Documente Academic
Documente Profesional
Documente Cultură
Growth
Shera, Adela . Journal of Information Technology and Economic Development ; Beverly Hills
Vol. 2, Iss. 1, (Apr 2011): 39-53.
ABSTRACT
Corruption, like cockroaches, has co-existed with human society for a long time. Corruption has received
significant attention among economists and international financial institutions during the last few decades, given
its implications for economic growth. There is an increasing volume of literature on the relationship between
corruption and economic growth, and the general conclusion is that corruption slows down the long-term growth
of an economy through a wide range of channels but also that it can have positive or beneficiary effect over
economic growth. It hampers economic growth, disproportionately burdens the poor and undermines the
effectiveness of investment and aid. Theoretical studies suggest that corruption may counteract government
failure and promote economic growth in the short run, given exogenously determined suboptimal bureaucratic
rules and regulations. As the government failure is itself a function of corruption, however, corruption should have
detrimental effects on economic growth in the long run. In this paper we will analyze and present: First, we will list
a number of possible causes and consequences of corruption, derived from a review of the recent empirical and
theoretical literature. Second, we will propose a model that can help explain the relationship between corruption
and the economic growth. In the third section, we will introduce several measures of corruption, certain empirical
correlations and their policy implications, we will then conclude in the last section.
FULL TEXT
Headnote
ABSTRACT
Corruption, like cockroaches, has co-existed with human society for a long time. Corruption has received
significant attention among economists and international financial institutions during the last few decades, given
its implications for economic growth. There is an increasing volume of literature on the relationship between
corruption and economic growth, and the general conclusion is that corruption slows down the long-term growth
of an economy through a wide range of channels but also that it can have positive or beneficiary effect over
economic growth. It hampers economic growth, disproportionately burdens the poor and undermines the
effectiveness of investment and aid. Theoretical studies suggest that corruption may counteract government
failure and promote economic growth in the short run, given exogenously determined suboptimal bureaucratic
rules and regulations. As the government failure is itself a function of corruption, however, corruption should have
detrimental effects on economic growth in the long run. In this paper we will analyze and present: First, we will list
a number of possible causes and consequences of corruption, derived from a review of the recent empirical and
theoretical literature. Second, we will propose a model that can help explain the relationship between corruption
and the economic growth. In the third section, we will introduce several measures of corruption, certain empirical
correlations and their policy implications, we will then conclude in the last section.
Keywords: Corruption, Economic Growth, Corruption Model, Investment, Bureaucratic Rules and Regulations.
(ProQuest: ... denotes formulae omitted.)
DETAILS
Location: Albania
Publication title: Journal of Information Technology and Economic Development; Beverly Hills
Volume: 2
Issue: 1
Pages: 39-53
Number of pages: 15
ISSN: 2153974X
LINKS
ABSTRACT
This paper adopts a productivity-based perspective in order to study how corruption conditions the efficiency
levels of the economies and their TFP growth rate. It attempts to identify the channels through which corruption
can influence productivity growth, whether by conditioning improvements in relative efficiency levels or by shifting
the production frontier. The results point out that corruption negatively affects both the levels of efficiency at
which the economies perform and the growth rate of TFP, suggesting that the negative impact of corruption on
technological progress manifest through its influence on human capital .[PUBLICATION ABSTRACT]
FULL TEXT
Port Econ J (2011) 10:109128
DOI 10.1007/s10258-010-0059-3
ORIGINAL ARTICLE
Received: 25 February 2009 / Accepted: 5 July 2010 / Published online: 23 July 2010 Springer-Verlag 2010
Abstract This paper adopts a productivity-based perspective in order to study how corruption conditions the
efficiency levels of the economies and their TFP growth rate. It attempts to identify the channels through which
corruption can influence productivity growth, whether by conditioning improvements in relative efficiency levels or
by shifting the production frontier. The results point out that corruption negatively affects both the levels of
efficiency at which the economies perform and the growth rate of TFP, suggesting that the negative impact of
corruption on technological progress manifest through its influence on human capital.
19A similar productivity index, based on Malmquist (1953), was proposed by Caves et al. (1982). However, these
authors assumed Dto(Xt, Yt) and Dt+1o(Xt+1, Yt+1) to be equal to unity, so that technical inefficiencies were not
considered.
20Method developed by Charnes et al. (1978), based on Farrell (1957) technical efficiency measures.
21Grifell-Tatj and Lovell (1995) show that the Malmquist index may not be an accurate measure of TFP change
when non-constant returns are assumed. Furthermore, the constant returns to scale assumption is sufficient
condition to guarantee that the optimization problem has a solution.
22Specifically,
Dto(Xt, Yt) = min{ : (Xt, Yt/) St}
= [max{ : (Xt, Yt) St}]1 = 1/Fto(Xt, Yt)
where Fto(Xt, Yt) is the Farrell output based measure of technical efficiency.
Corruption and total factor productivity: level or growth effects? 127
References
Abramovitz M (1986) Catching up, forgoing ahead, and falling behind. J Econ Hist 46(2):385406 Acemoglu D,
Zilibotti F (2001) Productivity differences. Q J Econ 116(2):563606Adkins LC, Moomaw RL, Savvides A (2002)
Institutions, freedom, and technical efficiency. South
Econ J 69:92108
Adser A, Boix C, Payne M (2003) Are you being served? Political accountability and quality of government. J Law
Econ Organ 19(2):445490Aghion P, Howitt P (1992) A model of growth through creative destruction. Econometrica
60(2):323351
Aidt T (2003) Economic analysis of corruption: a survey. Econ J 113:F632F652Aidt T, Dutta J, Sena V (2008)
Governance regimes, corruption and growth: theory and evidence.
J Comp Econ 36(2):195220Barro RJ (1990) Government spending in a simple model of endogenous growth. J Polit
Econ
98(5):103125
Barro RJ, Lee JW (2000) International data on educational attainment: updates and implications.
CID Working Paper 42Basu S, Weil D (1998) Appropriate technology and growth. Q J Econ 113(4):10251054
Benhabib J, Spiegel MM (1994) The role of human capital in economic development. J Monet
Econ 34(2):143174
Benhabib J, Spiegel MM (2005) Human capital and technology diffusion. In: Aghion P, Durlauf S
(eds) Handbook of economic growth. Elsevier, pp 935966Brunetti A, Weder B (2003) A free press is bad news for
corruption. J Public Econ 87(7):18011824 Caselli F (2005) Accounting for cross-country income differences. In:
Aghion P, Durlauf S (eds)
DETAILS
Volume: 10
Issue: 2
Pages: 109-128
ISSN: 1617982X
DOI: http://dx.doi.org/10.1007/s10258-010-0059-3
LINKS
ABSTRACT (ENGLISH)
Issue Title: Corruption Research in Vietnam
Corruption has long been considered a national illness in Vietnam but progress in fighting corruption has been
modest. In recent years, the Communist Party of Vietnam and the Government of Vietnam have strengthened their
efforts to prevent and fight corruption. Despite strong anti-corruption measures being implemented at the national
level, provincial authorities have shied away from tackling corruption. One of the reasons for this could be that it is
not clear to provincial authorities if and how corruption is affecting local development economically and socially.
This article demonstrates that corruption has a negative impact on private sector investment, employment and per
FULL TEXT
Crime Law Soc Change (2016) 65:325350
DOI 10.1007/s10611-016-9608-8
Abstract Corruption has long been considered a national illness in Vietnam but progress in fighting corruption has
been modest. In recent years, the Communist Party of Vietnam and the Government of Vietnam have strengthened
their efforts to prevent and fight corruption. Despite strong anti-corruption measures being implemented at the
national level, provincial authorities have shied away from tackling corruption. One of the reasons for this could be
that it is not clear to provincial authorities if and how corruption is affecting local development economically and
socially. This article demonstrates that corruption has a negative impact on private sector investment,
employment and per capita income at the provincial level in Vietnam. However, corruption is found to have no
significant impact on how income is distributed across provinces. The findings demonstrate that more effective
anti-corruption measures are necessary to promote the private sector and improve household income. The study
provides provincial leaders with empirical evidence and incentives for fighting corruption.
Introduction
Corruption is generally viewed in the larger academic literature as a hindrance to economic growth and
development. Corruption discourages investment by increasing costs and reducing certainty in the business
environment. It negatively impacts efficiency by diverting valuable resources from the public coffers into private
pockets, and from efficient projects to inefficient ones.
This article is an edited version of a paper by the same title, which was presented at the 2014 Vietnam Economist
Annual Meeting (VEAM), Ho Chi Minh City, 2425 June 2014 and the Costs of Corruption in Vietnam Research
Symposium, Hanoi, 2526 June 2014.
1 Central Institute for Economic Management (CIEM), Ministry of Planning and Investment,
http://crossmark.crossref.org/dialog/?doi=10.1007/s10611-016-9608-8&domain=pdf
In Vietnam, as pointed out in the 2012 Governments report to the National Assembly (Nguoi lao dong 2012)
corruption has long been named a national problem that threatens the existence of the regime. However, the
situation has not improved in recent years. The Vietnam Competitiveness Report 2010 1 ([1]: 14) states that
Bcorruption levels show few signs of falling^. By international standards, the situation of corruption in Vietnam
seems to have gotten worse in the past few years. In 2012, Vietnam dropped 11 places to the 123rd place in the
Transparency Internationals Corruption Perception Index (from 112th in 2011).
In recognition of the grave consequences of corruption, the Communist Party of Vietnam (CPV) and the
Government of Vietnam have in recent years stepped up efforts to prevent and fight corruption. This includes the
revision of the Law on Anti-corruption promulgated by the National Assembly in November 2012, and the
reallocation of the Central Steering Committee for Anti-Corruption from the Prime Ministers Office to the Partys
Secretary General. However, no major changes have yet been seen, possibly due to the fact that it takes time to
implement far-ranging policies such as anti-corruption measures. It can also be observed that local leaders in
Vietnam are not very active in fighting corruption. They seem to perceive that corruption happens somewhere else
and that anti-corruption initiatives have nothing to do with their mandate and responsibility.
This calls for a rigorous study that specifically demonstrates to local leaders how corruption can dampen local
development in Vietnam, which in turn can jeopardize or slow down their political careers. At the national level,
empirical studies of this kind can also provide relevant leaders and government agencies with valuable information
on the tangible costs of corruption in relation to socio-economic performance so that that they can remind their
provincial counterparts of the pernicious effects of corruption and demand that provincial leaders implement the
necessary changes.
