ‘community air carrier’ 11
Africa grouping with IAG and Lufthansa
air navigation service providers 75, 87
361–2 independent LCC subsidiary 274,
air transport liberalization 186–8, 281
199 passengers 70, 148
airline industry operating with KLM 143, 148, 152
evolution 188–93 air navigation service providers
infrastructure developments 197–9 (ANSPs)
persistent problems 194–7, 199 charges levied by 289, 291
potential for growth 185, 193–4 commercialization of 359–60
airlines Federal Aviation Administration
flag carrier 187, 195 attempts at reform 370–72
low-cost carriers 196 history, governance and funding
and market structures 232–5 366–8, 372
net post-tax profits 189 NextGen 369–70, 372
number operating 190 liberalization initiatives
top eight African 190–92 Australia 361
top eight non-African 192–3 New Zealand 360–61
population characteristics 228–31 overview 359–60, 362–3
revenue passenger kilometres 189, percolation 372
194 see also air traffic control and
trends in air transport 222–8 NAV CANADA
Yamoussoukro Decision 187–8, 190, air service agreements (ASAs)
193, 199 bilateral
Air Canada Europe 141, 151–2, 168
air transport policy as legacy of 27–8 between LAC countries 157
change of heart towards Canada and United States 39
deregulation 31 India 98
early regulation 28 international 35–8, 221
government policy 30–31 air traffic control see air traffic
impact of rents 47 management
involvement in metal neutral joint air traffic control officers (ATCOs)
venture 39, 42–3 107, 341–2, 346, 353, 365–6,
new Canadian air policy 32–3 369–72
privatization of 34–5 air traffic management (ATM)
transborder market 17, 40–41 Africa 197–8
Air France communication standards 347–8
ability to take over potential conceptual framework 343–4
competitors 89 evolution of
379

M4371-FINGER_9781786431851_t.indd 379 14/11/2017 11:14

need for 339–40 policy environments in which they

up to 2000 345–8 operate 132–5
function of air navigation services Latin America and Caribbean 165–6
provision 339, 340–43 Chile 177
India 105–7 Colombia 175
performance 349–50 Mexico 171–2
service offered by 341–2, 347 market structures 232–5
Single European Sky I 348–9 see also Africa and Asia
Single European Sky II 349 airport capacity utilization and
Single European Sky II+ 350–51 capacity constraints 276–82
technologies to overcome gridlock Airservices Australia 361
centralized services 354–5 American Airlines 12, 17, 41, 80, 145,
centralized services 354–5 244, 251
disruptive technologies 351 Ansett 51, 53, 55–8, 65–6
remote towers 353–4 Asia
socio-technical systems approach African traffic 191, 194–6, 199
340, 343–4 air traffic growth 269, 272, 280, 281
virtual air traffic control 352 airlines and market structures
United States 341, 345–6 232–5
units 341–2, 352–3 airports 271, 273
air transport liberalization see Australia traffic 51, 65–6
liberalization Europe traffic 139
Airbus 24, 52, 185, 199, 228, 254 population characteristics 228–31
Aireon 356, 362, 365, 373–4 safety performance 212, 214–16
airline deregulation see deregulation state involvement 299
Airline Deregulation Act (ADA) take-offs and CUIs 278–9, 281
10–11, 215, 236, 266, 288 trends 222–8
airline industry Association of European Airlines
competitive advantage 245, 250, 252, (AEA) 71, 74, 263
258, 262 Australia
concentration level air navigation service provider 361
measuring 245, 254–6, 260 airport security charges 325
overall US domestic and bagging and screening costs 317–18
international HHI 257 deregulation
of traffic between airports commencement of 266
globally and by region impacts 55–7
270–74, 281 domestic liberalization 53–7, 59–61
US domestic HHI 258–9, 262 international liberalization
US international HHI 258, Asia 65–6
260–61 Australia–New Zealand market
entry and exit 248–9, 262 44, 53, 59, 64–5
interdependence 247–8 Europe and sixth freedom rights
mergers and acquisitions 249–51, 61–3
262 nature 58–9
pricing implications 250, 252–4 Japan routes 63–4
strategic groups 246–7 North transpacific route and
airlines Singapore Airlines 64
India and China ownership limits on international
efficiency performances 128–32, carriers 43
134–5 passengers and fatalities 212, 216–7

