ACIL RET Review Modelling - Full Report

A C I L A L L E N C O N S U L T I N G
REPORT TO
RET REVIEW EXPERT PANEL
7 AUGUST 2014
RET REVIEW
MODELLING
MARKET MODELLING OF
VARIOUS RET POLICY
OPTIONS
ACIL ALLEN CONSULTING PTY LTD
ABN 68 102 652 148
LEVEL FIFTEEN
127 CREEK STREET
BRISBANE QLD 4000
AUSTRALIA
T+61 7 3009 8700
F+61 7 3009 8799
LEVEL TWO
33 AINSLIE PLACE
CANBERRA ACT 2600
AUSTRALIA
T+61 2 6103 8200
F+61 2 6103 8233
LEVEL NINE
60 COLLINS STREET
MELBOURNE VIC 3000
AUSTRALIA
T+61 3 8650 6000
F+61 3 9654 6363
LEVEL ONE
50 PITT STREET
SYDNEY NSW 2000
AUSTRALIA
T+61 2 8272 5100
F+61 2 9247 2455
SUITE C2 CENTA BUILDING

118 RAILWAY STREET
WEST PERTH WA 6005
AUSTRALIA
T+61 8 9449 9600
F+61 8 9322 3955
ACILALLEN.COM.AU
RELIANCE AND DISCLAIMER

THE PROFESSIONAL ANALYSIS AND ADVICE IN THIS REPORT HAS BEEN PREPARED BY ACIL ALLEN CONSULTING FOR
THE EXCLUSIVE USE OF THE PARTY OR PARTIES TO WHOM IT IS ADDRESSED (THE ADDRESSEE) AND FOR THE
PURPOSES SPECIFIED IN IT. THIS REPORT IS SUPPLIED IN GOOD FAITH AND REFLECTS THE KNOWLEDGE,
EXPERTISE AND EXPERIENCE OF THE CONSULTANTS INVOLVED. THE REPORT MUST NOT BE PUBLISHED, QUOTED
OR DISSEMINATED TO ANY OTHER PARTY WITHOUT ACIL ALLEN CONSULTING’S PRIOR WRITTEN CONSENT. ACIL
ALLEN CONSULTING ACCEPTS NO RESPONSIBILITY WHATSOEVER FOR ANY LOSS OCCASIONED BY ANY PERSON
ACTING OR REFRAINING FROM ACTION AS A RESULT OF RELIANCE ON THE REPORT, OTHER THAN THE ADDRESSEE.
IN CONDUCTING THE ANALYSIS IN THIS REPORT ACIL ALLEN CONSULTING HAS ENDEAVOURED TO USE WHAT IT
CONSIDERS IS THE BEST INFORMATION AVAILABLE AT THE DATE OF PUBLICATION, INCLUDING INFORMATION
SUPPLIED BY THE ADDRESSEE. UNLESS STATED OTHERWISE, ACIL ALLEN CONSULTING DOES NOT WARRANT THE
ACCURACY OF ANY FORECAST OR PROJECTION IN THE REPORT. ALTHOUGH ACIL ALLEN CONSULTING EXERCISES
REASONABLE CARE WHEN MAKING FORECASTS OR PROJECTIONS, FACTORS IN THE PROCESS, SUCH AS FUTURE
MARKET BEHAVIOUR, ARE INHERENTLY UNCERTAIN AND CANNOT BE FORECAST OR PROJECTED RELIABLY.
ACIL ALLEN CONSULTING SHALL NOT BE LIABLE IN RESPECT OF ANY CLAIM ARISING OUT OF THE FAILURE OF A
CLIENT INVESTMENT TO PERFORM TO THE ADVANTAGE OF THE CLIENT OR TO THE ADVANTAGE OF THE CLIENT TO
THE DEGREE SUGGESTED OR ASSUMED IN ANY ADVICE OR FORECAST GIVEN BY ACIL ALLEN CONSULTING.
© ACIL ALLEN CONSULTING 2014

AC I L AL L E N C O N S U L T IN G
Executive summary
The Commonwealth Government has appointed an Expert Panel to conduct the 2014
review of the Renewable Energy Target (RET). The Expert Panel is supported by a
secretariat in the Department of the Prime Minister and Cabinet (the Secretariat).
The RET is comprised of two separate, but related schemes, namely: the Large-scale
Renewable Energy Target (LRET) and the Small-scale Renewable Energy Scheme (SRES).
ACIL Allen Consulting (ACIL Allen) has been engaged to undertake detailed electricity
market modelling of the RET impacts on Australia’s electricity markets and emissions from
electricity generation. The modelling and analysis is designed to support the Expert Panel’s
deliberations and inform the Government’s response to the Review.
Cases and sensitivities

ACIL Allen has been tasked with modelling a range of policy scenarios and sensitivities as
required by the Expert Panel. These are:
 Reference case: This case provides projections for the status quo where legislation
underpinning the LRET and SRES schemes remains unchanged and the market
develops in accordance with baseline assumptions in terms of demand and supply. All
subsequent policy scenarios are compared against this Reference case.
 Repeal case: This case assumes that the SRES and LRET schemes cease to operate
from 1 January 2015 with 2014 being the last compliance year. This scenario assumes
that any mechanism introduced to compensate investments made under the RET (if
any) does not affect wholesale or retail price outcomes.
 Closed to New Entrants: This scenario assumes that the LRET scheme continues to
operate, but is closed to new installations from 1 January 2015. The SRES, which
operates under a deeming arrangement whereby installations receive certificates
upfront, does not continue to operate and is closed from 1 January 2015. Under the
LRET, installations receive certificates annually based on generation. Closure of the
scheme to new entrants (new accredited generators) means that creation of LGCs is
limited to existing or committed generators.
 Real 20% case: The Real 20% scenario involves two significant changes to the
current policy: Reducing the LRET 2020 target to 25,500 GWh (a level which, when
evaluated using the Panel’s methodology, represents a ‘Real 20%’ of expected
demand in 2020); Closing the SRES after 2020 and reducing the period of deeming for
solar PV from 15 years down to 10 years from 1 January 2015 (deeming period is
constant at 10 years through to the end of 2020).
 Real 30% case: This scenario involves modifying the LRET target level to 30% of
anticipated demand in 2030 and extension of the scheme to 2040. Annual targets from
2015 to 2030 follow a linear trajectory, reaching 52,500 GWh and are held constant at
this level until 2040. There is no change to the current SRES, with the scheme
terminating in 2030.
 50% Growth case: This scenario was undertaken after the stakeholder workshop in
late June where preliminary modelling results were presented. It involves moving the
LRET away from fixed annual targets to floating targets with each year reset based on
forecast demand growth for the year ahead. The LRET target would be increased each
ii
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
year based on 50% of the anticipated growth in market-facing demand; i.e. demand
growth net of that absorbed by behind the meter solar PV. The scenario assumes
SRES modifications as follows:
 A reduction in deeming for Small Generating Units (SGUs) to 10 years from 1
January 2015, with the deeming period for both SGUs and Solar Water Heaters
(SWHs) declining by one year each year and the scheme terminating at the end of
2020
 A reduction in the maximum size eligibility of small generating units for inclusion
under SRES down from the current 100 kW to 20 kW (systems above 20 kW would
be eligible for the LRET).
Figure ES 1 summarises the LRET annual targets across the policy cases. In the 50%
Growth case, the LRET annual targets are a function of demand growth and therefore vary
across the demand sensitivities examined. The Real 30% case also includes an extension
of the scheme to 2040 with targets held constant at the 2030 level until 2040 (not shown in
the figure). Table ES 1 summarises the SRES settings under each policy scenario.
Figure ES 1 LRET annual targets under the various policy scenarios
60,000
50,000
40,000
GWh
30,000
20,000
10,000
0
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Reference case Real 20% case Real 30% case

50% Growth case 50% Growth case: Low Demand 50% Growth case: High demand
Note: Under all scenarios the LRET terminates in 2030 except for the Real 30% which extends out to 2040
Source: ACIL Allen based on input settings provided by the Expert Panel
iii
Table ES 1SRES settings under the various policy scenarios

Policy scenario Scheme end Treatment of SGU Treatment of SWH
15 years upfront, with deeming 10 years deeming upfront, with

Reference case End of calendar year 2030 period declining by 1 year each deeming period declining by 1
year from 2017 year each year from 2022
Repeal case 2014 last compliance year No further subsidies No further subsidies
Closed to new entrants case 2014 last compliance year No further subsidies No further subsidies
10 years deeming from 1

Real 20% case End of calendar year 2020 January 2015 (10 years available No change to Reference case
until scheme end)
Real 30% case End of calendar year 2030 No change to Reference case No change to Reference case
10 years from 1 January 2015, 10 years from 1 January 2015,

50% Growth case End of calendar year 2020 with the deeming period with the deeming period
declining by one year each year declining by one year each year
Source: ACIL Allen based on input settings provided by the Expert Panel
Input assumptions for the modelling have been sourced from a range of publicly available
sources including AEMO and IMO for demand and BREE for capital costs and learning
rates. These have been supplemented by ACIL Allen’s own in-house assumptions for other
key inputs. Sensitivities have also been completed to test the effects of changes for a
number of the key input assumptions where they are subject to considerable uncertainty.
These include high and low demand growth; the potential introduction of other abatement
policies modelled through a shadow carbon price from 2021; high capital costs for
renewable energy technologies; and permanent retirements for incumbent generators which
mothball capacity.
Analysis and findings