Corruption is popular as a research subject. However, there are only a few evidence-based studies on corruption in
Vietnam. Perhaps this is due to the lack of data and attention from both government agencies and the academia.
Some noticeable studies on corruption in Vietnam include Nguyen and van Dijk [2], Gueorguiev and Malesky [3],
and the World Bank [4], as well as Rand and Tarp [5] and DEPOCEN [6]. This study hopes to contribute to the
limited knowledge of corruption in Vietnam, especially regarding its impact on local socio-economic development.
Corruption comes in many shapes and forms, and different types of corruption have different impacts on
development performance. Depending on the angles from which corruption is looked at, there are different ways to
1 This report was commissioned by the Government of Vietnam and written under the guidance of Professor
Michael Porter, Harvard Business School.
Conventional corruption studies focus on the impact of corruption on GDP or GDP per capita. However, in the
provincial context, and in view of the so-called legal target,2 GDP is not chosen as an indicator of economic
development, given that Vietnams national and provincial GDP estimates are notoriously incorrect and
inconsistent. Instead, this study focuses on three important issues for sustainable development in Vietnam,
namely, the development of the domestic private sector, income per capita and income inequality.
The two main variables in the analysis are the impact of corruption on i) domestic private sector investment and ii)
employment. Private sector investment has long been recognized as a significant driver of long-term growth, and
employment generated by private businesses is crucial in providing income to the population and maintaining
social order. As for social development, the analysis focuses on household per capita income and the Gini
coefficient. These indicators are important for local leaders because local economies are very much dependent on
private sector development, particularly after a wave of state-owned enterprises (SOE) privatization that took place
in the late 1990s and early 2000s. Household income and income inequality are also performance indicators that
are often stipulated in the five-year and annual socioeconomic development plans approved by the provincial
Peoples Council. If corruption can be shown to have a negative impact on these indicators, then these findings can
be used to pressure local leaders to implement more effective anti-corruption measures in their provinces.
The central research question of this study is how and to what extent corruption is affecting socio-economic
development in provinces of Vietnam, especially with regard to private investment, job creation, household income
and income inequality. In seeking to produce evidence about the supposedly negative impact of corruption on
socio-economic development in provinces of Vietnam, the ultimate goal of this study is to bring to the attention of
government agencies and the general public the costs of corruption in real terms so that leaders feel compelled to
take more immediate actions to fight corruption.
The next section reviews the literature regarding the impact of corruption on economic performance and income
inequality. It is followed by a discussion on the research approach and data employed in this study. The next
section presents and interprets the findings of the study. The final section concludes with some policy
discussions.
Literature review
Corruption is generally seen as an obstacle to investment and growth because it increases the costs of doing
business and reduces expected profits. In addition, corruption, which is secret in nature, makes the business
2 In Vietnam, every year the National Assembly passes a resolution on socio-economic development in which the
government is mandated to achieve some specific socio-economic targets which include, in particular, GDP growth
rate; hence the term Blegal target^. The same exercise is carried out at the provincial level.
more difficult for investors to make decisions, which in turn prevents investments from occurring. Corruption can
also exert a negative impact on economic growth by shifting valuable resources from efficient users to inefficient
ones, which can generate detrimental consequences for the economy. Researchers often refer to the negative
impacts of corruption on development and economic growth as the sanding the wheel effect.
On the contrary, some researchers have proposed that corruption may help development because it creates
incentive for low-paid bureaucrats to work harder in providing public services to firms. This is often referred to as
the greasing the wheel effect. Lui [8] analyses a model in which corruption works similarly to an auction where the
most efficient firms pay the highest price and get served, thus making the economy more efficient. However, these
theoretical arguments face serious criticisms. For example, corrupt bureaucrats may create delays to extract
bribes when they know firms are ready to pay more. Luis analytical work suffers from many shortcomings such as
the assumption of information symmetry (i.e., everybody knows about the auction of the services) and lack of
dynamic interaction between economic and political factors(i.e., inefficient but politically powerful firms may win
the deal). There also is some evidence in support of the greasing the wheel approach in the work of Dreher and
Gassebner [9]. They find that corruption facilitates firm entry in highly regulated economies where it takes more
than 50 days to start a business. However, this finding alone cannot lead to a conclusion about the overall impact
of corruption on economic growth because entry of some firms may limit entry of others and entry by itself does
not necessarily ensure increased investment and long-term economic growth.
In the current literature, there is ample empirical evidence about the negative impact of corruption on growth.
Mauro [10] shows that corruption reduces investment and lowers growth rates for a sample of 67 countries.
Campos et al. [11], while attempting to explain the East Asian growth-corruption irregularity, also confirms that in
any case, more corruption means less investment.
Corruption can have a negative impact on growth in an indirect way through its impact on the efficiency of
investments.3 Swaleheen [12] shows that corruption has a negative effect on efficiency which is measured by the
ratio of the increase in investment to the increase (or decrease) in GDP. Likewise, Lambsdorff [13] provides
evidence that corruption affects efficiency in the use of the entire stock of capital in a country. Specifically, that
article concludes: B[h]ad investments in Tanzania been undertaken at the low level of corruption that prevails in
the United Kingdom, the total output would have been 20 % higher.^
At the sub-national level, Del Monte and Papagni [14] study the relationship between corruption and growth across
Italian regions and find that corruption has a negative effect on growth, both through its effect on private
investment and on the efficiency of public expenditures.
At the firm level, Fisman and Svensson [15] show that corruption is negatively correlated with firm growth for a
sample of Ugandan firms. Using data of about 10,000 firms collected by the World Bank in 19992000, Asiedu and
Freeman [16] show that
corruption has a negative and significant effect on investment growth of firms in transition countries and is the
most important determinant of investment in these countries. In Vietnam, Nguyen and van Dijk [2] analyze a
sample of Vietnamese firms and find that corruption hampers the growth of private sector firms but not the state
sector ones. This study uses data collected by the World Banks 2005 Productivity and Investment Climate Survey,
which covers 24 provinces of Vietnam, and the Provincial Competitiveness Index (PCI) survey of 2005.
Studies have also shown the negative effects of corruption on foreign direct investment (FDI). Javorcik and Wei
[17] analyze firm data from 22 transition economies in Central and Eastern Europe and show that corruption
reduces the probability that a foreign firm looking for an investment destination would choose a certain country to
invest in. With a sample of both OECD and non-OECD countries, Egger and Winner [18] also show a negative impact
of corruption on FDI flows.
For a specific firm, corruption can either be an advantage or a disadvantage. On the one hand, corruption can be an
unbearable cost for a firm to expand to a new location or a new line of business. On the other, corruption can
provide another firm means of accessing valuable resources or great business opportunities. In other words,
corruption creates winners and losers. However, the aforementioned studies defending the sand the wheel
approach suggest that the net effect of corruption on a national or provincial economy is expected to be negative.
In theory, when political or grand corruption happens, it creates the possibility for the rich to get richer by
influencing policies to align with their own interests, thus worsening the inequality problem. Grand corruption can
lead to low government investment in education, which reduces access to education by poor people and limits
their opportunities to find good jobs and get out of poverty [19]. In addition, grand corruption can lead to higher
inequalities by creating incentives for higher investment in capital-intensive projects and lower investment in labor-
intensive projects [20].
There is empirical evidence on the negative impacts of corruption on income distribution. Gupta et al. [21] study
the impact of corruption in a cross section of countries and conclude that high and rising corruption increases
income inequality and poverty. Li et al. [22] examine the impact of corruption on inequality in a large sample of
countries and show that corruption is an important determinant of inequality with corruption alone explaining half
of Latin AmericaOECD Gini differential and all of the AsiaOECD Gini differential. Dong &Torgler [19], in turn, analyze
Chinas provincial level data and find that corruption increases income inequality and reduces government
spending on education. On the contrary, in the context of Latin America with a high level of informality, Dobson and
Ramlogan-Dobson [23] demonstrate that lower corruption is associated with higher income inequality. The
explanation is that lower corruption is associated with formalization, higher costs and lower employment for the
poor.
With respect to petty corruption, it can reduce private investment and job creation, thus preventing the poor from
getting jobs and income from the growth of the private sector [15, 16]. Petty corruption as measured by firm
Research approach
Most empirical studies on corruption that we are aware of utilize cross-sectional data at either the national or firm
level. This study greatly benefits from the availability of panel data extracted from the Provincial Competitiveness
Index (PCI), with observations of 63 provinces over the 20062012 period. Given this availability of data, panel with
unobserved effects is a favorable model because it can take into account unobserved province-specific
characteristics (that can have an impact on provincial economic performance) such as traditional and natural ones
(geography, climate, business tradition, etc.). If we do not take into account these variables that influence the
dependent variable but cannot be observed, our estimation will be biased because of the missing variable problem.
4 In this type of model, the unobserved effects are assumed to stay constant over time. Therefore, we apply the
following panel data model to empirically examine the relationship between corruption and development
performance:
In the above equation, Yi,t is the dependent variable which can be one of the four development indicators
discussed previously (i.e., investment, employment, household per capita income and Gini coefficient). Corruption
is one of several corruption variables which will be discussed in detail below. Controls is a vector of control
variables which are included to avoid omitted variable problems and obtain unbiased estimates of the effect of
corruption variables. ui is the unobserved province-specific effect as mentioned above and i,t is the idiosyncratic
error term (which is assumed to satisfy the classical OLS assumptions).
The control variables we use in this study are area, population, distance to Hanoi or Ho Chi Minh City, and
infrastructure development as proxied by the number of landline and post-paid mobile subscribers. These control
variables are included in the model because they are found to have significant effects on development indicators.
Naturally, land area and population are proxies for market size of the province which is normally proxied by GDP
per capita but this variable is not reliable in the Vietnamese context.
Distance to economic centers is a classical explanatory variable that is often included in research on investment
and economic growth (see, for example, [17]). The World Bank Economic Development Report 2009 dedicates an
entire chapter to the importance of economic distance and its implications for national economic development.
The level of infrastructure development is always important for investment because surveys show that firms often
cite infrastructure quality as an obstacle to doing business. In this study, the level of infrastructure development is
proxied by the number of landlines and post-paid mobile subscribers per thousand
4 If a variable has the potential to influence the dependent variable but it is not observed, hence not included, in the
estimation equation, the effects of included variables will carry the effect of the unobserved variable and the
estimated coefficients of included variables will be biased.