M4371-FINGER_9781786431851_t.indd 380 14/11/2017 11:14

Index 381
privatization 325 Canada

reluctant liberalizer 51, 66 air navigation service provider
social air service provision 237–8 359–60, 363–6, 368, 370–73
terminal access 56–7 air transport policy
Two Airline Policy 53–5, 206, 215 defining public interest for 46–7
Averch–Johnson effect 83, 363 government involvement in 27–8
aviation impacts of new approach to 45–6
drivers of transformation 2 legacy of 27–8, 45
early development of international new, 1984 31–3
accessibility 8–10 air transport trends 222–8
airlines and market structures 232–5
bankruptcy 78, 143–4, 248, 250 Blue Skies agreement with Ireland
Barrett, S. 73–4, 90 81
Baumgartner, M. 340, 345, 356–7 costs of aviation security 316–18,
Bermuda agreements 9–10 324–5, 329, 334
Berster, P. 267–8 deregulation
Bilotkach, V. 43, 139 future considerations 28–9, 31–3,
Bitzan, J. 14–15, 19, 22–3 46–7
Boeing 24, 30, 52, 58, 76, 80, 83, 95–6, large-scale 330
150, 185, 187, 193–4, 196, 198–9, international air service 35–7
228, 254, 290, 300 privatization 35
Boeing Commercial Aeroplanes 193–4, foreign ownership restrictions 43–4
196, 199 government policy
Borenstein, S. 15, 252, 300 Air Canada 30–31
Brazil and regulation 29
air service agreement with EU 168 international air service agreements
airport security charges 325 35–7
economic impact of Chile’s Open liberalization 206, 214–17
Skies policy 178 National Civil Aviation Security
foreign ownership limits 44 Program 309, 334
freight exchanges with EU 159–60 open skies agreement with EU 37–9
International Aviation Safety safety 79, 216–7
Assessment programme 169 social air service provision 237
military-run system 361 transborder market 39–42
passenger fatalities 214, 216 see also NAV CANADA and Air
private participation in airport Canada
infrastructure 163 Canada Transportation Act (CTA)
size of air transport market 123, 125, 29, 45
157–9 Canadian Transport Commission
social air service provision 239–40 (CTC) 28, 31–2, 34
sovereign state of LAC 156 Chicago Convention on International
Brexit implications 90, 139–40, 151–4 Civil Aviation (1944) 1, 9, 157,
British Airways 58, 61, 69, 71, 75, 89, 169, 187, 286, 341, 346–7, 359
143, 148, 152, 192, 244, 260 Chile
Bureau of Transportation Statistics airlines 43, 177, 215
(BTS) 24, 254, 257–61 airports 177, 179–80, 325
Button, K.J. 10–12, 24, 86, 90, 139, liberalization
159, 168, 186–9, 194, 200, 206, beginnings 176
248, 250, 266–7, 285–8, 291–2, embracing 180–81
294–5, 297, 334, 362, 373 Open Skies Agreements 176–8

M4371-FINGER_9781786431851_t.indd 381 14/11/2017 11:14

China sustainability 27–35, 39–42, 45–7

aviation market tendering 238
domestic 143 competitive advantage 245, 250, 252,
evolution 116–20 258, 262, 267, 274
growth 59, 65, 112 computable general equilibrium
comparison with India modelling 59
efficiency performances of airlines Coughlin, C. 310, 312, 333
128–32, 134–5 Cristea, A. 7, 13, 22
overview 123–8
policy environments 132–5 Dempsey, P. 10–11, 362
developing airports in Africa 198–9 deregulation
restrictions on use of airspace 361 Australia
safety 169, 212, 214–6 consequence of incomplete 60–61
security-related charges 324–5 and impacts of 55–7
cloud-based services 352–3 lengthy process of 66
Colombia Canada
airlines 175 future considerations for 46–7
airports 175–6 large-scale 330
liberalization international air service 35–7
beginnings 174, 215 privatization 35
embracing 180–81 timing 28–9, 31–3
Open Skies Agreements 174–5 China 112–3, 115, 117–18, 132
competition impact on Mexican airline industry
Africa 187–8, 191, 193, 199 172
Australia India 94–7, 112, 121–2, 124
domestic airlines 53–5, 66 safety 206–7, 217
from Gulf airlines 51 United States
and low-cost carriers 55–6 impacts 220–21, 223, 249
regulator of 60 mergers 244, 250
route to North America 64–5 movement toward 10–12
between low-cost airlines 297 pricing implications 250–52
China 117–18, 132–3 social air services 236–7
Europe 142–3, 153–4 see also Airline Deregulation Act;
extent of, on city-pair routes 233 liberalization; and competition
India 97, 100–101, 122 easyJet 70, 74–5, 77, 89, 148, 150–52,
Ireland 82–5, 87 245, 296
Marshall’s perfect 284, 292–3 economic perspectives on aviation
Mexico 170–72 security see security
oligopolization of markets 244–50, empty cores 292, 294–5, 300
252–3, 255–6, 262 environmental impact 107–8
route Etihad Airways 52, 61, 63, 102, 260
barriers to 23 Eurocontrol 341–2, 348–51, 354–5, 357
bilateral agreements enhancing 11 Europe
factors affecting effectiveness of Africa 186–7, 190, 192–4, 197
19 airports 271
findings 21–2 air traffic control 2, 355–6
lack of information on non-US airline profits 189
carriers 14 Australia 58–63
US liberalization policy 7–8 city-pair routes 232
stability 299–300 cross-border airlines 44