Currently renewable generation accounts for an estimated 16.1% of generation (at the end
of calendar year 2014). Under the Reference case where the RET remains unchanged,
renewable energy is projected to reach 26.3% by 2020 as shown in Figure ES 2.1
Under the Reference case assumptions, ACIL Allen’s modelling projects the renewable
energy target can be met by new renewable developments with the LRET fully subscribed
throughout the period to 2030. Much of the anticipated large-scale renewable development
occurs over the period 2016 to 2021, with around 7,650 MW of wind developed throughout
the NEM and SWIS regions and around 1,400 MW of utility-scale solar PV developed in the
regional grids of the NWIS, DKIS and Mt Isa. Owing to the subdued demand conditions in
electricity markets, the introduction of large volumes of renewable capacity results in a
mothballing of generating plant by incumbent operators, with much of this capacity returning
to service over time as demand grows.
1 This assessment has been undertaken using a formula provided by the Expert Panel and excludes the displacement from
solar water heaters (SWH). If displacement from SWH was to be added to both the renewable energy component (the
numerator), and to aggregate electricity demand (the denominator), aggregate renewables would be around one
percentage point higher at 27.3% by 2020.
iv
Figure ES 2 Proportion of renewables in Australia’s energy mix: Reference case
80,000 28%
26.3% 26.3% 26.3% 26.4% 26.4% 26.4% 26.4% 26.3% 26.3% 26.2% 26.1%
70,000 26%
23.6%
24%
60,000
Share of Australian energy

21.7%
22%
50,000 19.7%
GWh
20%
40,000 17.9%
17.0% 18%
16.1%
30,000
16%
20,000
14%
10,000 12%
0 10%
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Calendar year
Existing baseline generators SRES solar PV generation
Current LRET target Proportion of renewables
Note: Proportion of estimated total Australian electricity demand

Source: ACIL Allen
Across Australia, a total of $26.8 billion (real 2014 dollars) or $15.9 billion (in present value
terms) in capital expenditure on new generating capacity is projected to occur over the
period to 2040. Wind investment is projected to account for around 62% ($16.4 billion in real
2014 dollars or $12.1 billion in present value terms) of new large-scale generation
investment in the period to 2040. Gas-fired peaking plant and utility-scale solar PV each
account for around 11% of the total ($3 billion in real 2014 dollars). In present value terms,
solar accounts for $1.8 billion compared with $0.7 billion for peaking gas as adoption times
differ. Several categories of fossil fuel generation collectively account for the remaining 16%
($4.3 billion in real 2014 dollars or $1.3 billion in present value terms).
Small-scale systems (solar PV and solar water heaters) under the SRES are projected to
see strong growth with solar PV capacity rising from 4,133 MW at the end of 2014 to just
under 13,000 MW by 2030. Cumulative SWH installations are projected to increase from an
estimated 915,000 at the end of calendar year 2014 to over 1.5 million systems by 2030. A
total of $30.4 billion (real 2014 dollars) or $18 billion (present value terms) of new
investment is projected to occur in relation to solar PV and SWHs over the period to 2030.
The majority of this is solar PV ($20.6 billion in real 2014 dollars or $12.6 billion in present
value terms).
However, the subsidies paid to the renewable energy industry through the RET to bring
about this investment are high. Over the period to 2030, the projected total direct RET cost
(projected number of certificates multiplied by price) is $37.8 billion in real 2014 dollars
($22.4 billion in present value terms), of which, over 80% is associated with the LRET.
The modelling also shows that much of this additional capacity developed under the LRET is
surplus to market needs. Under the Repeal scenario, the modelling projects a net reduction
in the development of generating capacity of around 8,500 MW. Given the current levels of
oversupply in most electricity grids and muted demand growth, the existing generation fleet
is almost sufficient to meet expected demand for the foreseeable future.
v
Generation sector resource costs

Figure ES 3 below presents a summary of the present value of aggregate generation sector
resource costs over the period 2015-40 across each of the scenarios and sensitivities
modelled. This measure can be interpreted as the cost to society of generating electricity for
consumption by consumers and provides an indication of the sector’s labour and capital
productivity under each scenario and sensitivity when viewed on a per MWh basis.
For the Reference case (under core assumptions) costs total $121.9 billion in present value
terms over the period to 2040 using a discount rate of 7% pre-tax real. Under the core
assumptions, all of the policy variants examined resulted in a reduction in sector resource
costs, indicating capital and/or labour productivity gains for the economy. The Real 30%
scenario has almost the same aggregate cost as the Reference case because the deferral
of wind development early is offset by an overall larger amount of renewable development in
the longer-term.
The Repeal case has the lowest projected resource costs, as expected, as there are no
RET subsidies distorting supply costs and competitive wholesale electricity markets are left
to determine the most efficient, least cost plant mix to meet demand. This was one of the
fundamental intentions in the establishment of the NEM, with its rules and principles being
deliberately technology agnostic. Another reason for the development of the NEM was to
impose competitive disciplines on participants in order to avoid the large oversupply in
generation that had occurred through state governments using electricity supply to support
other industries and policies. In a market with little or no demand growth, the RET is creating
the same oversupply in generation that the NEM was designed to correct. In the absence of
the RET policy, the market determines the optimal level of generation investment, rather
than having arbitrary targets imposed upon it. In a market environment where capacity is
already oversupplied and demand may continue to decline, it is desirable (and efficient) for
no new investment in capacity to occur.
Figure ES 3 Aggregate generation sector resource costs (NPV 2015-2040): All scenarios/sensitivities
132.9
138.1
126.7
123.8
123.2
122.8
121.9
121.8
160
112.7
112.5
111.8
111.4
108.9
108.6
108.2
108.0
107.9
106.4
106.0
140
96.0
90.8
90.7
Real 2014 $ billion
120
100
80
60
40
20
0
Real 20%
Real 30%
Real 20%
Real 30%
Repeal
Repeal
Repeal
Repeal
Repeal
Closed to New Entrants

50% Growth
50% Growth
50% Growth
50% Growth
50% Growth
Reference
Reference
Reference
Reference
Reference
Reference
Core Carbon High capital costs High demand Low demand Permanent
retirements
Note: Measure includes capital expenditure (on both generating capacity and any interconnector expansions/augmentations); refurbish ment of
existing and new generators for life extension beyond initial economic life; fixed operating costs (fixed costs associated with normal operation
and stay in business capital expenditure associated with existing and new generating capacity); variable operating costs (fuel costs and variable
O&M costs for existing and new generation) and unserved energy. NPV calculated using a 7% real discount rate.
Source: ACIL Allen
vi
Emissions and cost of abatement

The RET policy delivers emissions abatement through displacing fossil fuel based
generation with renewable generation. The level of abatement achieved is projected to be
higher under the current market conditions relative to previous assessments because of the
reduced role of gas-fired plant (increasing gas prices) and the repeal of the carbon price,
both of which increase the competitiveness of coal fired plant within the generation mix.
Figure ES 4 below shows that the policy scenarios which include the RET or an expansion
to the RET (the Real 30% case) consistently result in the lowest emissions outcomes across
assumption sets. Conversely the Repeal of the RET is projected to lead to higher emissions;
between 8% and 14% relative to the Reference case over the period to 2040.
Figure ES 4 Aggregate emissions from electricity generation: 2015-2040: All scenarios/sensitivities
5,784
5,481
5,345
7000
5,117
5,114
5,114
5,072
5,072
4,917
4,917
4,909
4,859
4,594
4,593
4,557
4,525
4,494
6000
3,973
3,952
3,781
3,463
3,393
5000
4000
Mt CO2-e
3000
2000
1000
0
Real 30%
Real 20%
Real 30%
Real 20%
Repeal
Repeal
Repeal
Repeal
Repeal

Reference
50% Growth
Reference
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
retirements
Note: Excludes non-scheduled generation in NEM regions, own-generation in the SWIS and off-grid generation ‘Other’ category includes
cogeneration, liquid fuels, CCS-equipped technologies, biomass and geothermal.
Source: ACIL Allen
As with any Government expenditure or program, an important consideration is whether the

policy offers value for money relative to alternatives. Two methods for calculating the cost of
abatement from the policy have been used. Method 1 calculates abatement costs as the
present value of the change in resource costs divided by the discounted change in
abatement. Method 2 is the same, except the emissions in the denominator are not
discounted.2 Using method 1, estimated abatement costs for the RET range from $59/tonne
under the 50% growth case (core assumptions) to $77/tonne (Reference case High capital
costs).3 Under method 2, these costs are lower, ranging from $30/tonne to $40/tonne under
the same scenarios. Whilst the policy is somewhat effective in the abatement of emissions,
it is at high cost compared to current global pricing and is therefore not the most efficient
means of emissions abatement. There is also a large difference between calculated
abatement costs for the LRET and SRES components, with the abatement costs for the
SRES being at least 2 to 3 times higher than the LRET. Therefore policy scenarios which
tend to reduce subsidies provided for solar PV will tend to lower the overall RET abatement
cost.
2 ACIL Allen has used the second method to calculate abatement costs at the request of the Expert Panel. However, ACIL
Allen considers that the second method does not appropriately reflect the costs of emissions abatement on an inter-
temporal basis.
3 This range excludes the 50% Growth low demand case which is an outlier with a much higher cost of $164/tonne.
vii
Impacts on retail prices