This research makes use of three major data sources. The first is the annual enterprise censuses that have been
conducted since 2000 by the General Statistics Office (GSO). This dataset gives the sum of firms long-term and
fixed assets by ownership type and by province which is used to measure private sector investment at the
provincial level. This dataset also provides the number of employees in domestic private sector firms. The second
is a series of household living standard surveys (VHHLSS) conducted every other year since 2002 by the GSO with
technical assistance from the World Bank. These VHHLSS are carried out in accordance with international
standards and can be considered the most reliable data source in Vietnam. GSO uses these VHHLSS to generate
the provincial average income per capita and the Gini coefficients, which this study employs to assess the impact
of corruption on income and inequality. The third data source is the Provincial Competitiveness Index (PCI) survey
conducted by the Vietnam Chamber of Commerce and Industry (VCCI) with technical and financial assistance from
the US Agency for International Development (USAID). This survey covers almost all aspects of the provincial
business environment and reflects the quality of local economic governance, including corruption. Specifically, the
three survey questions that are used in this study to measure corruption are:
1. Question 9, Section D of the PCI 2012 survey5: Do you agree with this statement: It is common for firms in my
line of business to have to pay some irregular additional unofficial payments? The percentage of firms in a
province that answers Agree or Strongly agree is used as the first corruption variable which indicates the
prevalence of corruption in a province. This variable is denoted as d10_new.
2. Question 10, Section D of the PCI 2012 survey: On average, what percentage of income do firms in your line of
business typically pay per annum in unofficial payments to public officials? The percentage of firms in a province
that answers 10 % or higher is used as the second corruption variable which measures the burden of corruption.
This variable is denoted as d11_new.
3. Question 11, Section D of the PCI 2012 survey: If a firm has made unofficial payments, are transactions
conducted exactly as they wish? There are five answers to this question: (1) Always; (2) In most of the cases; (3)
Sometimes; (4) Rarely;
5 Numbering may be different from one year to another, but the wording of the question has not changed since
2006.
and (5) Never. The percentage of firms in a province that answer Always or In most of the cases is used as the
third corruption variable, indicating the predictability of corruption. This variable is denoted as d12_new.6
These questions are standard ones that have been used in surveysincluding the World Bank Enterprise Surveys,7
Table 1 below shows descriptive statistics of the variables used in this study. There was a drop in private sector
investment and employment in 2009 as compared to 2008. The average income per capita increased quickly over
the period 20062012, but this is nominal income with the inflation effect not accounted for. The Gini coefficient
increased over the 20062010 period from 0.374 to 0.386, but fell to 0.38 in 2012. One thing to note here is the
small variation in Gini coefficient which may render the estimation of the impacts of corruption on inequality
meaningless due to insignificant coefficient estimates.
On average, the indicator of the prevalence of corruption (d10_new) shows signs of improvement over the years.
The average percentage of firms agreeing that paying bribery is common practice in their line of business falls
from 69 % in 2006 to 53.5 % in 2012. However, the burden of corruption (d11_new) decreases from over 13.47 % in
2006 to almost 8 % in 2011. Table 1 also shows that predictability of corruption improves over the period with the
average percentage of firms agreeing that transactions are done as expected, increasing from 47 % in 2006 to 62 %
in 2011 before falling to 58.4 % in 2012.
Table 2 shows the average investment, employment, income per capita and GINI by corruption quartile. Corruption
quartiles are created based on a corruption index which is the simple average of the prevalence of corruption
(d10_new) and the burden of corruption (d11_new). These two variables are observed consistently over the period
and are highly correlated. The corruption index ranges from 0 to 10, with 10 representing little to no corruption. As
we can see, the level of investment and employment rises as corruption levels fall, except for the 4th quartile.
However, income per capita consistently increases as we move from the 1st quartile to the 4th quartile. The
average Gini coefficient also falls as the severity of corruption lessens, except for the 4th quartile.
In the Appendix, we provide some figures that show the position of provinces in the corruption-development
sphere. As in the quartile comparison, the corruption index used here is the simple average of prevalence of
corruption and burden of corruption and is rescaled on a 010 range with 10 being the cleanest. What stands out of
these figures is that Ha Noi and Ho Chi Minh City have very high levels of
6 In addition to these perception questions, the PCI survey also makes use of the Unmatched Count Technique
(UCT) or List question to generate a more precise measure of corruption prevalence, but the response rate for
some provinces is not high enough to build a statistically significant measure for cross-province comparison. See
Coutts and Jann [27] and Blai and Imai [28] for details about UCTs.
7 Visit http://www.enterprisesurveys.org/
Web End =http://www.enterprisesurveys.org/ for details about the surveys and a list of studies that use the survey
data.
8 See, for example, Fisman and Svensson [15], Campos et al. [11], and World Bank [29].
Dependent variables
Subscribers Distance
(km)
mean std. dev. mean std. dev. mean std. dev. mean std. dev.
private investment, private employment and income per capital, but they suffer from considerable corruption which
is close to the national average. In 2012, the corruption index scores of Ha Noi and Ho Chi Minh City are 6.35 and
7.13 while the national average is 6.63. Some provinces that are doing well in both control of corruption and
private sector development are Binh Duong, Dong Nai, Da Nang and Long An. The provinces that are lagging
behind in both private investment and control of corruption are Northern mountainous provinces such as Bac Kan,
Cao Bang, Hoa Binh and Ha Giang.
Table 3 shows the correlation matrix of variables used in this study. The Gini coefficient is not significantly
correlated with any explanatory variables, which
Mean Std. dev. Mean Std. dev. Mean Std. dev. Mean std. dev.
Control variables
(thousand)
suggests that the estimation may not yield meaningful results. It can be seen that the measure of prevalence of
corruption (d10_new) and the measure of burden of corruption (d11_new) are positively and significantly
correlated with each other while they are not significantly correlated with measure of predictability of corruption
Based on the model (1) and data characteristics, we revised the detailed model specification as follows:
In Eq. (2), Corri,t 1 is the first lag of either the measure of prevalence of
corruption or the measure of burden of corruption. Because prevalence and burden of corruption are highly
correlated we do not use them in the same equation to avoid the problem of multicollinearity. Predi,t 1 is the first
lag of the measure of predict
ability of corruption. Regioni is a vector of regional dummy variables which represent seven geographical regions
of Vietnam9 and i is the vector of corresponding coefficients to be estimated. These regional dummies are added
to control for regional differences that cannot be measured but may have effects on the dependent variables. Areai
is the total land area of each province which does not change over time and Popi,t is the population size of each
province. Subi,t 1
is the first lag of the number of landline and post-paid mobile subscribers which is the proxy for infrastructure
development. Disti is the distance from the province to either Ha Noi or Ho Chi Minh City, with the closer economic
center used as a benchmark.
Model (2) is a standard panel data model with unobserved individual-specific effect which is usually estimated by
fixed effect (FE) or random effect (RE) methods. The difference between the two methods is that in case the
explanatory variables and ui are
9 The seven regions are: Red River Delta, Northern Uplands, Northern Central Coast, Southern Central Coast,
Central Highlands, Southeast, and Mekong Delta. The Red River Delta is the reference region in this study.
d10_newd11_newd12_newInvestmentJobIncomeGINIAreaPop.Distance
Note:*meanssignificanceat5%level.d10_new,d11_newandd12_newaremeasuresofprevalence,burdenandpredictabili
tyofcorruption,respectively
Table3Correlationmatrix
d10_new1
d11_new0.42*1
Investment0.13*0.41*0.32*1
Employment0.0040.39*0.17*0.91*1
Income0.41*0.48*0.29*0.81*0.63*1
GINI0.080.110.010.020.110.0591
Area0.16*0.31*0.010.17*0.21*0.27*0.091
Population0.13*0.21*0.16*0.57*0.43*0.65*0.010.0151
Distance00.22*0.14*0.25*0.26*0.22*0.050.45*0.12*1
Subscriber0.14*0.14*0.12*0.65*0.72*0.35*0.080.090.31*0.27*
correlated, the FE estimator is consistent while the RE estimator is not. However, if the explanatory variables and
ui are uncorrelated, both FE and RE estimators are consistent while the RE estimator is more efficient because it
makes use of the information about the structure of the variance of the composite error term ui +
i,t. The assumption that the explanatory variables and ui are uncorrelated is too strong in many cases and in our
study it seems the assumption is arguable, at least. The reason is because ui is meant to represent geographical
and cultural features that can be correlated with corruption practices, firms perception or tolerance for corruption,
and population size. That means the theoretical ground for RE is weak, leaving FE as a preferred choice. However,
FE has a disadvantage of not estimating coefficients of time-invariant variables because before estimation the
data is transformed such that ui is removed, hence the removal of time-invariant variables. In addition, both FE and
RE requires that all explanatory variables are not correlated with the idiosyncratic error term i,t in order to produce
unbiased estimators. To address this issue, we use the first lags of corruption variables,
Corri,t 1 and Predi,t 1, and the infrastructure variable Subi,t 1 as shown in the Eq. (2).
When endogeneity is suspected and/or confirmed by empirical tests, the usual alternative is to find instrumental
variables not included in the model and to apply the instrumental variable method (IV) to have consistent
estimators. However, it is notoriously difficult to instrument for such a complex variable as corruption because the
instrumental variable has to satisfy two conditions to be a valid instrument: (i) it has to be correlated with the
instrumented variable; and (ii) it cannot be correlated directly with the dependent variable. To our knowledge, in
Vietnam there are no available measured variables that can be valid instruments for corruption.
In this context, we decide to make use of the Hausman-Taylor estimation method (H-T) as it presents several
advantages. H-T offers consistent estimators when some explanatory variables can be correlated with ui while
time-invariant variables are not dropped.10 Essentially, to estimate Eq. (2) by H-T we will make two assumptions:
(i) except for Corri,t 1, Predi,t 1, and Subi,t 1 all other explanatory variables are not correlated with ui; (ii) all
In this section, we will present estimation results obtained by FE, RE and H-T methods at the same time. Tables 4,
5, 6 and 7 below show the estimation results of Eq. (2) with four dependent variables as described above. In these
tables, FE, RE and H-T estimations show consistent results.
10 Refer to http://www.stata.com/manuals13/xtxthtaylor.pdf
Web End =http://www.stata.com/manuals13/xtxthtaylor.pdf for an explanation and examples about this method.
Basically, this method uses exogenous variables as instruments to estimate coefficients of time-invariant
endogenous variables.