M4371-FINGER_9781786431851_t.indd 382 14/11/2017 11:14

Index 383
disruptive technologies 351–5 study on airport ownership 241

fatality rates comparison 216–17 Transport Directorate 242
hub premiums 252 European Union (EU)
Ireland 69, 71–5, 78, 82, 89 and Africa 187, 192, 197
liberalization agreement with India 98
argument 69 airlines and market structures 232–5
gradual 140–42 charter airline fares 71
outcomes 142–9 creation of single airline market
three packages 206, 208, 215 138–9
literature review 139 cross national mergers 43
low-cost carriers 138, 149–51 failure to assess mergers and
market shares 262–3, 269 takeovers from public
new market entrants 70–71 perspective 87
operating margins of airlines implications of UK leaving 90,
289–90 139–40, 151–4
passengers carried and passenger Ireland’s consolidation of aviation
fatalities 212 with 89
private investment drawn to 337 large number of low-cost carriers
secondary airports in 86–7 emerged 29, 262–3
security costs at airports 318–20 Latin America
single aviation market 44, 53, 59, flow of goods 159–61
64–5, 215 horizontal aviation agreements
Single European Sky 340, 342–5, 167–8, 181
348–56 perishables and exotics 180
and sixth freedom issue 61–3 safety issues 169
switch to point-to-point systems liberalization
266–7 gradual 140–42
take-offs and CUI values 278–9 outcomes 142–9
traffic concentration largely staff numbers 74
unchanged 272, 280–81 steps focusing on 43
traffic shares 223 three packages of reforms 286,
traffic structure development 275 288
US–international alliances 12 year of 69
European Commission Open Skies Agreements
British Airways cascade analysis 71 Canada 37–9
on civil aviation agreement with efforts to create network of 139
India 98 Morocco 191
deregulation driven by 1 Ukraine 153–4
efforts to create a ‘Single European United Kingdom 151–2
Sky’ 340, 342, 348–51 United States 11–12, 80–81, 152
on EU carriers 11 opposition to Ryanair’s takeover of
Latin American countries Aer Lingus 77, 89
compliance with safety standards passengers and security costs 320
168 performance 349–50
horizontal aviation agreements population characteristics 228–31
with 167, 181 revenue passenger kilometres 123
legislative force 238 social air service provision 238–9
as obstacle to liberalization at state aid rules 85, 87
secondary airports 87 trends in air transport 222–8
Ryanair Airport state aid cases 88 weak economic growth 269