The public analysis of the costs and benefits of the RET scheme has been dominated by
views on the net benefits that the RET scheme provides to electricity consumers. These net
electricity consumer benefits are generally calculated by assessing wholesale electricity
price (pool) and RET certificate price changes for the market with and without the RET
scheme (i.e. the modelled impact of the subsidised renewable generation on wholesale
electricity market prices). Assessing the net consumer benefits limited to a specific
economic sector cannot be considered to be either a social cost benefit analysis or an
economy wide assessment of the RET scheme. Considering only the benefits that flow to
consumers ignores the opportunity costs of the capital and labour involved and the other
welfare effects of the policy.
Figure ES 5 below shows the projected aggregate cost for an average Australian household
on electricity over the period 2015-40 in NPV terms. In most cases, moving from the
Reference case to the Repeal case (the most extreme policy variant) results in projected
household electricity costs rising in net terms (the reduction in direct compliance costs is
outweighed by increases in wholesale electricity prices). This indicates that wealth transfers
are occurring from existing generators to both new renewable energy projects and
consumers.
Interestingly, this pattern of price changes does not hold under low demand conditions. This
is because new renewable generation is incapable of supressing wholesale prices below
levels which are sustainable for incumbent generators to keep operating. Under these
conditions, removal of the direct compliance costs is not offset by any increases in
wholesale prices and consumers are better off under a Repeal scenario.
The impact on retail electricity prices is subject to uncertainty in the modelled components.
Pool prices are inherently uncertain. This is because many of the drivers of pool prices are
uncertain, such as:
 Weather driving demand is unpredictable and highly variable
 Plant performance (outages) is also stochastic (random)
 Fuel prices may vary over time although most fossil fuel fired plant tend to contract over
several years and so these prices tend to be reasonably certain on an annual basis
 Participant behaviour (mothballing, plant retirement, strategic bidding, etc.) may swamp
other effects over time.
A key factor in the uncertainty around future electricity prices is participant behaviour. As
electricity demand has fallen in recent years, an increasing willingness of participants to
mothball or close generation plant has been observed. Closing or mothballing plant can
cause a significant rebound in pool prices and may fully offset any downward pressure from
renewable plant. While we have incorporated some mothballing of plant in the analysis,
participants may have different objectives and take quite different views to mothballing and
plant closure than we have taken. This could substantially change the net benefit to
electricity consumers through net changes in retail prices. The Permanent retirement
sensitivity, which includes a larger amount of incumbent capacity withdrawal, demonstrates
that the impact on retail electricity prices can easily be reversed, with consumers benefiting
from moving from the Reference case policy to a 50% Growth scenario. Directionally, the
same outcome would also be seen if the scheme was closed to new entrants or fully
repealed.
viii
Figure ES 5 NPV of average household total expenditure on electricity (2015-2040): All

scenarios/sensitivities
20,260
20,451
20,074
20,037
21,000
19,887
19,771
19,754
19,634
19,464
19,453
20,500
19,358
19,349
19,305
19,213
19,193
19,181
19,182
19,119
19,119
19,092
19,003
20,000
NPV of retail bills ($)
18,706
19,500
19,000
18,500
18,000
17,500
Real 30%
Real 20%
Real 30%
Real 20%
Repeal
Repeal
Repeal
Repeal
Repeal
50% Growth
50% Growth
50% Growth
50% Growth
50% Growth
Reference
Reference
Reference
Reference
Reference
Reference
retirements
Note: NPV of annual residential bills for average household over the period 2015-40. Uses a 7% real discount rate
Source: ACIL Allen
Regardless of direction, the impact on retail electricity prices is small, even when considered
over the period to 2040. Under the core assumptions, moving from the Reference case to a
complete repeal of the scheme is projected to increase a typical household’s expenditure on
electricity over the period to 2040 by 0.6% in present value terms. By comparison, moving
from the Reference case to a Repeal case under low demand conditions is projected to
reduce a typical household’s expenditure on electricity (over the same period) by 2.1% in
present value terms. In all cases examined, the benefits or costs are a very small
percentage of the total electricity bill and could easily be swamped by the range of
uncertainties in pool prices, especially the changes in the behaviour of generation
participants.
Assessing the RET’s impacts on retail electricity prices in isolation does not provide a solid
basis for economically evaluating the RET policy. That the RET may lower electricity prices
for consumers does not mean that its benefits outweigh its costs when considered in society
wide terms. The diversion of capital and labour from other productive activities to the
electricity sector imposes real costs on other sectors of the economy. Other policies such as
subsidising fossil fuels or fossil fuel generators would also likely have the effect of lowering
costs to electricity consumers and probably at significantly lower resource costs, yet few
would advocate these as being good policy positions. An economic evaluation of the policy
would not normally include wealth transfers where either producers or consumers benefit at
the expense of each other. This makes projected changes to retail electricity prices mostly
irrelevant in any economic assessment of the policy.
In ACIL Allen’s view, the main focus of any evaluation should be on the cost of abatement
achieved through the policy and whether this represents an efficient means of achieving
abatement objectives.
ix
Executive summary ii
1 Introduction and background 1

1.1 Approach to the modelling 1
1.2 Overview of scenarios examined 2
1.3 Scope of modelling 4
1.4 Report structure 5
2 Reference case 7
2.1 Scenario description 7
2.2 Results 9
2.2.1 RET outcomes 14
2.2.2 Resource costs, profitability and emissions 19
2.2.3 Retail price outcomes 22
3 Repeal case 25
3.2 Results 25
4 Closed to new entrants 33

4.2 Results 35
5 Real 20% by 2020 38

5.2 Results 40
6 Real 30% by 2030 51


6.2 Results 52
7 50% of growth case 64

7.2 Results 65
8 Sensitivities 76
8.1 Low demand 76
8.1.1 Reference case results 77
8.1.2 Other policy cases 80
8.2 High demand 87
8.3 Shadow carbon price 93
8.4 High capital costs 99
8.5 Permanent retirements 106
9 Abatement costs 113

9.1.1 Methodology for calculation 113
9.1.2 Abatement costs across policy cases to 2030 114
9.1.3 Abatement costs across policy cases to 2040 117
10 Summary and conclusions 120

10.1 Summary of modelling outcomes 120
10.1.1 Electricity generation sector resource costs 120
10.1.2 Greenhouse gas emissions and abatement costs 121
10.1.3 Investment in renewable generation 123
10.1.4 RET compliance costs 125

10.1.5 Residential retail prices 127
10.1.6 Assessing net electricity consumer benefits 128
Appendix A Methodology and assumptions A1

Glossary
Acronym or term Explanation
Australian Energy Market Operator, the entity that manages dispatch and planning in the National
AEMO
Electricity Market.
Australian Energy Technology Assessment, an analysis of future generation costs from various
AETA
electricity supply technologies undertaken by BREE in 2012 and updated in 2013.
BREE Bureau of Resources and Energy Economics, a Commonwealth Government research agency.
Combined-cycle gas turbine, a gas turbine generator where waste heat from the turbine exhaust is
CCGT
captured and used to drive an auxiliary steam turbine.
Carbon capture and storage, the capturing of carbon dioxide produced in the process of generating
CCS
electricity (or some other industrial process) and storing
Computable General Equilibrium modelling, a form of modelling that relates the inputs and outputs of
CGE different industries within an economy to determine a ‘general equilibrium’ outcome across all
industries when inputs or assumptions are varied.
CLFR Concentrated Linear Fresnel Reflector, a form of solar thermal generation technology.
A cogeneration plant generates both electricity and steam, with the steam typically being used for
industrial process applications. Cogeneration plants can be based on either a typical steam turbine,
Cogeneration, or ‘cogen’ with lower pressure steam being diverted for use as heat rather than for electricity generation, or on a
gas turbine, where the gas turbine itself generates electricity but waste heat is captured to generate
steam for use as process heat.
CO2 Carbon dioxide, the most common greenhouse gas
CO2CRC The Cooperative Research Centre for Carbon Capture and Storage.
Darwin-Katherine Interconnected System, the interconnected electricity grid servicing the main
DKIS
population centres of the northern part of the Northern Territory.
DUOS Distribution use of service: charges for transmitting electricity through distribution networks
Engineered geothermal system, a form of geothermal generation technology also sometimes known
EGS
as ‘hot fractured rocks’.
Fixed operating and maintenance costs. These are represented in ACIL Allen’s modelling as a fixed
FOM
annual payment required to keep a power station operational.
HEGT High efficiency gas turbine.
HHV Higher heating value
HSA Hot sedimentary aquifer, a form of geothermal generation technology.
Integrated gasification combined cycle, a form of generation technology that uses coal as the fuel, and
IGCC
which converts the coal to a synthetic gas to drive a gas turbine through an integrated process.
Independent Market Operator, the entity that manages dispatch and planning in the South-West
IMO
Interconnected System.
Load duration curve, a representation of the variation in electricity demand over a period of time
LDC
created by ordering the electricity demand (or ‘load’) in descending order.
LP Linear programming
Large-scale Renewable Energy Target, the Commonwealth Government’s scheme to promote large-
LRET scale renewable electricity generation. Formerly known as the Mandatory Renewable Energy Target
(MRET), and sometimes referred to simply as the RET.
Marginal loss factor, the level of transmission losses between a given generator and the point of
MLF market settlement attributed in dispatching bids for electricity supply and therefore in calculating
electricity prices.
MW Megawatt, a unit of (instantaneous) electricity output or consumption
Megawatt-hour, a unit of electricity output or consumption measured over time, which is equivalent to
MWh
one megawatt being produced/consumed continuously for one hour.
National Electricity Market, the interconnected electricity grid covering most of Queensland, New
NEM
South Wales, Victoria, Tasmania and South Australia.
NSLP Net system load profile
North-West Interconnected System, the interconnected electricity grid covering the Pilbara region of
NWIS
north-western Western Australia.
Open cycle gas turbine, a gas turbine generator where waste heat is vented to the atmosphere rather
OCGT than captured to generate electricity or steam, as in a combined-cycle gas turbine (CCGT) or
cogeneration plant.
PC Pulverised coal. See also ‘pf’
pf Pulverised fuel, typically coal. See also ‘PC’.
Photovoltaic, a form of generation that converts solar radiation to direct current electricity using semi-
PV
conductors that exhibit the photovoltaic effect.
Acronym or term Explanation
Solar Flagships, a Commonwealth Government program to promote large-scale solar generation