VARIABLES (1) RE_d10 (2) FE_d10 (3) HT_d10 (4) RE_d11 (5) FE_d11 (6) HT_d11
Predictability (lagged) 0.031*** 0.030*** 0.031*** 0.035*** 0.034*** 0.035***(0.003) (0.003) (0.003) (0.003) (0.003)
(0.003)
Population 0.001*** 0.002*** 0.001*** 0.001*** 0.002*** 0.001***(0.000) (0.000) (0.000) (0.000) (0.001) (0.000)
Subscribers (lagged) 0.000 0.000 0.000 0.001 0.000 0.001(0.000) (0.000) (0.000) (0.001) (0.001) (0.000)
Northern Central Coast 0.740** 0.738 0.601* 0.611(0.324) (0.650) (0.338) (0.966)
Southern Central Coast 0.005 0.039 0.031 0.105(0.405) (0.738) (0.357) (1.095)
Constant 8.833*** 7.139*** 8.853*** 6.956*** 4.714*** 6.670***(0.327) (0.697) (0.469) (0.252) (0.756) (0.575)
Observations 377 377 377 377 377 377 Number of pci_id 63 63 63 63 63 63 R-squared of overall model 0.720 0.600
0.728 0.600chi2 1378 . 538.2 1029 . 417.7P value of model significance 0 0 0 0 0rho 0.606 0.834 0.790 0.550 0.911
0.887 F for Hausman-Taylor model 44.85 34.81
Note: Robust standard errors are in parentheses. ***, ** and * mean p <0.01, p <0.05 and p <0.1 respectively. rho is
the ratio of the variance of the unobserved fixed effects, u_i, to the sum of variance of u_i and error terms _it. The
reference region is the Red River Delta. The Northern Uplands is sometimes divided into Northeast and Northwest
Table 4 shows regression results for private sector investment. We use the H-T estimation results in columns (3)
and (6) to interpret the impact of corruption. As
Variables (1) RE_d10 (2) FE_d10 (3) HT_d10 (4) RE_d11 (5) FE_d11 (6) HT_d11
Predictability (lagged) (0.003) (0.003) (0.003)0.009*** 0.009*** 0.009*** 0.010*** 0.010*** 0.010***
Population (0.000) (0.000) (0.000) (0.000)0.001*** 0.000** 0.001*** 0.001*** 0.001*** 0.000***
Subscribers (lagged) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)0.000 0.000 0.000 0.000 0.000 0.000*(0.000)
(0.000) (0.000) (0.000) (0.000) (0.000)
Northern Central Coast 0.277 0.272 0.236 0.234(0.231) (0.336) (0.226) (0.318)
Southern Central Coast 0.512 0.516 0.502* 0.510(0.312) (0.381) (0.291) (0.361)
Constant 10.483*** 10.109*** 10.487*** 9.964*** 9.462*** 9.962***(0.180) (0.311) (0.226) (0.154) (0.277) (0.196)
Observations 377 377 377 377 377 377 Number of pci_id 63 63 63 63 63 63 R-squared of overall model 0.764 0.629
0.775 0.667chi2 829.1 . 447.7 651.6 . 451.8P value of model significance 0 0 0 0 0 0rho 0.867 0.917 0.856 0.839
0.893 0.835 F for Hausman-Taylor model 37.31 37.65
Note: Robust standard errors are in parentheses. ***, ** and * mean p <0.01, p <0.05 and p <0.1 respectively. rho is
the ratio of the variance of the unobserved fixed effects, u_i, to the sum of variance of u_i and error terms _it. The
reference region is the Red River Delta. The Northern Uplands is sometimes divided into Northeast and Northwest
the Bsanding the wheel^ theory predicts, the coefficient of both prevalence of corruption and burden of corruption
are significant and negative. This result is
Variables (1) RE_d10 (2) FE_d10 (3) HT_d10 (4) RE_d11 (5) FE_d11 (6) HT_d11
Predictability (lagged) 0.015*** 0.014*** 0.015*** 0.017*** 0.018*** 0.019***(0.002) (0.002) (0.002) (0.002) (0.002)
(0.002)
Population 0.000 0.001* 0.000 0.000 0.001* 0.000(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Subscribers (lagged) 0.000 0.000 0.000 0.000* 0.000 0.000(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Northern Central Coast 0.045 0.041 0.004 0.005(0.095) (0.388) (0.099) (0.562)
Constant 7.134*** 6.140*** 7.122*** 6.358*** 4.981*** 6.116***(0.186) (0.594) (0.294) (0.134) (0.641) (0.343)
Observations 188 188 188 188 188 188 Number of pci_id 63 63 63 63 63 63 R-squared of overall model 0.651 0.265
0.609 0.210chi2 788.3 . 254.4 626.3 . 183.9P value of model significance 0 0 0 0 0 0rho 0.174 0.901 0.873 0.113
0.943 0.926 F for Hausman-Taylor model 21.20 15.32
Note: Robust standard errors are in parentheses. ***, ** and * mean p <0.01, p <0.05 and p <0.1 respectively. rho is
the ratio of the variance of the unobserved fixed effects, u_i, to the sum of variance of u_i and error terms _it. The
reference region is the Red River Delta. The Northern Uplands is sometimes divided into Northeast and Northwest
consistent across three methods. In other words, the prevalence and burden of corruption reduces private
investment at the provincial level. In line with what is
Predictability (lagged) 0.000 0.000 0.000 0.000 0.000 0.000(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Population 0.000 0.000 0.000 0.000 0.000 0.000(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Subscribers (lagged) 0.000** 0.000 0.000 0.000 0.000 0.000(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Southern Central Coast 0.013 0.016 0.016 0.018(0.016) (0.016) (0.016) (0.017)
Constant 0.343*** 0.370*** 0.352*** 0.364*** 0.393*** 0.366***(0.019) (0.031) (0.020) (0.013) (0.025) (0.014)
Observations 188 188 188 188 188 188 Number of pci_id 63 63 63 63 63 63 R-squared of overall model 0.268
0.00917 0.250 0.00975chi2 53.96 . 30.27 32.91 . 26.66P value of model significance 2.78e-07 0.645 0.00254
0.00100 0.819 0.00863 rho 0.436 0.607 0.404 0.444 0.600 0.448F for Hausman-Taylor model 2.523 2.222
Note: Robust standard errors are in parentheses. ***, ** and * mean p <0.01, p <0.05 and p <0.1 respectively. rho is
the ratio of the variance of the unobserved fixed effects, u_i, to the sum of variance of u_i and error terms _it. The
reference region is the Red River Delta. The Northern Uplands is sometimes divided into Northeast and Northwest
found in Campos et al. [11], at the same level of prevalence or burden of corruption, the predictability of corruption
(d12_new) has a positive impact on
private investment at the provincial level as well. More specifically, the coefficient of prevalence of corruption is
0.036 (column 3). This means that if the percentage of firms in a province agreeing with the statement that paying
bribes is common practice in their line of business is reduced by 1 %, the amount of private investments in that
province will be increased by 3.7 %.11 Likewise, if the percentage of firms agreeing with the statement that paying
bribes is common practice in their line of business is reduced by 10 %, private investments will increase by 43.3 %.
Regarding the burden of corruption, column (6) shows that the coefficient of burden is 0.062 which is much larger
than that of prevalence of corruption in terms of magnitude. Accordingly, a decrease of 1 % in the percentage of
firms saying that informal charges costs them 10 % or more of their revenue will lead to an increase of6.2 % in
private investments.
ity of corruption (d12_new) has a positive and significant effect on private investment with the coefficient equal to
0.031 or 0.035 when it is entered into the estimation equation with prevalence of corruption or burden of
corruption, respectively. This means, for the same level of burden of corruption, if the percentage of firms agreeing
that they can rely on bureaucrats to deliver services or contracts as agreed after receiving bribes goes up by 1 %,
then private investment will increase by 3.6 % (column (6)). This effect is similar to that of prevalence of corruption
but considerably smaller than that of burden of corruption. A possible explanation for the positive impact of
predictability of corruption is that the local business community has more confidence in the local business
environment when officials keep to their promise.
Table 5 shows the estimation results for private sector employment. It is not a surprise that the results here are
similar to those in Table 4 in terms of the coefficient sign because investment and job creation are highly
correlated. A one percent decrease in d10_new (prevalence of corruption) can help increase private sector
employment by 1 % (Column 3). The corresponding number for d11_new (burden of corruption) is 1.8 % (Column 6).
As for predictability of corruption (d12_new), a one percent increase will lead to a 1 % increase in private sector
jobs (column (3) &(6)).
Household income
Table 6 shows the estimation results for income per capita. As we can see, the prevalence and burden of
corruption also have large and negative impacts on
11 Because private investment is log-transformed before the estimation, this percentage value is calculated by
converting log value to percentage value by this mathematical transformation: ln_INVt ln_INVt_1 = 0.0.36
ln(INVt /INVt_1)=0.036 INVt -INVt_1 =e0.036 (INVt -INVt_1)/INVt_1 = e0.036 - 1.
income per capita, perhaps due to their impacts on private sector development. If a province can reduce the
prevalence of corruption (d10_new) by 1 %, then it can expect the local per capita income to increase by 1.5 %. As
for the burden of corruption, the impact is two times larger, with the expected growth of income per capita
standing at 2.3 % for a 1 % reduction in the burden of corruption. Improvement in the predictability of corruption is
also helpful in improving per capita income, since the expected increase in income is 1.5 % (column (3)) or 1.9 %
(column (6)) for 1 % increase in predictability of corruption.
Income distribution
Finally, Table 7 shows that corruption does not have significant impacts on distribution of income in Vietnamese
provinces. This is suggested by the scatterplot of corruption on the Gini coefficient in the Fig. 5 in the Appendix.
The reason is, as noted in the data description section, the variation of Gini coefficients at the provincial level is
very small. This is not a surprising result because previous studies (e.g. [30]) also show that there is no significant
relationship between investment and income inequality.
The control variables do not have significant impacts on inequality either, except for some regional dummies.
Table 7 suggests that Northern Uplands, Southeast and the Mekong Delta regions have more inequality than the
reference region of Red River Delta.