M4371-FINGER_9781786431851_t.indd 383 14/11/2017 11:14

fare differentials for US–international Gudmundsson, S.V. 244, 246, 248, 250,
routes 7–8, 18–23 262
fares studies
airline liberalization 7–8, 10, 12–24 Herfindahl–Hirschman concentration
empty cores 294 index (HHI)
low fares, as strategy to reduce Canadian 40
market instability 296–7 explanation 255–6
oligopolization of markets 247–8, US domestic 258–9
250, 252–4, 256 US international 258, 260–61
in theories of market instability The Hindu 99, 102–3, 106
292–3 Hooper, P. 93–6, 115, 121, 266
Federal Aviation Administration horizontal aviation agreements 98,
(FAA) 167–8, 181
attempts at reform 370–72 horizontal relationships 309–12
comparison with NAV CANADA hub-and-spoke (HS) systems
359, 368, 372–3 airport capacity utilization/capacity
history, governance and funding constraints 276–82
366–8, 372 background 266–8
Hub Classification scheme 225, hub classification scheme 225, 227
241–2 traffic concentration 268–9, 270–76,
International Aviation Safety 280–81
Assessment programme 168–70
Ireland’s categorization 79 India
Modernization and Reform Act 236 Air Corporations Act 96, 121
NextGen 369–70, 372 Air Deccan 97, 100–101, 115–16,
space-based satellite navigation 373, 122, 124
375 Air India 92, 96–8, 115–16, 121,
fifth freedom 9, 36–7, 41, 45, 48, 63, 123–4, 126, 128–32, 134, 307,
81, 141, 188 337
Findlay, C. 59–60, 62, 115, 121, 123 Air Taxi system 94–5, 97, 121, 124
Finger, M. 206, 340, 343, 356 air traffic management 105–7
foreign ownerships Aircraft Act 103–4
eliminated limitations in Colombia Airports Authority of India (AAI)
175 102–4, 106–7
raised limit in China 118–20 aviation market post-deregulation
restrictions 100–102
Canada 38, 43–4 civil aviation industry
efficiency challenge 23 evolution 92–3, 112, 121–13, 124
post-Brexit scenario 151–2 literature review 115–16
setting limits 172 comparison with China
Forsyth, P. 52–7, 60, 162, 206 efficiency performances of airlines
Fu, X. 24, 114, 118 128–32, 134–5
full-service network carriers (FSNCs) overview 123–8
274–7, 281 policy environments 132–5
Director General of Civil Aviation
Gelhausen, M.C. 268, 277 (DGCA) 92, 96, 104, 106–7,
Gillen, D. 308, 337 126
Goetz, A. 76–9, 85, 266 East West Airlines 94–6, 121, 124
Goldstein, A. 115, 121, 123 environmental impact 107–8
greenfield airports 103–5, 107 foreign ownership limits 44

M4371-FINGER_9781786431851_t.indd 384 14/11/2017 11:14

Index 385
GoAir 97, 99, 116, 122, 124–6, defence of cornerstones 73

130–31 GDP as driver 269
Greenfield Airport Policy 93 airport codes 278
greenfield airports 103–5, 107 aviation security
Indian Airlines 92, 94–7, 115–16, developing move to ‘next
121–4, 134 generation’ 335
IndiGo 97–9, 101–2, 107, 116, limitation to plans for 336
122–6, 130–32 on responsibility for 331
Jet Airways 92, 95–8, 102, 107, commercial airline statistics 189, 290
121–4, 126, 130–32 criteria for defining bilateral
Kingfisher Airlines 97, 101, 122–4, agreements 37
126 fare recommendations 59
liberalization forecast of China’s air passengers
airport infrastructure policy 102–4 112
benefits 109 information on slot-controlled
impact on larger economy 99–100 airports 14
onset 93–5 location identifiers 225
safety 216 regional classification 222
‘Licence Raj’ 93 International Airline Group (IAG) 69,
low cost carriers 75, 77, 86–7, 89, 143, 148, 152,
effects of presence of 115–16, 274
122–3 International Civil Aviation
entry of 97–9, 124 Organization (ICAO)
fast growth of 125 Asia Pacific 106
market share 112 aviation security 308
policy environments in which they criteria for defining bilateral
operate 132–5 agreements as liberal 37
revenue passenger kilometres after Ireland 79–80
101 Latin America 173, 179
technical efficiency 129 meeting 335
Ministry of Civil Aviation 102, 107, provision of data 209–13, 215–17
129, 133, 239 safety and security regulations
National Civil Aviation Policy 108, 168–70, 180, 346–8, 360, 362
122, 239 social air service provision 235–6
Paramount 97, 101, 116, 124 International Civil Aviation Policy
passenger fatality rates 214, 216, (ICAP) 59, 62
307, 337 International Convention for Air
public–private partnerships 103–5, Navigation (ICAN) 345
108–9 Ireland
revenue passenger kilometres 98, CityJet 70, 77
101, 123–5, 129 Dublin–London route 69, 73–4, 77
social air service provision 239 EU member state 140
SpiceJet 97–100, 107, 116, 122, growth in air passenger traffic 145–6
124–6, 130–31 Irish Aviation Authority (IAA)
International Air Services Commission 79–80
(IASC) 60 liberalization
International Air Transport airport competition 82–5
Association (IATA) aviation safety 79–80
air passenger statistics 195, 306 European aviation, leading role
air transport development in 69