SF
projects.
Short-Run Marginal Cost, an economic interpretation of the extent to which production costs, in this
SRMC case electricity generation costs, vary at the margin when key inputs, particularly the capital
equipment comprising the generator, cannot be varied.
SWCJV South-West Cogeneration Joint Venture
South-West Interconnected System, the interconnected electricity grid covering south-western
SWIS
Western Australia. Also known as the Wholesale Electricity Market, or WEM.
TUOS Transmission use of service: charges for transmitting electricity through transmission networks
Variable operating and maintenance costs. These are represented in ACIL Allen’s modelling as costs
VOM which vary linearly with the amount of electricity produced by a given power station (i.e. as a cost in
$/MWh).
1 Introduction and background

The Commonwealth Government has appointed an Expert Panel to conduct the 2014
review of the Renewable Energy Target (RET). The Expert Panel is supported by a
secretariat in the Department of the Prime Minister and Cabinet (the Secretariat).
To assist with this review, ACIL Allen Consulting (ACIL Allen) has been engaged to
undertake detailed electricity market modelling of the impacts of the RET. The modelling
and analysis will support the Expert Panel’s deliberations and inform the Government’s
response to the Review.
This report represents the results of ACIL Allen’s modelling of the scheme and the key
deliverable under the engagement.
1.1 Approach to the modelling

Figure 1 provides an overview of the modelling process used to undertake the projections
for the Expert Panel under this project. The modelling involves the use of three key models
within ACIL Allen’s energy market modelling suite:
 PowerMark LT: a long-term dynamic least cost planning model which models all
electricity grids simultaneously as well as Australia-wide policy settings
 SRES projection model: projects uptake of small-scale PV systems based on historical
relationships of installs against system financial paybacks
 Retail price model: bottom-up build of retail cost components: network (TUOS &
DUOS), wholesale energy, LRET, SRES, losses, green scheme costs, metering
charges, retail operating costs, pool fees, ancillary service costs, retail margin.
Figure 1 Modelling framework
Market characteristics
Off-grid database
- existing generator inputs
- new entrant options and costs - detailed breakdown of existing
SRES model generation ~ 10,000 GWh
Coverage:
- projected SWH - current renewable composition
NEM
- projected SGU (solar PV) approximately 2%
SWIS
DKIS
NWIS and Mt Isa
Demand inputs LRET settings

- market facing sent-out energy - targets and banking rules
and peak demand - existing generators
- 100 point sampled load PowerMark LT - new entrant generators
duration curves - certificate multipliers per MWh
Model outputs
- Emissions (CO2-e)
- Generation/capacity mix Retail electricity pricing model
- Fuel use - network costs
- Wholesale electricity prices - retail costs and margin
- LGC prices - hedging costs (based on load shape)
- LRET market surrenders/shortfalls - retail series for households
- New entry technology and build timing - retail series for SMEs/large users
- Retirements and refurbishments
- System resource costs
Source: ACIL Allen
1
A more detailed description of the modelling approach and key assumptions used is
presented in Appendix A.
1.2 Overview of scenarios examined

ACIL Allen has been tasked with modelling a range of policy scenarios and sensitivities as
required by the Expert Panel. These are:
 Reference case: this case provides projections for the status quo where legislation
underpinning the LRET and SRES schemes remains unchanged and the market
develops in accordance with baseline assumptions in terms of demand and supply. All
subsequent policy scenarios are compared against this Reference case.
 Repeal case: Under the Repeal case it is assumed that the SRES and LRET schemes
cease to operate from 1 January 2015 with 2014 being the last compliance year. This
scenario assumes that any mechanism introduced to compensate investments made
under the RET (if any) does not affect wholesale or retail price outcomes.
 Closed to New Entrants case This scenario assumes that the LRET scheme
continues to operate, but is closed to new installations from 1 January 2015. The
SRES, which operates under a deeming arrangement whereby installations receive
certificates upfront, does not continue to operate and is closed from 1 January 2015.
Under the LRET, installations receive certificates annually based on generation.
Closure of the scheme to new entrants (no new accredited generators) would result in
the supply of LGCs being limited to existing or committed generators with annual LRET
targets adjusted accordingly.
 Real 20% case: The Real 20% scenario involves two significant changes to the
current policy: Reducing the LRET 2020 target to 25,500 GWh (a level which, when
evaluated using the Panel’s methodology, represents a ‘real20%’ of expected demand
in this year); Closing the SRES after 2020 and reducing the period of deeming for solar
PV from 15 years down to 10 years from 1 January 2015 (deeming period is constant
at 10 years through to the end of 2020).
 Real 30% case: This scenario involves modifying the LRET target level to 30% of
anticipated demand in 2030 and extension of the scheme to 2040. Annual targets from
2015 to 2030 follow a linear trajectory, reaching 52,500 GWh and are held constant at
this level until 2040. No change to the current SRES is made, with the scheme
terminating in 2030.
 50% Growth case: This scenario was undertaken after the stakeholder workshop in
late June where preliminary modelling results were presented. It involves moving the
LRET away from fixed annual targets to floating targets, reset each year based on
forecast demand growth for the year ahead. Specifically, the LRET target would be
increased each year to 2020 based on 50% of the anticipated growth in market-facing
demand over the year — that is, demand growth net of that absorbed by behind the
meter solar PV.4 The LRET would terminate in 2030. The scenario also implements
modifications to the SRES as follows:
 A reduction in deeming for SGUs to 10 years from 1 January 2015, with the deeming
period for both SGUs and SWHs declining by one year each year and the scheme
terminating at the end of 2020.
4 Under this scenario LRET targets are not reduced if demand is expected to decline. That is the LRET target for the following
year is the maximum of the previous year’s target or a base year’s target plus 50% of the growth in energy (net of solar PV
energy increase).
2
 A reduction in the size eligibility of small generating units for inclusion under SRES
down from the current 100 kW to 20 kW (systems above 20 kW would be eligible for
the LRET).
Sensitivities have also been completed testing the impact of a number of the key input
assumptions which are subject to considerable uncertainty. These sensitivities examine a
change to one input variable holding all others constant as per the base assumption set.
Sensitivities examined include:
 Low demand growth: use of the AEMO and IMO low economic growth energy and
demand forecasts, whilst keeping load growth in regional grids and off-grid sources
flat. Under this scenario, policy cases which involve a present day recalibration of
LRET targets (Real 20% and Real 30% cases) are maintained based on base demand
assumptions as these represent the current expected outcome. This sensitivity tests
the impact of demand outcomes being well below current expectations.
 High demand growth: Like the 50 per cent growth scenario, this sensitivity analysis
was undertaken following the stakeholder workshop held in June. It used the AEMO
and IMO high economic growth energy and demand forecasts with a slight increase to
off-grid and regional grids relative to base assumptions. Under this scenario, policy
cases which involve a present day recalibration of LRET targets (Real 20% and Real
30% cases) are maintained based on base demand assumptions as these represent
the current expected outcome. This sensitivity tests the impact of demand outcomes
being well above current expectations.
 Shadow carbon price: Current Government policy includes repealing the carbon tax
legislation and meeting Australia’s 2020 emissions reduction target of a 5 per cent cut
through other means. To understand the effect of policies that could potentially be
introduced after 2020 to meet future emissions reduction goals, a sensitivity has been
modelled which introduces a shadow carbon price commencing 2021 with pricing
based on a projection of current European emission permit prices.
 High capital costs: This sensitivity analysis was undertaken following the stakeholder
workshop held in June. It examines the impact of much slower cost reductions in costs
for renewable technologies, specifically solar and wind which are expected to the
primary technologies deployed under the current RET
 Permanent retirements: This sensitivity analysis was undertaken following the
stakeholder workshop held in June. In response to new renewable capacity entering
under the RET it is expected that some incumbent fossil fuelled plant will withdraw
capacity from the market due to market revenues not covering avoidable costs. This
sensitivity tests the possibility of more capacity being withdrawn (relative to the
Reference case) and for that capacity not to be reintroduced to the market in later
years based on engineering and technical constraints around long-term mothballing.
Figure 2 provides a matrix of the policy cases and assumption sets which have been
undertaken. Due to time constraints, not all potential combinations were run.
3
Figure 2 Scenario / sensitivity matrix
1.3 Scope of modelling