This study set out to investigate the possible impacts of corruption on private sector development, household
income and inequality at the provincial level in Vietnam. The objective was to find evidence of the costs of
corruption at the provincial level in order to better understand the kinds of incentives and pressures that are
required to fight corruption. By analyzing provincial level data from the GSO enterprise censuses and the PCI
surveys, this study has
i. The Impact of corruption on private sector development: Prevalence and burden of corruption have significant
and negative impacts on private sector development, as indicated by the level of private sector investments and
the number of jobs. Therefore, control of corruption brings about sizeable benefits for provincial economic
development. If the percentage of firms that agree that paying bribes is common in their line of business
(corruption prevalence) is reduced by 1 %, then private sector investment is expected to increase by 3.7 %. If the
percentage of firms agreeing that informal charges costs them 10 % or more of their revenue (corruption burden) is
reduced by 1 %, then private investment is expected to increase by 6.4 %. For job creation, the corresponding
effects of corruption prevalence and corruption burden are 1 % and1.8 %.
For the same level of corruption, the more predictable the outcomes of corruption are, the more investments are
made and jobs created. That is to say, the more firms believe they can get transactions done as expected after
paying bribes, the more investments are made and more jobs are created. One percent increase in the
predictability of corruption will lead to a 3.6 % increase in private investment and 1 % increase in private sector
jobs.
It is clear from this empirical analysis that maintaining predictability of corruption is not a good strategy to attract
investment because the prevalence and burden of corruption have stronger negative effects on investment and job
creation. It is imperative that local governments reduce corruption practices while making the local business
environment more predictable and transparent as well.
(ii) Impact of corruption on income per capita: Corruption prevalence, corruption burden and corruption
predictability all have significant impacts on the level of average per capita income. A one percent decrease in the
measure of corruption prevalence (d10_new) will lead to a 1.5 % increase in income per capita. A one percent
decrease in the measure of corruption burden (d11_new) will lead to a 2.3 % increase in income per capita. And a
one percent increase in the measure of corruption predictability will lead to an increase of at least a 1.5 % in
income per capita.
(iii) Corruption and income distribution: This study does not find any significant impact of corruption on the Gini
coefficient. Perhaps the main reason is the low level of variation in the provincial Gini coefficients. Cross-country
studies such as Barro [30] have found few links between inequality and investments. Analyses at district or
commune levels may
yield significant results because of greater variations in district or commune Gini coefficients.
This study also suggests that predictability in delivery of public services in particular and in overall regulatory
environment in general is very important for developing the private sector and raising peoples income. Therefore,
local governments should develop a culture of regulatory consistency and predictability to attract more investment
and achieve higher economic growth.
government apparatus from central to local level should get involved. Local leaders should be aware of the costs
of corruption because it has a significant and negative impact on the development performance of their province
and therefore on their career prospects as well. If we are to believe a recent World Bank study ([4]: 63), addressing
corruption at the local level could also be the right entry point, since local leaders at the commune, district and
province level tend to be more positive than officials at the central level about the effectiveness of anti-corruption
measures, such as increased openness and transparency, or the implementation of entitlement norms and
standards, administrative reforms, etc.
Appendix
Note: Corrup on Index is a simple average of corrup on prevalence and corrup on burden, converted to 0-10 scale.
Fig. 3 Corruption and private employment (ln), 2012. Note: Corruption Index is a simple average of corruption
prevalence and corruption burden, converted to 010 scale
Note: Corrup on Index is a simple average of corrup on prevalence and corrup on burden, converted to 0-10 scale.
Fig. 4 Corruption and income per capita (thousand VND per month), 2012. Note: Corruption Index is a simple
average of corruption prevalence and corruption burden, converted to 010 scale
Corrup on and income per capita (thousand VND per month), 2012
348 D. Quang Vinh
Fig. 5 Corruption and Gini coefficient, 2012. Note: Corruption Index is a simple average of corruption prevalence
and corruption burden, converted to 010 scale
2. Nguyen, T. T., &van Dijk, M. A. (2012). Corruption, growth, and governance: private vs. state-owned firms in
Vietnam. Journal of Banking and Finance, 36, 29352948.
3. Gueorguiev, D., &Malesky, E. (2012). Foreign investment and bribery: a firm-level analysis of corruption in
Vietnam. Journal of Asian Economics, 23, 111129.
4. World Bank. (2012). Corruption from the perspectives of citizens, firms and public officialsResults of
sociological surveys. Hanoi: National Political Publishing House.
5. Rand, J., &Tarp, F. (2010). Firm-level corruption in Vietnam. Economic Development and Cultural Change, 60(3),
571595.
6. DEPOCEN. (2012). Report on current status of corruption in business sector in Viet Nam. Hanoi: DEPOCEN.
7. Tanzi, V. (1998). Corruption around the world: causes, consequences, scope, and cures. IMF Working Paper
WP/98/63.
8. Lui, F. T. (1985). An equilibrium queuing model of corruption. The Journal of Political Economy, 93(4), 768781.
9. Dreher, A., &Gassebner, M. (2013). Greasing the wheels? the impact of regulations and corruption on firm entry.
Public Choice, 155, 413432.
10. Mauro, P. (1995). Corruption and growth. Quarterly Journal of Economics, 110(3), 681712.11. Campos, J., Lien,
D., &Pradhan, S. (1999). The impact of corruption on investment: predictability matters. World Development, 27(6),
10591067.
12. Swaleheen, M. U. (2007). Corruption and investment choices: a panel data study. Kyklos, 60(4), 601616.13.
Lambsdorff, J. G. (2003). How corruption affect productivity. Kyklos, 56, 457474.14. Del Monte, A., &Papagni, E.
(2001). Public expenditure, corruption, and economic growth: the case of Italy. European Journal of Political
Economy, 17, 116.
15. Fisman, R., &Svensson, J. (2007). Are corruption and taxation really harmful to growth? firm level evidence.
Journal of Development Economics, 83, 6375.
16. Asiedu, E., &Freeman, J. (2009). The effect of corruption on investment growth: evidence from firms in Latin
America, Sub-Saharan Africa, and Transition Countries. Review of Development Economics, 13(2), 200214.
17. Javorcik, B. S., &Wei, S.-J. (2009). Corruption and cross-border investment in emerging markets: firm-level
evidence. Journal of International Money and Finance, 28, 605624.
18. Egger, P &Winner, H. (2006). How corruption Influences foreign direct investment: a panel data study.
Economic Development and Cultural Change No.54.
20. Rose-Ackerman, S. (1997). Corruption and good governance, UNDP discussion paper series no. 3. New York:
United Nations Development Program.
21. Gupta, S., Davoodi, H., &Alonso-Terme, R. (2002). Does corruption affect income inequality and poverty?
Economics Governance, 3, 2345.
22. Li, H., Xu, L. C., &Zou, H. F. (2000). Corruption, income distribution and growth. Economics and Politics, 12(2),
155182.
23. Dobson, S., &Ramlogan-Dobson, C. (2010). Is there a trade-off between income inequality and corruption?
Evidence from Latin America. Economics Letters, 107, 102104.
24. Chakraborty, C., &Nandi, B. (2003). Privatization, telecommunications and growth in selected Asian countries:
an econometric analysis. Communications and Strategies, 52, 3147.
25. Barro, R. J. (2000). Inequality and growth in a panel of countries. Journal of Economic Growth, 5(1), 532.
References
26. Islam, N. (1995). Growth empirics: a panel data approach. The Quarterly Journal of Economics, 110(4),
11271170.
27. Coutts, E., &Jann, B. (2011). Sensitive questions in online surveys: experimental results for the Randomized
Response Technique (RRT) and the Unmatched Count Technique (UCT). Sociological Methods &Research, 40(1),
169193.
28. Blai, G., &Imai, K. (2012). Statistical analysis of list experiments. Political Analysis, 20, 4777.29. World Bank.
(1997). World development report: The state in a changing world (p. 10016). New York: Oxford University Press,
Inc.
30. Barro, R. J. (1991). Economic growth in a cross section of countries. The Quarterly Journal of Economics,
106(2), 407443.
DETAILS
Company / organization: Name: International Bank for Reconstruction &Development--World Bank; NAICS:
928120; Name: United Nations Development Programme; NAICS: 928120
Classification: 1200: Social policy; 2147: social problems and social welfare; sociology of crime;
9141: political economy; political economy; 9105: politics; national-level politics;
4300: Law; 1110: Economic conditions &forecasts; 9550: Public sector; 9130:
Experimental/theoretical; 9179: Asia &the Pacific
Volume: 65
Issue: 4-5
Pages: 325-350
Publication subject: Social Services And Welfare, Criminology And Law Enforcement
ISSN: 09254994
DOI: http://dx.doi.org/10.1007/s10611-016-9608-8
ABSTRACT (ENGLISH)
Fighting corruption is a vital aspect of good governance. When assessing government performance voters should
thus withdraw electoral support from government parties that turn a blind eye to or even engage in corrupt
practices. Whereas most accounts of performance-based voting focus on economic outcomes, we analyse
whether and to what extent voters punish incumbents for high levels of corruption. Using data from the
Comparative Study of Electoral Systems, we find that while voters perceiving high levels of corruption punish
incumbents, corruption performance voting depends on individual-level attributes and the electoral context: it is
most likely for non-partisans, for voters who believe that government turnover will bring about change, and in
systems where corruption is a salient issue. Yet, corruption performance voting is not moderated by the clarity of
political responsibility. Studying these conditions helps us to understand why corruption is more persistent in
some contexts than in others.
FULL TEXT
(ProQuest: ... denotes formulae and/or non-US-ASCII text omitted; see image)
Introduction
In representative democracies citizens hold the government accountable for its actions. By punishing poor and
rewarding good performance, voters express their preferences and force political elites to perform in line with their
preferences. Most empirical research on electoral accountability and the impact of retrospective performance
evaluations focuses on the economy (Lewis-Beck and Stegmaier, 2000; Anderson, 2007; Duch and Stevenson,
2008; Singer and Carlin, 2013). However, as voters can assign responsibility to various government outcomes,
there is no reason to believe that electoral accountability should be restricted to economic issues (Singer, 2011a,
b). In this article we focus on corruption - another dimension of the growing performance voting literature.