M4371-FINGER_9781786431851_t.indd 385 14/11/2017 11:14

North Atlantic routes 80–81 outcomes 362–3

on-going challenges 85–8 percolation 372
value to Irish economy 88–9 air transportation markets 285–7
passenger shares of airports 82–3 Australia
post-Brexit scenario 153 domestic 53–7
Ryanair international 58–66
emergence of 69–75 reluctant 51, 66
impact 75–6 India
wider geographical context 76–9 airport infrastructure policy 102–4
benefits 109
Jain, R.K. 115–16, 129 impact on larger economy 99–100
Japan 63–4, 66 onset 93–7, 124
Ireland
Kahn, A.E. 79, 85, 89, 247, 249, 285, airport competition 82–5
288 European aviation 69
Krishnan, R.T. 94, 96–7 North Atlantic routes 80–81
on-going challenges 85–8
Latin America and Caribbean (LAC) value to Irish economy 88–9
air transport market movement toward 10–12
institutional constraints 161 worldwide safety record in era of
overview 157–61 207–14, 218
airlines 165–6 Lijesen, M.G. 252–4, 256
Chile 177 low-cost carriers (LCCs)
Colombia 175 Africa 196
and market structures 232–5 Canada 46
Mexico 171–2 China 113–14, 119, 127, 129, 132–4
airports 162–5 Europe
Chile 177, 179–80 business model 144, 149–51, 274
Colombia 175–6 emergence of 29, 142
Mexico 172–4 explosive growth of 138, 148
Chile case study 176–81 post-Brexit scenario 151–2
Colombia case study 174–6, share of traffic 271, 273
180–81 hub concentration opening doors
commercial airlines net post-tax for 267
profits 189 India
Mexico case study 170–74, 181 effects 115–16, 122–3
Open Skies Agreements 166–8 entry 97–9, 124
Chile 176–8 growth of 125
Colombia 174–5 market share 112
Mexico 170–71 policy environments 132–5
pace of liberalization 156–7 revenue passenger kilometres 101
population characteristics 228–31 technical efficiency 129
social air service provision 239–40 introduction in Australia 55–8
sovereign states 156 Laker Airways 62
trends in air transport 222–8 in Latin America 161, 170–72, 175
LCCs see low-cost carriers (LCCs) lower dependence on hub traffic 274
liberalization and market instability 297–8
Africa 186–8, 199 policy environments in which they
air navigation services operate 132–4
initiatives 359–63 traffic concentration 281

M4371-FINGER_9781786431851_t.indd 386 14/11/2017 11:14

Index 387
traffic structure development and United States routes 16–18,

274–7 20–21
United States 144–5 year of liberalization 215
US–Mexico routes 17 Middle East
Lufthansa African connections 190–91, 194,
average hub premium 252 199
concerns over dominance 77 airports 271
consolidations 143 airlines and market structures
‘fifth freedom’ rights 141 232–5
grouping with IAG and Air France capacity situation 280
75, 87 commercial airlines net post-tax
hubs 144 profits 189
independent LCC subsidiary 274, impact on US 260, 262
281 passenger fatality rates 212
major European airline 58 population characteristics 228–31
metal neutral joint venture 39 share of independent FSNCs 275
operating costs and easyJet top non-African airlines 192
150–51 traffic growth
owning smaller European carriers continued positive 269, 281
152 traffic concentration increasing
passenger 70, 148 with 272–3
potential competitors 89 trends in air transport 222–8
top non-African airlines (available Morrison, S.A. 34, 59, 142, 285, 298
seats) 192 Morrison, W.G. 308, 337
market instability Nagpal, R. 102, 115–16, 128–9