In this exercise, ACIL Allen has been engaged to undertake sectoral modelling of the impact
of the RET upon electricity sector outcomes. This includes detailed analysis of the impacts
of the RET and how the RET may influence Australia’s electricity mix, electricity prices,
emissions and some of the direct costs associated with those impacts. This is done under
the range of scenarios and sensitivities outlined above.
Extending this analysis through to the broader economy requires different but
complementary types of modelling and analysis to be undertaken. The economic impacts of
a government policy/program are usually assessed using one or more of the following
economic analysis tools:
 Social Benefit-Cost Analysis (SBCA)
 Computable General Equilibrium (CGE) modelling
 Input-Output (I-O) analysis.
Importantly, additional spending or direct employment associated with a policy/program will
not always have a positive economic impact on the local, state or national economy.
Although various aspects of a policy/program – such as the number of jobs or the size of the
investment expenditure – are of relevance to particular stakeholders, the key aggregate
measure of the macroeconomic impact of a policy/program is the extent to which the total
income of the economy has changed as a result of the policy/program. Typically this is
measured by real gross national disposable income (RGNDI), although real gross domestic
product (GDP) and consumer surplus (among others) can also be important aggregate
measures depending on the nature of the policy/program being analysed.
The main factors that need to be considered when analysing the macroeconomic impacts of
a policy/program include:
 The direct and indirect contribution to the economy as a result of the activities
associated with the policy/program
 Any crowding out implications as resources are potentially diverted from other productive
activities to undertake the policy/program being analysed
 Any productivity effects generated as a direct result of the policy/program activities –
particularly any enduring productivity changes or productivity impacts on other activities
not directly associated with the policy/program
 Any changes to the factors of production in the economy
 Any implications associated with changes in terms of trade or foreign income transfers
 The extent of any dynamic aspect of any of the above effects (for example, associated
with different phases of the policy/program).
4
Figure 3 shows these components functionally. Some of these effects may be negligible
while others may be very significant. An understanding of the effects helps determine the
most appropriate tool(s) for the analysis.
Figure 3 Estimating the macroeconomic impact of a policy/program
Total economic impact

(real income/RGNDI)
Economic output impact Terms of Foreign income

(GDP) trade effects transfer effects
Direct Indirect
economic economic
contribution contribution
Factors of
Supply chain production
impacts impacts
Taxes less
Value added subsidies
Crowding out Productivity
impacts impacts
Source: ACIL Allen
ACIL Allen has not been tasked with estimating the broader economic impacts of the policy
and therefore this report does not make any conclusions regarding the RET in this regard.
We understand that the Review is drawing on information from other sources through the
consultation process.
1.4 Report structure

This report is ACIL Allen’s primary deliverable under this assignment for the Expert Panel. It
provides an overview of the methodology, key input assumptions and key results from each
of the policy cases and sensitivities examined.
The remainder of the report is structured as follows:
 Chapter 2 presents modelling results from the Reference case – that is, the outlook for
the relevant markets should there be no change to the status quo legislation and the
markets developed based on mid-line (core) assumptions in relation to demand and
costs. The chapter discusses projected key trends in the electricity markets both at a
wholesale and retail level under the core assumptions used.
 Chapters 3 through 7 examine each of the alternate policy cases, providing a more
detailed scenario description and presenting results from the modelling. Results are
presented both on an absolute basis and relative to the Reference case.
 Chapter 8 provides the results of the market modelling for each of the sensitivity cases,
comparing the results against the Reference case and the Core scenario from which
the sensitivity was derived.
 Chapter 9 provides the estimated cost of abatement from the LRET/SRES under each
of the scenarios/sensitivities
 Chapter 10 provides a summary of the results across the scenarios/sensitivities and
discusses conclusions which can be drawn from the market modelling.
5
 Finally, Appendix A provides a more detailed description of the methodology employed

in the analysis and key assumptions.
It should be noted that the results herein are projections based on the modelling
methodologies employed and the assumptions used as set out in Appendix A. Many of the
elements projected are subject to significant uncertainty, whilst the models operate under
the assumptions of full information and perfect certainty.
6
2 Reference case
2.1 Scenario description

This case provides projections for the status quo where legislation underpinning the LRET
and SRES remains unchanged and the market develops in accordance with baseline
assumptions in terms of demand and supply.
Figure 4 provides the estimated proportion that renewable generation would account for in
Australia’s energy mix under Reference case assumptions provided the LRET is fully met.
Currently renewable generation accounts for an estimated 16.1% (at the end of calendar
year 2014), with this projected to rise to 26.3% by 2020. The three components to
renewable generation considered in this calculation are:
 Estimated long-term average below baseline output from pre-existing generators
(approximately 16,150 GWh). This is slightly less than thecurrent aggregate baseline
of around 16,600 GWh and is held constant into the future5
 Projected contribution from behind the meter solar PV under SRES which rises from
5,140 GWh in 2014 to 9,920 GWh by 2020
 The current mandated LRET annual targets which increase from 16,100 GWh in 2014
to 41,000 GWh in 2020 (excluding WCMG volumes).
This calculation excludes voluntary renewable generation, for example energy associated
with the GreenPower scheme or energy associated with desalination plants. LGCs created
from renewable energy generators for the purpose of supplying these loads are voluntarily
surrendered to the Clean Energy Regulator (CER) and are therefore incremental to
mandated volumes through the LRET.
The denominator in this calculation is ACIL Allen’s estimate of total Australian electricity
demand including native energy for major grids; energy sourced from embedded generators
or through self-generation and off-grid energy. A breakdown of these demand components
is provided in Table 1.
ACIL Allen estimate total electricity demand as being 232.6 TWh in 2014, with this projected
to rise to 255.2 TWh by 2020 (equating to a cumulative average growth rate of 1.6%). Of the
22.6 TWh of projected growth, 9.5 TWh is expected from the NEM, with 4.7 TWh from
increased output from behind the meter (rooftop) solar PV and 4.2 TWh from off-grid
sources (increases include self-generation at new LNG developments on Curtis Island QLD
and offshore WA).
5 Below-baseline generation by large-scale hydro and biomass-based generators is inherently uncertain. This figure is around
1,000 to 2,000 GWh higher than some previous estimates, which have been based on historical data. See Appendix
section A.1.8 for more detail on how this level was determined.
7
Figure 4 Proportion of renewables in Australia’s energy mix: Reference case
80,000 28%
26.3% 26.3% 26.3% 26.4% 26.4% 26.4% 26.4% 26.3% 26.3% 26.2% 26.1%
70,000 26%
23.6%
24%
60,000
Share of Australian energy

21.7%
22%
50,000 19.7%
GWh
20%
40,000 17.9%
17.0% 18%
16.1%
30,000
16%
20,000
14%
10,000 12%
0 10%
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Calendar year
Existing baseline generators SRES solar PV generation
Current LRET target Proportion of renewables
Note: Proportion of estimated total Australian electricity demand

Source: ACIL Allen
We note that previous assessments of the proportion of renewables in Australia’s electricity

mix have also considered displacement through solar water heaters as part of the
renewable energy component.6 Under the Reference case, aggregate displacement from
SWHs is estimated to be 2,870 GWh in 2014, rising to just under 3,500 GWh by 2020. If
displacement from SWH was to be added to both the renewable energy component (the
numerator), and to aggregate electricity demand (the denominator), aggregate renewables
would be around one percentage point higher at 27.3% by 2020.
Table 1 Projected energy by grid: Reference case
Grid / demand segment Annual energy (TWh) Average annual growth rate
2014 2020 2040 2014-20 2014-40

NEM 181.1 190.6 204.2 0.9% 0.5%
SWIS 18.5 20.7 29.1 1.9% 1.8%
NWIS 2.9 3.6 4.9 3.6% 2.0%
DKIS 1.6 1.8 2.9 2.0% 2.2%
Mt Isa 2.3 2.5 4.0 1.6% 2.2%
Rooftop Solar PV 5.0 9.8 22.0 11.7% 5.8%
Embedded/self-gen 10.3 11.1 13.8 1.4% 1.1%
Off-grid 10.9 15.1 20.4 5.6% 2.5%
232.6 255.3 301.3 1.6% 1.0%
Note: Energy values exclude estimated displacement from SWH.
Source: ACIL Allen based on forecast data from AEMO, IMO and other sources
The Reference case outcomes which are detailed in the following sections provide the base
scenario against which outcomes from alternate policy cases are compared in subsequent
chapters.
6 ACIL Allen has been instructed by the Expert Panel to exclude SWH for modelling purposes.
8
2.2 Results
Figure 5 shows the wholesale price outcomes from the Reference case. Results for
individual regions have been combined into two price traces:
 NEM: comprising of a weighted average price outcome for NSW, Queensland, South
Australia, Tasmania and Victoria.
 Other grids: comprising of a weighted average price outcome for the SWIS, NWIS,
DKIS and Mt Isa.
NEM regions commence with prices around $44/MWh in calendar year 2014 and fall to
mostly below $33/MWh in 2015 due to it being the first full year without carbon pricing.
Prices rise slightly through 2016 and 2017, influenced by additional demand in Queensland
and reduced output from gas-fired generation. Through 2017 to 2020 significant amounts of
new wind capacity enter the market driven by the LRET and this tends to hold prices at an
average of around $30/MWh until around 2025. Some incumbent capacity is mothballed late
in the decade due to low profitability as observed within our simulation model. Capacity
withdrawal is required to accommodate the additional wind entry and to increase wholesale
prices to a sustainable level for incumbent plant operators.
Prices begin to slowly rise from 2025 onwards as demand growth has largely absorbed the
additional renewable capacity and mothballed plant is reintroduced to service. Prices in the
NEM remain well below levels required for new baseload/intermediate fossil fuel plant entry
in the period to 2040.
Prices in the ‘Other grids’ start at a higher level than NEM regions, driven by higher fuel
prices and relatively poor load factors. Prices decline to 2021 as a result of wind entry being
greater than load growth during this period. Prices do trend upward earlier which is a result
of the higher demand growth forecast — particularly the SWIS. Prices remain relatively flat
in real terms after 2030. Significant volumes of utility-scale solar PV are projected to be
developed in regional grids under the Reference case, driven by the underlying energy
prices and high LGC prices which prevail under this scenario.
Figure 5 Wholesale price outcomes: Reference case