Corruption comprises activities such as bribery, embezzlement of public funds, misuse of public party funding, and
clientelism. Its severe detrimental effects on the economy and society (Mauro, 1995; Rose-Ackerman, 1999)
should motivate citizens to make vote choices based on their perceived level of corruption. We employ a reward-
and-punishment framework to study its electoral consequences. Corruption is a valence issue (Stokes, 1963, 1992)
where voters evaluate parties based on their ability to achieve a generally desired policy goal. As a result, citizens
should reward government parties for low levels of corruption and punish them for high levels. However, corruption
is also a highly persistent phenomenon (e.g. Alesina and Weder, 2002), which suggests that this simple reward-
and-punishment framework to 'throw the rascals out' is not the modus operandi in all contexts. Drawing on the rich
The empirical analysis is based on individual-level data on corruption perceptions and voting behaviour in 20
elections across Western and Central Eastern Europe from Module 2 of the Comparative Study of Electoral
Systems (CSES) survey. Past research on the effects of corruption on incumbents' electoral performance is often
based on macro-level data and addresses variation in corruption across countries (Ferraz and Finan, 2008; Krause
and Mendez, 2009; Choi and Woo, 2010). Micro-level analyses, in turn, are often linked to specific scandals and
certain national contexts (Peters and Welch, 1980; Welch and Hibbing, 1997; Vivyan et al., 2012; Basinger, 2013;
Praino et al., 2013; Winters and Weitz-Shapiro, 2013; Wagner et al., 2014). In this article, we employ the CSES data
to analyse corruption performance voting and its constraints under various micro- and macro-level factors. The
results show some similarities between economic and corruption performance voting, but also point to important
differences. As with economic issues, voters generally punish governments if they perceive high levels of
corruption. Moreover, corruption perceptions influence vote choice if voters believe that bringing in a new
government will make a difference. Non-partisans are more likely to make use of their corruption perception than
party-loyal voters. We also find that corruption voting is stronger in national contexts where corruption is a severe,
very salient policy issue. However, we find no evidence that corruption voting hinges on a political system's 'clarity
of responsibility', that is, the degree to which voters can assign responsibility to particular political actors (Powell
and Whitten, 1993).
These findings suggest that while the state of the economy is a major benchmark that allows voters to evaluate
incumbent parties, it is not the only yardstick they use to reward or punish incumbents. Democracies in Central
Eastern but also Southern Europe often face high levels of corruption and patronage that may overshadow
problems in the economic realm (see, e.g. Anderson and Tverdova, 2003; Kopecký et al., 2012). In these contexts,
including corruption perception enriches the performance voting framework and our knowledge of voting
behaviour. At the same time, clarity of responsibility as an alternative context factor seems to be less relevant for
corruption performance voting than for the economy. As a result, the context-dependent nature of corruption
performance voting allows us to better understand why corruption is highly persistent in some countries.
We explore whether vote choice is influenced by perceived levels of corruption. Conceptually, this approach relates
to the broad theoretical framework of performance voting models (Key, 1966; Kramer, 1971; Ferejohn, 1986).
Performance voting is based on the simple idea that voters evaluate parties based on their achievements in office.
It assumes that government performance enters a voter's decision-making process retrospectively. In Kramer's
(1971: 134) words: 'if the performance of the incumbent party is "satisfactory" [...], the voter votes to retain the
incumbent governing party in office [...]; if the incumbent's performance is not "satisfactory", the voter votes
against the incumbent to give the opposition party a chance to govern'. Thus, performance voting is based on
voter evaluations of the government parties' past performance irrespective of how they evaluate the opposition
parties' potential performance. The implicit assumption is that voters first and foremost care about 'throwing the
rascals out' if their performance was unsatisfactory.
So far, existing studies on performance voting have mainly focused on the economy (Key, 1966; Kramer, 1971;
Fiorina, 1981; Powell and Whitten, 1993; Whitten and Palmer, 1999; Lewis-Beck and Stegmaier, 2000; van der Brug
et al., 2007; Duch and Stevenson, 2008; Becher and Donnelly, 2013). There are various reasons for its
A thriving or deteriorating economy is, however, not the only criterion voters can use to evaluate the performance
of government parties (Fiorina, 1981; Stokes, 1992). As Stokes (1992: 147) notes:
'The classic illustration [for a valence issue] is good economic times and bad. But the parties and leaders are at
times linked to peace and war, internal order and crime, and many other conditions that are positively or negatively
valued. Although their bonding to economic conditions goes back to the early nineteenth century, the conditions
that provide the symbolic content of valence politics have progressively expanded with the scope of government in
the twentieth century.'
Thus, if voters care about monitoring officeholders, they might also condition their vote choice on other
dimensions of good governance such as public safety or peace. Here, we focus on one such non-economic issue:
the level of corruption. Corruption is commonly defined as 'the misuse of public office for private gains' (Treisman,
2000: 399) or as 'an act by a public official (or with the acquiescence of a public official) that violates legal or
social norms for private or particularistic gains' (Gerring and Thacker, 2004: 300). In both definitions public
officials are the major actors, while the extent and type of corruption are left open. Corruption thus comprises
activities such as bribery, embezzlement of public funds, misuse of public party funding, and clientelism. Gerring
and Thacker's definition also includes the toleration of corrupt practices. Hence, political elites may be held
accountable not just for themselves abusing power and money, but also for failing to limit corrupt behaviour in
general (Tavits, 2007).
Recent research highlights the role of corruption perceptions on political behaviour, in particular its effect on voter
turnout (Peters and Welch, 1980; Stockemer and Calca, 2013; Stockemer et al., 2013). Some studies show that
high levels of corruption decrease turnout (e.g. McCann and Dominguez, 1998; Stockemer et al., 2013) as high
levels of perceived corruption lead to more negative evaluations of political authorities and the political system in
general (Anderson and Tverdova, 2003). Yet, others find a positive effect of corruption on turnout (Stockemer and
Calca, 2013). Kostadinova (2009), in turn, argues that both effects may occur: there is a direct, mobilizing effect of
corruption on turnout and an indirect, negative one that affects turnout through political efficacy. Here, we will
extend this research on corruption and study its effect on incumbent voting.
We posit that voters evaluate government parties based on the perceived level of corruption for two reasons. First,
corruption is a valence issue (Stokes, 1963, 1992) and thus fits the reward-and-punishment logic of the
performance voting framework: voters should prefer less to more corruption and should evaluate the extent to
which the incumbent government has performed effectively.1Second, corruption is particularly relevant in some
Eastern and Southern European (e.g. Anderson and Tverdova, 2003; Kopecký et al., 2012) and Western European
countries (Singer, 2011b). For example, countries such as Romania or Bulgaria do poorly in Transparency
International's Corruption Perception Index (CPI), and voters in these countries also name corruption as one of the
most important issues (Singer, 2011b). Thus, in these contexts corruption should be a major element in voters'
In fact, the reward-and-punishment mechanism may even be more persuasive for corruption levels than for the
economy. Economic policies also entail aspects where policy means rather than goals are important for making
decisions. For example, voters may not only value parties that are competent in ensuring economic growth or low
unemployment rates, but also those that advocate policies (e.g. income tax, minimum wages) that are in line with
their personal policy preferences. These positional elements of party competition are less relevant for voters'
perceptions of corruption. In contrast to economic policies, parties differ less in how they propose to fight
corruption. Instead, corruption can almost exclusively be discussed in valence terms, making the reward-and-
punishment logic even more appealing than for economic policies. Moreover, corruption is an issue where it is
crucial that voters monitor the incumbents. This is due to the fact that political elites enjoy private gains when
exploiting their power and positions for their own good. They thus have clear incentives to pursue their own goals,
and these differ from those of voters. Finally, corruption is a particularly suitable issue for performance voting
because corruption, as defined above, is directly attributable to the political elites. Thus, it should be relatively easy
for the average voter to evaluate incumbent performance on corruption and to punish those who fail to deliver the
desired outcome. This leads us to expect that corruption perceptions have a strong effect on vote choice.
However, previous research has shown that performance voting is context-dependent (Powell and Whitten, 1993;
Whitten and Palmer, 1999; Hellwig 2001, 2011; Anderson, 2007; van der Brug et al., 2007; Duch and Stevenson,
2008, 2010). Hence, we expect that the connection between voter perceptions of corruption and vote choice is also
likely to be contingent on individual-level attributes as well as on macro-level context factors. We argue that some
of these contingency-effects will differ from those for the economy as a performance voting issue. While both
issues share many commonalities leading to similar effects and mechanisms, we also observe differences that
should affect the circumstances under which economic and corruption performance voting are most likely.2
A first moderating factor is partisanship: performance voting is more likely for non-partisans than for voters who
lean towards a particular party (van der Brug et al., 2007; Kayser and Wlezien, 2011). This argument goes beyond
the widely known findings that partisanship affects voter perceptions of politics (Campbell et al., 1960; Bartels,
2002; Anduiza et al., 2013) and vote choice (see e.g. Merrill and Grofman, 1999). Rather, it states that partisanship
moderates the impact of (potentially biased) perceptions of politics on vote choice (Vivyan et al., 2012). Voters
with strong partisan attachment should be less responsive to short-term factors including performance
evaluations because of their clear-cut party preferences. Although these voters might be dissatisfied with the
performance of their most preferred party, their strong preferences for that party make party switching less likely
(Kayser and Wlezien, 2011; Eggers et al., 2014). In contrast, non-partisans lack these strong predispositions, and
thus even small differences in their perceived party performance can have significant effects on their vote choice
(van der Brug et al., 2007). Following the literature on economic voting, we test whether non-partisan voters put
greater weight on the government's record in tackling corruption:
Hypothesis 1. The influence of corruption performance voting is higher for non-partisans than for voters who feel
close to a party.
Retrospective voting should also depend on voters' belief that throwing the rascals out will not bring new rascals
in. Punishing government parties by voting for opposition parties presumes that voters believe in the effect of
government turnover. This belief is captured in several theoretical concepts. It relates to voter perceptions of
government responsiveness (Kölln et al., 2014) and is reflected in what Downs (1957) has labelled the 'expected
Hypothesis 2. The influence of corruption performance voting is higher if voters expect significant change when
the government composition changes.
Turning to factors in the electoral context, we test whether the effect of corruption performance voting depends on
the severity of corruption. One of the core elements of issue voting models is that voters evaluate parties based on
their most important problems and concerns (Krosnick, 1988, 1990). Thus, economic performance voting is most
likely in hard economic times and among those who are hit hardest by unemployment and recessions (Stevenson,
2002; Singer, 2011a, b).
Following this argument, we test whether voters pay more attention to their corruption perception when evaluating
government performance in electoral contexts where corruption is a prominent problem. In these contexts, the
misuse of public goods is more relevant to voters when evaluating government performance. Thus, it is here where
voters can use their retrospective evaluation of corruption as an information shortcut (Fiorina, 1981). In contrast,
the effect of corruption perception might be weaker in national contexts where corruption is less severe. Here,
voters are more likely to focus on other issues (e.g. the economy) to assess the performance of government
parties. Therefore, corruption performance voting should be more likely in contexts where corruption is a
prominent problem. This leads to:
Hypothesis 3. The influence of corruption performance voting is higher in electoral contexts where corruption is a
prominent issue.