coordination of suppliers 299 Natarajan, R. 115–16, 129
empty cores 292, 294–5, 300 Nathan Economic Consulting India
imperfections in models 284 94, 96, 134
liberalization 285–7 NAV CANADA
market shocks 289–92 comparison with FAA 359, 368,
present in all sectors 284–5, 299 372–3
public policy reaction 298–9 cost effectiveness 368
strategies to reduce 295–8 poster child of commercialization
theories of inherent 292–3 366
transition processes 287–8 private non-profit user cooperative
market oligopolization see 360
oligopolization of markets privatization 363–5
mergers and acquisitions 249–51, 262 undergoing institutional reform 359
metal neutral joint agreement 39, Neiva, R. 139, 362–3, 373–4
42–3 New Zealand
Mexico air navigation services liberalization
airlines 171–2 360–61
airports and slot allocation 172–4 Airways New Zealand 291, 359–61
as highly regulated market 170 baggage/screening costs per
lengthy and protectionist process of passenger 317
liberalization 181 hub connections 234
Open Skies Agreements 170–71 not-for-profit government
passenger fatality rates 214, 216 corporation 287
take-offs and CUI values 279–80 ownership limit 43

M4371-FINGER_9781786431851_t.indd 387 14/11/2017 11:14

single aviation market 44, 53, 59, UK–EU 151–2

64–5 UK–US 152–3
year of liberalization 215 Open Skies policy
‘next generation’ security 335–6 India in 1990s 94–7, 124
NextGen 369–70, 372 US, with all countries 46, 59
Njoya, E.T. 188–9, 192 origin–destination (OD) systems
North Atlantic routes 80–81 airport capacity utilization and
North Transpacific route 64 capacity constraints 276–82
background 266–8
O’Connell, J.F. 115, 122, 190–92, 194, study conclusions 280–82
200, 267 traffic concentration 270–74, 281
oligopolization of markets traffic growth 268–9, 280–81
industry concentration 256–62 traffic structure development 274–6,
industry entry and exit 248–9, 262 281
interdependence 247–8 OSAs see Open Skies Agreements
measuring concentration 245, 254–6, (OSAs)
260 Oster, C.V., Jr. 206, 212–13, 362
mergers and acquisitions 249–51, Oum, T.H. 24, 65, 181, 249
262
oligopoly perishables and exotics (P and E)
conditions 245 159–61, 167, 180
definition 244 Pitfield, D.E. 11–12, 24, 294, 298
pricing implications 250, 252–4 point-to-point (P2P)
strategic groups 246–7 low-cost carriers 97, 142–5, 148–51,
US domestic market 244–5 153, 247–8, 274
Open Skies Agreements (OSAs) systems 266–7
agreement as template 11 traffic 276, 281–2
Australia–New Zealand 64, 65 Porter Airlines 27, 45–6
Canada–EU 37–9 public franchise, aviation security as
Canada–US 39–41, 48 308–12
Canada’s interest in 35–6 public–private partnerships
EU–Morocco 191 Africa 198–9
EU–Ukraine 153–4 global 220, 298
EU–US 11–12, 80–81, 152 India 103–5, 108–9, 124
EU’s efforts to create network of 139 United States 237
flights covered by Pustay, M.W. 40–41, 246–7
data and empirical approach
14–20 ‘Q Code’ 345, 356
international fare findings 20–22 Qantas 52, 55–8, 60–65
past findings on 7, 10, 12–13
India–US 98 remote towers (RTWR) 353–4
Latin America and Caribbean revenue passenger kilometres (RPK)
166–8 98, 101, 123–5, 129, 189, 194
Chile 176–8 Reynolds-Feighan, A. 224, 267
Colombia 174–5 Roberts, T. 254–6
Mexico 170–71 Round, D.K. 60, 113, 117–18, 132
lower fares and greater service Ryanair
frequency 23 Aer Lingus 84–5, 89
prelude 32 barriers to entry at route level 245
replacing bilateral rules 330 Buzz acquisition 253