100
90
80
Real 2014 $/MWh
70
60
50
40
30
20
NEM Other grids
10
0
2023
2038
2014
2015
2016
2017
2018
2019
2020
2021
2022
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2039
2040
Note: Prices presented on an equivalent basis to time-weighted annual averages. Weighted average
prices for grids (NEM = NSW, QLD, SA, TAS and VIC; Other grids = SWIS, NWIS, DKIS and Mt Isa).
Source: ACIL Allen
Despite the influences of the RET, Australia’s generation mix doesn’t change dramatically
over the period to 2040 as shown in Figure 6. Wind entry over the period 2016-2020 is
significant and displaces primarily black coal generation. Once the wind build necessary to
meet the LRET target is completed however, the future fuel mix is relatively static
throughout the remainder of the modelling horizon, with most growth met by increased
output from existing coal-fired stations. Additional large-scale renewable generation (beyond
9
that required to meet the LRET) is not projected to enter the market by 2040 as costs, whilst
falling, remain above price levels in the market. A key limiting factor is the dispatch-weighted
prices available to both wind and solar technologies which are both well below average
time-weighted prices shown above. This is due to the output from these technologies
occurring at times of low pool prices (wind generally skewed toward overnight periods and
solar PV increasingly depressing midday prices) whereas time-weighted prices are more
representative prices for baseload technologies.
Figure 6 Dispatch by fuel type and fuel shares for modelled grids: Reference case
Wind Other
Dispatch by fuel type Hydro 3.7% 2.9% 2014
Baseload 7.5%
300,000 Gas
8.4%
Peaking Gas
1.3%
250,000 Black coal
53.3%
Brown coal
23.0%
200,000
GWh sent-out
150,000
100,000
Solar Other
1.1% 1.1% 2030
Hydro Wind
50,000 6.2% 14.4%
Baseload Black coal

Gas 49.3%
0 5.5%
Brown coal
Peaking Gas 21.8%
0.7%
Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other
Note: Excludes non-scheduled generation in NEM regions, own-generation in the SWIS, off-grid generation and rooftop solar in all regions.
‘Other’ category includes cogeneration, liquid fuels, CCS-equipped technologies, biomass and geothermal.
Source: ACIL Allen
Figure 7 shows the new entrant capacity developed by the model over the period to 2040. A
total of around 7,650 MW of wind is developed throughout the NEM and SWIS regions over
the period 2016 to 2021.
Utility scale solar PV installations occur in the regional markets (NWIS, DKIS and Mt Isa) in
the period 2018-2021 due to the higher wholesale prices prevailing in these regions. Utility
scale solar accounts for around 24% of total generation (and 30% of total capacity) in these
markets by 2020 which represents a significant shift away from the currently dominant fuel,
natural gas.
Aside from the incentives provided through the LRET and several small solar plants that are
supported through other policies, no other renewable developments occur in the regional
grids (aside from those already committed) as costs remain above wholesale dispatch-
weighted price levels achievable by these technologies. For utility-scale solar PV in major
grids, dispatch-weighted prices are well below average prices due to the influence of
behind-the-meter solar PV which severely depresses mid-day prices. Similarly for wind,
wind farms increasingly earn a substantial discount to average pool prices due to
correlations of wind regimes in VIC-SA and within NSW.
Conventional fossil-fuelled capacity is projected to enter markets from around 2024 with
capacity largely being gas-based (900 MW of baseload gas; 2,980 MW of peaking gas by
2040). A total of 640 MW of new coal capacity is projected to be developed in the SWIS.
This is the only region to see new coal being developed in the Reference case.
10
Also shown in Figure 7 is a breakdown of the $26.8 billion (real 2014 dollars) in capital
expenditure on new generating capacity projected to enter the various modelled grids over
the period to 2040. Wind investment accounts for around 62% ($16.4 billion) of new large-
scale generation investment in the period to 2040. Gas-fired peaking plant and Utility-scale
solar PV each account for around 11% of the total ($3 billion). Several categories of fossil
fuel generation collectively account for the remaining 16% ($4.3 billion).
Figure 7 New entry capacity and capital expenditure: Reference case

New entrant capacity developed New entrant capital expenditure
16,000 6,000
14,000
5,000
12,000
4,000
Real 2014 $m
10,000
MW
8,000 3,000
6,000
2,000
4,000
1,000
2,000
0 0
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other Hydro Wind Solar Other
Source: ACIL Allen
Figure 8 shows new entrant capacity by grid. NEM-based regions only see wind entry in the
near-term, with QLD, SA and VIC also seeing some gas-fired peaking capacity in the long-
term (after 2034).
The modelling projects NSW to get around 4,000 MW of new wind in the period to 2020
which is the majority of that installed in the NEM. NSW has a reasonably good wind
resource in certain locations (most of the wind development to-date has occurred around the
ACT) which has favourable diurnal characteristics. NSW is also a large demand region able
to incorporate significant intermittent generation, however over time, the high correlations of
wind regimes results in deteriorating dispatch-weighted prices for wind farms.7
Whilst to-date Queensland has been relatively overlooked for wind development, rising gas
prices and the strongest demand growth of all NEM regions may see this change. Despite
its relatively poor wind resources, Queensland is projected to get around 550 MW of wind
developed, with its higher energy prices offsetting the effect of lower capacity factors.
South Australia is projected to receive an additional 820 MW of wind, further increasing the
already high level of penetration of intermittent generation sources into this region. Projected
lower energy output from gas-fired baseload/intermediate generators in South Australia as a
result of gas price increases allows the additional wind to be accommodated. The expansion
of the Heywood interconnector also allows greater exports to Victoria during period of high
wind output.
7 As more wind development occurs in correlated wind regimes, prices during windy periods increasingly become depressed,
resulting in lower pool revenues for these wind farms relative to average market prices.
11
Victoria sees around 800 MW of new wind developed. Development is limited due to the
high correlation with South Australian wind regimes (which results in low dispatch-weighted
prices for wind across the two regions) and planning restrictions which have resulted in
some wind proposals being abandoned and forced developers to look for more remote sites
which incur larger connection costs.
Tasmania, which is forecast to have declining demand growth, has ample generation
capacity to meet its needs and increasingly will be constrained by the limits on the Basslink
interconnector to export power to the mainland. While the modelling has a range of
proposed wind projects in Tasmania, the modelling sees the combined capital cost of wind
and interconnector capacity augmentation being higher than alternative renewable projects
elsewhere throughout Australia. As a result, these projects are not projected to proceed.
The IMO’s strong demand growth assumption results in a large amount of new entrant
capacity entering the SWIS. This takes the form a mix of wind initially followed by gas and
coal-fired capacity. It is noted that there is some doubt regarding the ability to develop new
intermittent renewable projects in the SWIS over the short to medium term, given the level of
retail concentration and uncertainty associated with potential reforms to the wholesale
electricity market.
Regional grids see a significant shift away from gas to large-solar PV in the period to 2020.
This is due to the high cost of wholesale electricity in these regions being the most attractive
for utility scale solar. However, there are some real world commercial constraints on the
uptake of solar in the near term. As no formal markets exist in these areas, supply is
acquired through long-term power purchase agreements (PPAs). To the extent that current
supplies are already underpinned by PPAs, penetration of solar may be limited to capturing
demand growth rather than displacing existing generation facilities as occurs within the
modelling.
12
Figure 8 New entrant capacity by grid: Reference case

NSW QLD SA
4,500 1,400 1,400
4,000 1,200 1,200
3,500
1,000 1,000
3,000
2,500 800 800
MW
MW
MW
2,000 600 600
1,500
400 400
1,000
500 200 200
0 0 0
2024
2038
2022
2036
2020
2034
2014
2016
2018
2020
2022
2026
2028
2030
2032
2034
2036
2040
2014
2016
2018
2020
2024
2026
2028
2030
2032
2034
2038
2040
2014
2016
2018
2022
2024
2026
2028
2030
2032
2036
2038
2040
Black coal Brown coal Peaking Gas Black coal Brown coal Peaking Gas Black coal Brown coal Peaking Gas
Baseload Gas Hydro Wind Baseload Gas Hydro Wind Baseload Gas Hydro Wind
Solar Other Solar Other Solar Other
TAS VIC SWIS

1 1,600 4,500
1,400 4,000
1 3,500
1,200
3,000
1 1,000
2,500
MW
MW
MW
800
2,000
0 600
1,500
400 1,000
0
200 500
0 0 0
2024
2038
2022
2036
2020
2034
2014
2016
2018
2020
2022
2026
2028
2030
2032
2034
2036
2040
2014
2016
2018
2020
2024
2026
2028
2030
2032
2034
2038
2040
2014
2016
2018
2022
2024
2026
2028
2030
2032
2036
2038
2040
NWIS DKIS Mt Isa