Another factor that might hinder performance voting is the clarity of political responsibility. In order to punish
incumbent parties, voters need to be sure that bad performance was truly a result of the officeholders'
mismanagement - and not a result of exogenous factors or actions taken by other political actors. One of the major
findings in the economic voting literature is that evaluating parties and candidates is much easier in systems
where performance can clearly be attributed to individual actors (Powell and Whitten, 1993; Whitten and Palmer,
1999; Anderson, 2007; Duch and Stevenson, 2008). It is easier for voters to evaluate the responsible parties, for
example, if the government composition is relatively stable and if the incumbent parties have a majority in
parliament. Moreover, the consensus-seeking common in coalition governments makes it difficult for voters to
identify the party or parties that are responsible for policy outputs, and this challenge increases with the increasing
number of coalition partners.
We test whether the same moderating effect exists for corruption performance voting. Following the arguments
from the economic voting literature, corruption performance voting should be more likely in electoral contexts
where government parties have unified control over policy-making. Higher clarity of responsibility should make it
easier for citizens to identify corruption and scandals, and a high level of information about corrupt practices is
linked to corruption performance voting (Winters and Weitz-Shapiro, 2013). There are of course differences
Hypothesis 4. The influence of corruption performance voting is higher in electoral contexts where the clarity of
political responsibility is high.
There are only a few cross-national studies on corruption performance voting using individual-level data. Most
studies focus on particular events (e.g. an expenses scandal) to study the effects of bribery on voters' opinions
and vote choice (Peters and Welch, 1980; Vivyan et al., 2012; Wagner et al., 2014). Moreover, many analyses use
survey experiments (Anduiza et al., 2013; Winters and Weitz-Shapiro, 2013; Eggers et al., 2014; Wagner et al., 2014)
to provide insights on the causal mechanisms in decision-making processes. Yet, analyses based on a single
national context do not allow us to study cross-national differences even though such differences are likely to
exist (van der Brug et al., 2007; Duch and Stevenson, 2008). Comparative analyses are rare and largely limited to a
macro-level perspective (Krause and Mendez, 2009; Choi and Woo, 2010). These analyses run the risk of
committing ecological fallacies and are not well-suited to testing the causal mechanism linking citizens'
perceptions of corruption and reward-punishment behaviour.
We study individual voting behaviour in 20 parliamentary elections in Western and Central Eastern Europe using
the CSES Module 2 survey data (CSES, 2007).3These data enable us to study the effect of corruption perception
using identical question wording in various political and economic contexts.
Dependent variable
Our dependent variable is vote choice, distinguishing respondents who voted for an incumbent government party
(1) and those casting a vote for an opposition party (0).4The data on the government status of each political party
running in the respective election is retrieved from the ERD data archive (Andersson et al., 2014).
Independent variable
The key independent variable captures each individual's perceived level of corruption and is based on the following
survey item:
How widespread do you think corruption such as bribe taking is amongst politicians in [COUNTRY]: very
widespread (4), quite widespread (3), not very widespread (2), it hardly happens at all (1)?5
Moderating variables
Since our interest lies primarily in the contingent effect of corruption perception, our empirical models feature
interaction terms with four moderating variables. First, we test whether partisanship moderates the effect of
corruption performance voting: Non-partisans should be more likely to vote based on their performance
evaluations than voters who feel close to a political party. We identify voters' partisanship using the CSES question
item 'which party do you feel closest to?' and use indicator variables to distinguish between those who feel close
to government parties, opposition parties, and non-partisans. As above, the data on government participation of
political parties are retrieved from the ERD data archive.6The second individual-level characteristic refers to the
voters' belief that government turnover will bring about change. We capture this belief with an item (measured on a
five-point scale) indicating whether respondents think that 'it makes a difference who is in power' (5) or not (1).
For macro level effects, we measure the severity of corruption using the CPI provided by Transparency
International. The CPI is a composite index and combines surveys and assessments of corruption by several
institutions and organizations into a single measure. It is generally considered the most valid and reliable cross-
national estimate of corruption and is thus widely used in comparative studies (see, e.g., Treisman, 2000; Persson
et al., 2003; Krause and Mendez, 2009; Choi and Woo, 2010).7For each election we capture the severity of
corruption via the country's average CPI score over the entire parliamentary term. In this context, we inverse the
original eleven-point scale so that the maximum score (10) corresponds to high levels of corruption, while a
minimum score (0) indicates low corruption levels.
Our last moderating variable is clarity of responsibility. There is no established measure of clarity of responsibility
in the existing literature (see, e.g. Whitten and Palmer, 1999; Tavits, 2007; De Vries et al., 2011). Hobolt et al. (2013)
argue that attributes of the incumbent government (which they label 'government clarity') are more relevant than
institutional factors (such as bicameralism and the committee structure in parliament). Thus, we follow their
approach by compiling a composite index of government clarity ranging from 0 (low clarity) to 1 (high clarity). This
index is based upon each government's (i) single-party status, (ii) the extent of cohabitation within semi-
presidential systems, (iii) a measure of its ideological cohesion, and (iv) the main governing party's share of
cabinet posts.
Control variables
The incumbent government's economic performance is captured via the level of unemployment in the year before
the election. Ideally, including measures of the perceived state of the economy at the individual level would allow
us to test the effect of corruption performance voting side-by-side with economic performance voting.
Unfortunately, such individual-level measures are not included in the CSES Module 2 survey data.8We resort to
using the level of unemployment as a macroeconomic indicator that is likely to have the most direct effect on each
Empirical analysis
Given the hierarchical structure of our data with individuals nested within elections and the binary nature of the
dependent variable, the subsequent empirical analysis rests upon a series of multilevel logistic regression models
of incumbent voting with random intercepts accounting for unobserved heterogeneity between elections.10The
results of these hierarchical logistic regression models are shown in Table 1. Model 1 shows the overall effect of
corruption performance on vote choice. Model 2 includes moderating effects for both individual-level factors
(Hypotheses 1 and 2). We test the electoral context effects of the systemic level of corruption (Hypothesis 3;
Model 3) and of clarity of responsibility (Hypothesis 4; Model 4) separately. All estimated marginal effects and
associated marginal-effect plots are based on a full model (Model 5) including all covariates.
Table 1
Close to
Reference -3.053***
H1 Non-partisan opposition
category (0.244)
party
Close to
-3.058*** -3.077*** 2.855***
-3.073*** (0.247) government
(0.244) (0.247) (0.195)
party
Corruption
perception×
2.842*** 2.851*** 0.118
2.862*** (0.194) close to
(0.196) (0.194) (0.083)
opposition
party
Corruption
perception×
0.120 0.127 0.081
0.126 (0.084) close to
(0.083) (0.084) (0.071)
government
party
Corruption
perception×
0.261*** 0.259*** 0.260*** perceived
0.261*** (0.051)
(0.051) (0.051) (0.051) effect of
government
turnover
Corruption
0.116 0.133 perception×
(0.085) (0.083) corruption
severity
-0.034* -0.033*
H4
(0.015) (0.015)
-0.311 -0.393
Government clarity
(0.596) (0.593)
-0.135 -0.120
Corruption perception×government clarity
(0.116) (0.116)
N
19,782 19,782 19,782 19,782 19,782
(elections)
Log
20 20 20 20 20
likelihood
Likelihood
ratio test
25,075.1 16,287.5 16,286.3 16,288.5 16,287.3 vs. logistic
regression
(P-value)
We start by analysing whether voters resort to corruption as a simple means of retrospective voting. As expected,
corruption perception has a significant negative effect on incumbent voting. Thus, individuals perceiving higher
levels of corruption are less likely to vote for a party in government. Increasing the level of corruption perception by
one unit (on a four-point scale) lowers the probability of incumbent voting by ~5.4 percentage points, with all other
independent variables held at their mean or mode, respectively.11To put this effect in perspective, we compare this
estimate with that of a one unit change of voters' economic perceptions. Studying 163 different estimates of
economic perception voting, Duch and Stevenson (2008: 64) find that a one-unit change in economic perceptions
(on a three-point scale) decreases the probability of incumbent voting on average by 5.0 percentage points.
Despite the difficulties of comparing effect sizes across the different models,12we may tentatively conclude that
the substantive impact of corruption perceptions on vote choice is similar to that of economic perceptions.
This baseline effect of corruption perception on incumbent voting may serve as a starting point to evaluate the
conditional factors in Hypotheses 1-4. Starting with the individual-level factors that may affect performance voting,
our results indicate that the effect of corruption perception on incumbent voting is characterized by a strong
partisan bias. Figure 1 juxtaposes the particularly influential effect of corruption perception for non-partisan voters
with that of voters feeling close to an opposition party, and those who feel close to the incumbent government.13
Corroborating our first hypothesis, the perceived level of corruption has only little influence on partisans. In fact,
individuals closely tied to the incumbent are likely to support them regardless of their corruption perception, as
Figure 1
Corruption perception effect depending on party affiliation (H1). Note: This figure shows the estimated marginal
effect of a one-unit change in corruption perceptions on the probability of incumbent voting for non-partisans,
voters feeling close to an opposition party, and voters feeling close to a government party. All point estimates
(dots) and corresponding 95% confidence intervals (vertical lines) are based on Model 5.
A second moderating factor is whether respondents believe that government turnover will yield changes. In line
with Hypothesis 2, Models 2 and 5 show a negative interaction between corruption perception and the perceived
differences of who is in power, and this effect is statistically significant at conventional levels. Thus, the extent to
which voting behaviour is influenced by corruption perceptions increases as respondents believe that opposition
parties will govern differently than the incumbent government. Figure 2 shows the marginal effect of corruption
perception dependent on the perceived effect of government turnover, holding all other variables at their mean or
mode, respectively. Here, we observe that corruption perceptions have almost no impact on the likelihood of
incumbent voting (increase by ~0.2 percentage points, insignificant) for individuals who have little faith in
government turnover. In fact, a minimum sense that 'it makes a difference who is in power' is necessary so that
corruption perceptions influence one's voting behaviour.14For these voters, an increase in the level of perceived
corruption negatively affects the probability to vote for incumbents by ~3.7 percentage points, and this effect
becomes stronger as the perceived effect of government turnover increases. For individuals who strongly believe
in the changing effect of government turnover (five scale points), increasing the level of corruption perception by
one unit decreases the likelihood of incumbent voting by 7.8 percentage points.