M4371-FINGER_9781786431851_t.indd 388 14/11/2017 11:14

Index 389
Charleroi 86–7, 89 financing

emergence 69–75, 75–6, 77 public finance 330–34
Irish airline financial instability 77 security-related charges 324–9,
low-cost model 97, 296 335
passengers global attacks inside aircraft 306–7
EU’s largest airline 69, 75, 89, 148 governance 329–34
planes and staff 70, 73–4 New Zealand security charges 326
post-Brexit scenario 151–2 next generation (NEXTGEN) 335–6
rapid expansion factors 75–6 production 312–14
share of transfer passengers at public franchise 308–12
London Stansted 267 public good 332–4
state aid cases 88 regulations in LAC countries
turnaround time 76 168–70, 180
under-utilized airport capacity 82 resource allocation problems 305,
use of internet 75 313, 334
VivaColombia 175 risk-based 335–6
US costs security 316–8, 324, 327,
S-curve effect 42 331
safety see also safety
air navigation services 360–62 Serebrisky, T. 164–5, 175
concerns, and liberalization 205–7, Single European Sky (SES)
217 beginning of project 345
deregulation 206–7, 217 conceptual framework 344
fatalities 79, 209–14, 216–17, 306–7, European Commission’s initiative
330, 337 340, 342
global record 207–14, 218 first package 348–9
Ireland 79–80 gridlock 343, 351–5
Irish airlines 79–80 performance dimension 349–50
LAC countries 168–70, 180 second package 349
oversight 157, 360–62, 367 standards 348
persistent problem for Africa 196–7 towards SES II+ 350–51
pillar of Single European Sky 349 sixth freedom rights 36, 41, 45–6, 51,
pre- and post-liberalization 214–18 58–9, 61–3
safety and security regulations slot-controlled airports 14, 17
168–70, 180 small communities
Ryanair 79–80 air transport service provision
Single European Sky I and II 348–9 changes since 1970s 220–21
UK and the North Atlantic 79 impact of regulation 221
United States 205–6, 216–7 airlines and market structures 232–5
see also security city pairs
Saranga, H. 102, 115–16, 128–9 average state length 233–4
security busiest 228–30
background 305–7 percentage of single-carrier 233
costs served by region 232–3
passenger screening 320–24 formation of multi-country trade
European airports 318–20 blocs 221
measuring and comparing 314–18 funding infrastructure provision 220
output 312–14 growth of traffic movements across
data and analysis 334, 336 hub classes 234–5
efficiency 320–24, 334–5 major trends in air transport 222–8

M4371-FINGER_9781786431851_t.indd 389 14/11/2017 11:14

number being served by multiple Trans Australian Airlines (TAA) 53

airports 225–7 transatlantic joint venture (TJV) 42
number of each type of hub by transborder joint agreements 42–3
region 227 transborder market 39–42
population characteristics 227, 231 Two Airline Policy
role of hubs in providing cessation of 215
connectivity 227–8 under scrutiny 54–5
social air services underlying theme 53
provision 235–40
purpose 221, 241 United Airlines 12, 17, 30, 40, 85, 244,
social air services 251
provision United-Continental Holdings (UCH)
Australia 237–8 42–3
Canada 237 United Kingdom (UK)
definition 235 air navigation service provider 291,
European Union 238–9 360
framework 235–6 airport security charges 327, 329
India 239 airports privatized
Latin America and Brazil 239–40 fully 330
United States 236–7 price cap regulation 287
purpose 221, 241 Australia 61–2
Southwest Airlines compliance rate 79
competitive advantage 252, 262 Ireland 69–74, 76, 81, 84
large proportion of monopoly routes landmark agreement with US
248 9–10
market share 246 leaving EU 90, 139–40, 151–4
mergers and acquisitions 251 percentage growth of air passenger
mobility barriers 262 traffic 145
modelling 47, 76, 97 Regional Air Connectivity Fund
new entrant 246 239
point-to-point route structure 247 Ryanair and Aer Lingus 89
raising barriers to entry at route level United States
245 Africa 187, 197
top US domestic airlines 244 agreement with Canada 32, 48
space-based satellite navigation 361–2, airlines operating margins 289–90
365–6, 373, 375 air navigation service provider
Ssamula, B. 194, 196–7 366–73
Star Alliance 27, 37, 41–2, 65, 190–91, airlines and market structures
275–7 232–5
Strong, J.S. 362 commercial airlines net post-tax
profits 189
Tata Group 92, 102, 121–2 Department of Transportation
terminal access (USDOT) 12–14, 16–17, 21, 39,
importance of 56–7 236, 256, 367–8
privatization 51 domestic airline market
total factor productivity (TFP) 57, 165 average fares 254
traffic barriers to entry 245
concentration 270–74, 281 concentration level 257–9, 262
growth 268–9, 280–81 leaders in 244–5, 262
structure developments274–6, 281 oligopoly in 244