700 600 700
600 500 600
500 500
400
400 400
MW
MW
MW
300
300 300
200
200 200
100 100 100
0 0 0
2024
2038
2022
2036
2020
2034
2014
2016
2018
2020
2022
2026
2028
2030
2032
2034
2036
2040
2014
2016
2018
2020
2024
2026
2028
2030
2032
2034
2038
2040
2014
2016
2018
2022
2024
2026
2028
2030
2032
2036
2038
2040
Source: ACIL Allen
Figure 9 provides the generation capacity in aggregate for all modelled grids by fuel type
and the share of capacity for 2014 and 2030. Aggregate capacity is projected to increase to
around 60,000 MW, with wind entry accounting for most of the increase.
The incumbent fossil fuel fleet is projected to largely remain in service through to 2040 with
coal-fired generators being refurbished through life extensions under this scenario. Coal-
fired generation remains by far the lowest cost form of baseload power in the period to 2040.
The incentives therefore for owners of coal-fired stations to keep plants operating are high,
particularly in the absence of explicit carbon pricing and when there may be difficulties in
obtaining licences to develop new coal plants.
13
Figure 9 Generation capacity by fuel type: Reference case

Wind Other
Installed capacity
4.7% 4.6% 2014
70,000
Hydro Black coal
14.8% 36.1%
60,000
Baseload Gas
10.9%
50,000
Peaking Gas Brown coal
17.2% 11.7%
40,000
MW
30,000
Solar Other
20,000 2.3% 2.4% 2030
Wind
Black coal
10,000 18.4%
31.4%
Hydro
0 12.9%
Brown coal
11.1%
Peaking Gas
11.2%
Hydro Wind Solar Other Baseload Gas
10.3%
Source: ACIL Allen
2.2.1 RET outcomes

Figure 10 shows a summary of LRET market outcomes including the annual demand-supply
balance and the amount of LGCs banked. The market commences the 2014 compliance
year with a substantial balance of banked LGCs (estimated at around 26 million after
accounting for 2013 surrenders8) which, when added to projected certificate creation from
existing and committed generators, is sufficient to meet liable entities obligations (including
assumed demand from GreenPower and other sources) up to and including the 2017
compliance year (due for surrender in February 2018).
LGC creation from existing and committed accredited generators remains largely flat at
around 17 million per annum (assumptions on LGC creation from incumbents is undertaken
outside the modelling on a station-by-station basis).
New entrant LGC creation represents projected LGC supply from new large-scale
renewable plant constructed by the model. Supply of LGCs from these new entrants ramps
up quickly to 2020 although it lags slightly the demand for LGCs. This results in the amount
of banked LGCs decreasing each year as liable entities are required to draw down banked
LGCs to meet obligations in these years. Banked LGCs are exhausted in 2019 and a small
shortfall occurs in this year (around 1 million LGCs or around 2.3% of the aggregate
demand in this year). This shortfall is carried forward and met by excess LGCs produced in
subsequent years and hence no penalties are paid in this scenario.9
8 Also accounting for lags in creation for calendar year 2013

9 The RET Regulations allow for a 10% leeway provision on liabilities before Shortfall Penalties are levied. Any shortfalls up to
10% of liability are carried forward into future year liabilities for individual entities. In addition, if Shortfall Penalties are paid
(a liable entities shortfall is greater than 10% of its liability), liable entities have up to 3 years to provide certificates to meet
prior year shortfalls in order to receive a refund of any penalties paid.
14
Figure 10 LRET supply-demand balance and price outcomes: Reference case

LRET supply-demand balance Projected price outcomes
50,000 100
45,000 90
40,000 80
35,000 70
Real 2014 $/LGC

LGCs ('000)
30,000 60
25,000 50
20,000 40
15,000
30
10,000
20
5,000
10
0
0
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Incumbent LGC Creation New Entrant LGC Creation
Spot price Tax-adjusted Penalty price
Total LGC demand Banked LGCs
Note: Incumbent LGC creation is from existing, under construction or committed large-scale projects. Total LGC demand includes mandated
demand (including WCMG volumes) plus voluntary demand. Banked LGCs are as at end of calendar year. Spot LGC prices are those sufficient
to make new large-scale investments commercially viable. The penalty price represents the tax-adjusted penalty level (nominal $65 which is
non-deductible for tax purposes).
Source: ACIL Allen
The figure also provides the projected LGC spot price trajectory over the period to 2030.
The LGC price for 2014 is $54.90 – significantly higher than the spot price of around $30 at
the time of writing (mid-July 2014) which reflects significant policy uncertainty. The
modelling assumes that any changes to the policy are instantly reflected in market prices.
Prices trend upward to 2019 (the period in which banked LGCs exist) in accordance with the
assumed holding cost assumption of 5% real. Prices peak at around $70/LGC in 2020, then
trend down, reaching $40/LGC in 2030. The demand for LGCs is represented as a
constraint within the model’s linear programming formulation. This price series is derived
from the shadow price of this constraint which represents the additional revenue stream
necessary to make the marginal renewable project commercially viable.
Renewable energy developers and electricity retailers will tend to construct contract prices
which are flat in real terms and generally bundle the energy and LGC components into a
single price. The combined revenue streams of the underlying energy price (sometimes
referred to as ‘black’ energy) plus the LGC revenue stream are just sufficient to make the
marginal large-scale renewable project developed by the model commercially viable over its
economic life.
Figure 11 shows the composition of LGCs created by fuel source, annually and in aggregate
over the period to 2030. Wind is the dominant technology which benefits from the LRET,
accounting for almost 83% of all certificates created. Large-scale solar PV is the second
largest component, with contributions from already committed projects receiving additional
subsidies through solar flagships and other schemes and projected commercial installations
in regional grids and off-grid locations. Other sources such as bagasse, landfill and
sewerage gas are projected to increase over historical levels, but still remain a relatively
small component in the overall supply mix.
15
Figure 11 LGC creation by source: Reference case

LGC creation by source
50,000 Total share of LGCs (2014-30)
45,000 Geothermal
0.1%
40,000 Hydro Sewerage
3.7% gas
35,000 Bagasse 0.5%
Other 2.5%
LGCs ('000)
30,000 Landfill gas

0.1% 2.8% Solar
25,000 Wood 5.5%
1.1% WCMG
20,000
0.9%
15,000
10,000
5,000
Wind
0 82.8%
Bagasse Geothermal Hydro Landfill gas

Sewerage gas Solar WCMG Wind
Wood Other
Note: LGC creation by generation year. Other category includes food, crop and agricultural and municipal s olid wastes. WCMG = waste coal
mine gas.
Source: ACIL Allen
Figure 12 shows the projected STC creation from Small Generating Units (SGUs) 10 and
Solar Water Heaters (SWHs) over the period to 2030. Initial creation is relatively constant at
around 14 million STCs from 2015-2017 before declining virtually linearly to zero by the end
of the period. This is due to the declining deeming period offered by the scheme: a
maximum of 15 years for SGUs, 10 years for SWHs, declining annually by one year to
phase out at the scheduled end of the scheme in 2030. In all years the market price
assumed for STCs is $38/STC (nominal) which is slightly under the fixed clearing-house
price of $40/STC.
SWH systems are assumed to have an economic life of 10-15 years, with one third of
systems replaced at 10 years; one third at 12.5 years and the remaining third after 15 years.
Installations of new SWH are projected to occur at a relatively stable pace, with aggregate
installs increasing from an estimated 915,000 at the end of calendar year 2014 up to over
1.9 million systems by 2040. Net systems additions are projected to run at around 35,000-
45,000 systems per annum over the modelling horizon. STCs are created from solar water
heater installs regardless of whether the unit is a new installation or a replacement system.
The increase in STCs created from SWH systems from 2018 through early next decade
reflects the large replacement task based on the large surge of installs which occurred in the
period 2008-2010.
Energy generation/displacement from these systems is projected to increase almost at a
linear rate over the period to 2030, rising from around 8,000 GWh in 2014 to just under
21,000 GWh by 2030 and 28,000 GWh by 2040.
10 SGUs comprise of solar PV, micro wind and micro hydro units although the latter two categories are negligible.
16
Figure 12 SRES up-take and market outcomes: Reference case

STC creation Energy generation/displacement
18,000 30,000
16,000
25,000
14,000
12,000 20,000
STCs ('000)
GWh
10,000
15,000
8,000
6,000 10,000
4,000
5,000
2,000
0 0
2015
2014
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Solar PV SWH Solar PV output SWH energy displacement
Note: Solar PV generation for units installed under SRES (<100 kW).
Source: ACIL Allen
Figure 13 presents the RET compliance costs both annually (left) and on a cumulative basis
over various timeframes (right). These represent the direct RET costs levied on liable
entities under the schemes (generally retailers) and passed on to consumers. Under the
Reference case, annual compliance costs peak in 2020 at over $3.2 billion, with the LRET
accounting for over $2.8 billion of this. Over the period to 2030 the total direct RET costs
totals $37.8 billion (real 2014 dollars) of which 88% is associated with the LRET.
Figure 13 RET compliance costs: Reference case

RET Compliance costs Cumulative RET compliance costs
3,500 40.0
3,000 35.0 4.5 4.5
2,500 30.0
Real 2014 $m
Real 2014 $ billion
2,000 25.0
20.0
1,500
33.3 33.3
15.0
1,000 2.8
10.0
500
5.0 11.5
0
0.0
2015-20 2015-30 2015-40
LRET SRES LRET SRES
Note: Sum of projected LGC price multiplied by mandated LRET targets and projected STC creation by assumed STC price
Source: ACIL Allen
Table 2 provides the key figures for the LRET and SRES market outcomes in tabular form.
Line items include LGCs created and surrendered, banked certificates, shortfalls (if any) and
other values required for the calculation of RPP and STP values each year.
17
Table 2 LRET and SRES market outcomes: Reference case
Units 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
LRET market outcomes