Figure 2
Corruption perception effect depending on effect of government turnover (H2). Note: This figure shows the
estimated marginal effect of a one-unit change in corruption perceptions on incumbent voting, conditional on the
perceived effect of government turnover. All point estimates (dots) and corresponding 95% confidence intervals
(vertical lines) are based on Model 5.
Turning to the macro-level context factors (Models 3 and 5), we hypothesized that corruption performance voting
is more relevant in countries with high levels of corruption (Hypothesis 3). Figure 3 shows the corresponding
marginal effect of corruption voting for varying levels of corruption, again holding all other variables in the model
at their mean or mode, respectively. Indeed, we observe a strong punishment effect by the electorate in national
contexts where corruption is highly salient (e.g. Romania in 2004). If corruption perceptions increase by one unit in
such contexts, we would predict that the probability of voting for the incumbent declines by 9.2 percentage points.
This effect is approximately three times larger than the corresponding effect in elections where corruption is
Figure 3
Corruption perception effect depending on corruption severity (H3). Note: This figure shows the estimated
marginal effect of a one-unit change in corruption perceptions on incumbent voting, conditional on corruption
severity. All estimates (solid line) and the corresponding 95% confidence interval (dashed lines) are based on
Model 5. The empirically observed maximum on the Corruption Perception Index (CPI) 0-10 scale is 7.2.
Concerning Hypothesis 4, the results of Models 4 and 5 do not support the claim that corruption performance
voting varies with the clarity of government responsibility. While the corresponding coefficient for Corruption
perception×government clarity is negative, it fails to reach conventional levels of significance, indicating that
varying levels of government clarity do not affect the effect of corruption voting.15Thus, these results indicate that
individuals are able to clearly attribute and punish corrupt behaviour, regardless of factors blurring this link (see
Figure 4).
Figure 4
Corruption perception effect depending on government clarity (H4). Note: This figure shows the estimated
marginal effect of a one-unit change in corruption perceptions on incumbent voting, conditional on the level of
government clarity. All estimates (solid line) and the corresponding 95% confidence interval (dashed lines) are
based on Model 5.
Finally, Models 1-5 also contain the level of unemployment as a control variable. In line with the expectations of
the economic performance voting literature, the coefficient of unemployment is negative: an increase in
unemployment by one standard deviation above the mean decreases the probability to vote for incumbent parties
by 7.9 percentage points. This suggests that performance voting may encompass economic as well as non-
economic performance evaluations and that their combined effect is larger than either that of economic and of
corruption performance voting alone.
Conclusion
Are elected officials held accountable for high levels of corruption? The evidence presented here suggests that
voters indeed sanction governing parties if they perceive high levels of corruption. Yet, this effect depends on both
individual voter characteristics and the electoral context. These findings might help us to understand why
corruption is a persistent phenomenon in some contexts but not in others. Regarding individual characteristics, we
showed that corruption voting depends on partisanship. It is non-partisans who make vote choices based on
perceived levels of corruption. Partisans who lean towards government or opposition parties have a strong
predisposition towards these parties. For such voters, performance evaluations are far less important for their vote
choice. This is in line with previous research showing varying effects of economic voting for partisans and non-
partisans (e.g. Kayser and Wlezien, 2011), suggesting that partisanship moderates the effect of voters' attitudes
on vote choice.
The CSES data from 20 elections in different national contexts also allow us to analyse the electoral context of
corruption performance voting. Corruption voting increases in a setting with high corruption, indicating that
performance voting depends on the saliency of the issue voters use to evaluate government parties. The
normative implications of this finding are twofold: First, it implies that voters facing high and persistent levels of
corruption indeed try to bring down these high levels by closely monitoring the incumbent parties. Second, it
means that corrupt political elites might escape electoral punishment as long as corruption is not among the most
salient issues.
In contrast to the economic dimension of performance voting, we find no evidence that clarity of responsibility
affects the magnitude of corruption performance voting. This negative finding, which is robust to various
approaches to measuring clarity of responsibility, illustrates that performance voting is a multi-faceted
phenomenon: while being a strong moderating variable for the influence of voters' economic perceptions on
incumbent support (Powell and Whitten, 1993; Whitten and Palmer, 1999; Anderson, 2007; Duch and Stevenson,
2008), the same does not hold for corruption perception voting. Perhaps the signal-extraction problem, fostered by
power dispersion, is more severe for economic policies than for corruption.
Economic policies partly entail positional competition, and these conflicts affect legislation on economic policies.
Thus, power dispersion leads to a signal extraction problem and makes economic performance voting less likely.
In contrast, fighting corruption involves less ideological conflict and is less dependent on legislative politics.
Fighting corruption is largely a matter of executive politics, the enforcement of existing rules and laws, and
effective control in the public administration. Moreover, corruption scandals are often highly visible and clearly
linked to names of parties and individual politicians, irrespective of the political and institutional context. While
this remains speculation, extending the theoretical framework of performance voting to explain these differences
is a crucial step towards a better understanding of individual voting behaviour.
This study focused on the influence of corruption perceptions on incumbent support, but there are several further
implications for other aspects of electoral politics. For example, the results presented here may have implications
for the emergence and electoral success of new parties (Lucardie, 2000; Tavits, 2006; Sikk, 2012). Following
Lucardie (2000: 182), 'new parties need a political project that caters to social problems perceived as important by
significant numbers of voters in order to win support'. In contexts where voters perceive high levels of corruption,
corruption can serve as a fruitful ground for new parties that run on an anti-corruption image (see also Sikk, 2012),
with sustainable effects if these parties can defend their distinct profile (Bolleyer and Bytzek, 2013). Furthermore,
corruption perceptions not only affect party choice but also turnout (Stockemer et al., 2013). Yet, there is no
consistent evidence whether turnout increases or decreases together with levels of corruption. The results
presented here suggest that these differences depend on whether voters believe that replacing the government
would make a difference. Those voters who believe in different outcomes under a government led by the current
opposition parties punish incumbents for high corruption, while we find no such effect for those who perceive few
differences between government alternatives. Thus, the effect of corruption perceptions on turnout may well be
conditional on citizens' beliefs about whether government turnover actually makes a difference (see also
The present study also encourages research on temporal effects on performance voting. We focus on corruption
performance voting in different national contexts, and data constraints do not allow us to include cross-temporal
variation in corruption performance voting as well. Nevertheless, we believe that time-series data linking corruption
perceptions and vote choice could reveal further insights. For example, voters may be most responsive to sharp
increases in perceived corruption and reward incumbents if they perceive decreasing levels of corruption.
Including these factors would improve our understanding of voter behaviour in general, and corruption
performance voting in particular.
Acknowledgements
We gratefully acknowledge financial support from the Austrian Science Fund (FWF) (grants P25490, S10902-G11,
and S10903-G11) and thank Hannah Kieblspeck for excellent research assistance. We are indebted to Michael
Lewis-Beck, Markus Wagner, Eva Zeglovits, and the participants of the seminar series at the Department of
Methods in the Social Sciences and the Department of Government, University of Vienna, for helpful comments
and suggestions on earlier drafts. We also thank the three anonymous reviewers and the editors at EPSR for their
support and helpful suggestions.
Footnote
1. Clearly, any voter directly benefiting from clientelist linkages may be somewhat ambivalent in her evaluation of
corruption (Kitschelt, 2000). However, not all clientelist practices are necessarily corrupt. Also, individual level
measures of corruption largely reflect voters' daily personal experiences with petty acts of corruption and are
generally detached from material advantages in clientelist elite-voter linkages. See Singer (2011c) on the complex
relationship between clientelism and perceived levels of corruption, especially grand corruption by political elites
used to generate funds to finance clientelist practices.
2. The list of moderating factors in Hypotheses 1-4 is of course not exhaustive. For example, we could also test
whether a voter's political information, and thus the probability to recognize corruption among politicians, affects
vote choice. As there are much better data for political information in national surveys, however, we refrain from
testing this information hypothesis (Winters and Weitz-Shapiro, 2013).
3. See Table A.1 in the Supplementary material for detailed information on the parliamentary elections studied.
4. Our empirical results are robust to two alternative specifications of the dependent variable, namely a model
separating voters for the prime minister's party (1) vs. any other party (0), and a multinomial logit model based on a
tripartite coding of respondents' vote choice (abstention (0), opposition vote (1), or incumbent vote (2)). The latter
model indicates that increasing corruption perception significantly reduces the odds of voting for the incumbent
(vs. voting for an opposition party), supporting the basic idea of retrospective voting models (see Supplementary
material, Model 6 in Table A.2 and Model 10 in Table A.3).
5. We use inversed scores so that higher scores indicate more perceived corruption, while lower scores indicate
less perceived corruption. Note that this measure does not capture differences across countries. In
methodological terms, Hypothesis 3 accounts for these different national contexts by weighting the corruption
perception effect with the severity of corruption in the election context. In other words, we use corruption severity
as an anchor (King et al., 2004) to achieve higher levels of cross-national comparability of survey items.
6. Including partisanship as a moderating variable also accounts for potential biases in our estimates of corruption
performance voting in general (Duch and Stevenson, 2008; Tilley et al., 2008): Voters who feel close to a political
References
References
Alesina A. and Weder B. (2002 ), 'Do corrupt gssovernments receive less foreign aid? ', American Economic Review
92 (4 ): 1126 -1137 .10.1257/00028280260344669
Anderson C.J. (2007 ), 'The end of economic voting? contingency dilemmas and the limits of democratic
AuthorAffiliation
1
Department of Government, University of Vienna, Austria
2
Department of Methods in the Social Sciences, University of Vienna, Austria
3
Department of Government, University of Vienna, Austria
DETAILS
Location: Austria
Company / organization: Name: International Bank for Reconstruction &Development--World Bank; NAICS:
928120; Name: University of Vienna; NAICS: 611310
Volume: 8
Issue: 3
Pages: 333-354
Number of pages: 22
ISSN: 17557739
DOI: http://dx.doi.org/10.1017/S1755773915000053
Copyright: © European Consortium for Political Research 2015 This is an Open Access article,
distributed under the terms of the Creative Commons Attribution licence
(http://creativecommons.org/licenses/by/3.0/), which permits unrestricted re-use,
distribution, and reproduction in any medium, provided the original work is properly
cited.
LINKS