M4371-FINGER_9781786431851_t.indd 390 14/11/2017 11:14

Index 391
foreign ownership restriction 43 welfare gains; OSAs-covered flights

international accessibility 8–10 7, 10, 12–13, 14–18, 20–23
international airline market unmanned air vehicles 342
concentration 258, 260–62
Ireland 79 Vega, H.L. 159, 161, 167, 294
liberalization 10–12, 215 Virgin Australia 52, 55–8, 60–61, 63–5
main LCC players 144–5 virtual air traffic control (VATC)
mergers and acquisitions 87, 143, 352–3
251–3, 262 Vowles, T. 76–9, 85, 266
non-scheduled services 211
Open Skies Agreements welfare gains 7, 10, 12–13, 14–22
all countries 46, 59 WestJet 27, 40, 42, 46–8
European Union 11–2, 80–81, Williams, G. 115, 122, 267
152 Winston, C. 7, 10, 13, 22, 59, 142, 285
India 98
United Kingdom 152–3 Yamoussoukro Declaration/Decision
population characteristics 187–8, 190, 193, 199
228–31 Yan, J. 7, 13, 22
similarities with EU markets 142
social air service provision 236–7 Zhang, A. 113–14, 117–19, 135
subsidies 299 Zhang, Y. 113–15, 117–19, 132–3
trends in air transport 222–8 Zorn, C.K. 206

M4371-FINGER_9781786431851_t.indd 391 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 392 14/11/2017 11:14

Air Transport Liberalization A Critical Assessment - Finger2017

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Air Transport Liberalization A Critical Assessment - Finger2017

Încărcat de

Drepturi de autor:

Formate disponibile

Air Transport Liberalization

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 1 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 2 14/11/2017 11:14

Cheltenham, UK • Northampton, MA, USA

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 3 14/11/2017 11:14

All rights reserved. No part of this publication may be reproduced, stored

Edward Elgar Publishing, Inc.

A catalogue record for this book

Library of Congress Control Number: 2017947105

This book is available electronically in the

ISBN 978 1 78643 185 1 (cased)

Typeset by Servis Filmsetting Ltd, Stockport, Cheshire

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 4 14/11/2017 11:14

PART I COUNTRY AND REGIONAL REPORTS

2 Airline liberalization in the US 7

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 5 14/11/2017 11:14

PART II TOPICAL ISSUES

11 Aviation safety in the age of liberalization 205

PART III FUTURE CHALLENGES

16 Economic perspectives on aviation security 305

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 6 14/11/2017 11:14

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 7 14/11/2017 11:14

Clinton V. Oster Jr, Professor Emeritus, School of Public and Environmental

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 8 14/11/2017 11:14

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 1 14/11/2017 11:14

While national interests were initially of primary importance in large

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 2 14/11/2017 11:14

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 3 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 4 14/11/2017 11:14

Country and regional reports

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 5 14/11/2017 11:14

M4371-FINGER_9781786431851_t.indd 6 14/11/2017 11:14

Air transportation has become an essential mode of international travel for

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 7 14/11/2017 11:14

historical review of liberalization of US–international aviation regula-

EARLY DEVELOPMENT OF INTERNATIONAL

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 8 14/11/2017 11:14

the stated goal of promoting economic growth and development, as well as

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 9 14/11/2017 11:14

negotiations between the US and UK occurred in Bermuda. The agree-

MOVEMENT TOWARD LIBERALIZATION

After 30 years, the UK and US terminated the Bermuda I agreement, in

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 10 14/11/2017 11:14

Bilateral agreements ratified during the 13 years succeeding the enactment

Matthias Finger and Kenneth Button - 9781786431851

M4371-FINGER_9781786431851_t.indd 11 14/11/2017 11:14

Easing of restrictions negotiated in OSAs presents carriers of member

PAST FINDINGS ON WELFARE GAINS ASSOCIATED

2 Airline liberalization in the US 7

11 Aviation safety in the age of liberalization 205

16 Economic perspectives on aviation security 305

Table 2.1 Descriptive statistics on US carrier characteristics and route

Table 2.2 US carrier options for passengers on US–Mexico Origin–

after correcting for heteroscedasticity if market concentration is endog-

Table 2.3 Estimation of airfares for one-segment international routes