Incumbent LGC Creation ('000 LGCs) 15,051 16,748 17,123 17,172 17,211 17,248 17,282 16,563 16,594 16,628 16,666 16,711 16,765 16,830 16,910 17,010 17,093
New Entrant LGC Creation ('000 LGCs) 0 0 872 2,639 10,125 17,698 25,066 25,930 26,009 26,045 26,080 26,115 26,157 26,188 26,221 26,235 26,235
Bank Balance Year Start ('000 LGCs) 26,144 22,576 18,803 13,697 5,807 842 0 0 0 0 0 0 0 0 0 0 0
Created ('000 LGCs) 15,051 16,748 17,995 19,811 27,336 34,945 42,348 42,494 42,603 42,673 42,746 42,827 42,922 43,018 43,131 43,245 43,328
Net Banking ('000 LGCs) -3,569 -3,772 -5,106 -7,890 -4,965 -842 0 0 0 0 0 0 0 0 0 0 0
Surrendered ('000 LGCs) 18,620 20,520 23,101 27,701 32,301 35,787 42,348 42,494 42,603 42,673 42,746 42,827 42,922 43,018 43,131 43,245 43,328
Bank Balance Year End ('000 LGCs) 22,576 18,803 13,697 5,807 842 0 0 0 0 0 0 0 0 0 0 0 0
Total Supply ('000 LGCs) 41,196 39,323 36,798 33,508 33,143 35,787 42,348 42,494 42,603 42,673 42,746 42,827 42,922 43,018 43,131 43,245 43,328
Total LGC demand ('000 LGCs) 18,620 20,520 23,101 27,701 32,301 36,901 43,520 42,670 42,670 42,670 42,670 42,670 42,670 42,670 42,670 42,670 42,670
LGC Shortfall ('000 LGCs) 0 0 0 0 0 1,114 1,172 176 1,181 1,169 100 1,024 918 0 315 343 0
LGC Shortfall% % 0% 0% 0% 0% 0% 3% 3% 0% 3% 3% 0% 2% 2% 0% 1% 1% 0%
LGC Shortfall Refunds ('000 LGCs) 0 0 0 0 0 0 0 0 1,114 1,172 176 1,181 1,169 348 777 918 658
LGC Spot price Real 2014 $/LGC 55.61 58.39 61.31 64.38 67.60 70.98 68.28 65.37 62.72 60.07 57.17 54.56 51.93 49.00 46.39 43.75 40.79
LGC Penalty price Real 2014 $/LGC 92.86 90.59 88.38 86.23 84.12 82.07 80.07 78.12 76.21 74.35 72.54 70.77 69.04 67.36 65.72 64.11 62.55
SRES market outcomes
SGU capacity at year end MW 4,133 4,760 5,375 6,020 6,645 7,239 7,821 8,391 8,948 9,488 10,015 10,523 11,031 11,522 12,021 12,495 12,954
SWHs installed at year end ('000) 916 951 986 1,022 1,058 1,095 1,133 1,171 1,209 1,249 1,288 1,328 1,369 1,410 1,452 1,495 1,538
Solar PV output GWh 5,167 6,019 6,823 7,635 8,445 9,214 9,957 10,677 11,378 12,056 12,710 13,339 13,957 14,556 15,151 15,725 16,271
SWH energy displacement GWh 2,867 2,968 3,071 3,175 3,280 3,387 3,495 3,605 3,717 3,830 3,944 4,060 4,178 4,297 4,419 4,541 4,666
Total generation/displacement GWh 8,034 8,987 9,893 10,810 11,725 12,601 13,452 14,283 15,095 15,885 16,654 17,400 18,135 18,854 19,570 20,267 20,937
STC creation from SGU ('000 STCs) 14,077 12,685 12,125 11,818 10,610 9,286 8,327 7,378 6,477 5,553 4,703 3,866 3,173 2,404 1,768 1,040 387
STC creation from SWH ('000 STCs) 1,410 1,599 1,733 2,006 2,415 3,570 3,421 3,469 3,702 2,793 2,921 2,116 1,571 1,119 815 543 231
Total STC creation ('000 STCs) 15,487 14,283 13,858 13,824 13,025 12,855 11,748 10,847 10,180 8,346 7,624 5,983 4,744 3,523 2,583 1,583 618
STC price Real 2014 $/STC 38.00 37.07 36.17 35.29 34.43 33.59 32.77 31.97 31.19 30.43 29.69 28.96 28.26 27.57 26.89 26.24 25.60
RET summary
Relevant acquisitions GWh 205,266 206,756 212,186 215,662 217,408 218,820 220,308 221,991 223,488 224,863 226,353 228,155 230,181 232,279 234,540 236,970 238,944
PECs GWh 27,200 23,377 25,669 27,190 28,157 29,066 30,005 30,014 30,128 30,057 30,149 30,131 30,190 30,246 30,362 30,476 30,525
Reduced acquisitions GWh 178,066 183,379 186,517 188,472 189,251 189,753 190,303 191,976 193,360 194,806 196,204 198,023 199,991 202,032 204,178 206,494 208,419
RPP % 9.87% 10.28% 11.49% 13.81% 16.19% 18.57% 21.99% 21.36% 21.20% 21.05% 20.90% 20.70% 20.50% 20.29% 20.08% 19.86% 19.67%
Direct LGC cost Real 2014 $/MWh 5.49 6.00 7.05 8.89 10.94 13.18 15.02 13.96 13.30 12.64 11.95 11.30 10.65 9.94 9.32 8.69 8.02
STP % 10.48% 6.06% 7.43% 7.33% 6.88% 6.77% 6.17% 5.65% 5.26% 4.28% 3.89% 3.02% 2.37% 1.74% 1.27% 0.77% 0.30%
Direct STC cost Real 2014 $/MWh 3.98 2.25 2.69 2.59 2.37 2.28 2.02 1.81 1.64 1.30 1.15 0.88 0.67 0.48 0.34 0.20 0.08
Note: Presented on a calendar year basis. Total LGC demand includes mandated demand (including WCMG volumes) plus voluntary demand. Direct LGC cost = LGC spot price x RPP. Direct STC cost = STC
price x STP. PECs = Partial Exemption Certificates.
Source: ACIL Allen

2.2.2 Resource costs, profitability and emissions

Figure 14 presents the annual generation sector resource costs broken down by category
for modelled grids. Resource costs represent the annual expenditure by the sector on:
 Capital expenditure (on both generating capacity and any interconnector
expansions/augmentations11)
 Refurbishment of existing and new generators for life extension beyond initial economic
life
 Fixed operating costs (fixed costs associated with normal operation and stay in business
capital expenditure associated with existing and new generating capacity)
 Variable operating costs (fuel costs and variable O&M costs for existing and new
generation)
 Unserved energy (which, if occurs, is valued at the Value of Lost Load (VoLL) equivalent
market price cap).12
These costs exclude the cost of carbon, which only applies in the first half of 2014 in any
case (i.e. carbon costs are not included as part of the variable O&M cost component).
Aggregate resource costs for the Reference case total $121.9 billion in present value terms
over the period to 2040 using a discount rate of 7% pre-tax real.13 $102.9 billion of this total
is associated with large-scale wholesale energy and $19 billion is associated with
installation of small-scale solar PV and SWH systems. The largest cost components for the
system (other than the build cost) are fixed and variable O&M costs associated with fuel and
maintaining the existing generating fleet.
11 Capital expenditure on new-build capacity is pro-rated down (on a NPV equivalent annuity basis) based on the remaining
modelling horizon relative to the full economic life of technologies. No interconnector expansions occur in the Reference
case.
12 Unserved Energy is routinely calculated as part of sector resource costs. It represents the value of any energy which is not
supplied through non-voluntary load shedding. Whilst the value of unserved energy across all scenarios is zero it is shown
for completeness.
13 A higher 10% pre-tax real discount rate is applied to generation projects weighted average cost of capital in assessing
commercial viability.
19
Figure 14 Electricity generation sector resource costs: Reference case

Resource costs (to 2040) NPV (7%)
16,000
Total 121.9
14,000
Total Small-scale 19.0
12,000
SRES: SWHs 6.4
Real 2014 $m
10,000
SRES: SGUs 12.6
8,000
Total Large-scale 102.9
6,000 Unserved Energy 0.0
4,000 Variable O&M 46.9
2,000 Fixed O&M 38.3
0 Refurbishment 3.4
New build 14.4

New build Refurbishment Fixed O&M
0 50 100 150
Variable O&M Unserved Energy SRES: SGUs
NPV ($ billion)
SRES: SWHs
Note: Capital expenditure on new-build capacity and refurbishments is pro-rated down (on a NPV equivalent annuity basis) based on the
remaining modelling horizon relative to the full economic life of technologies.
Source: ACIL Allen
Figure 15 presents calculated profitability for generators grouped by fuel type. The
profitability measure which is similar to earnings before interest, tax, depreciation and
amortisation (EBITDA) value is calculated as modelled revenues (energy and LGCs) less
fixed operating and maintenance costs and variable generating costs. When projected
future values are discounted to present value, it provides a proxy for valuation purposes.14
The entry of wind from 2017 to 2020 can be seen to reduce profit margins for coal-fired plant
down to negligible levels in 2020. The value for this year implies generators are only just
covering avoidable costs and there is little net cash available for debt repayment, let alone
returns to equity. Over time margins increase as wholesale prices rise and generation
volumes increase.
The chart on the right shows resulting NPVs when this EBITDA stream is discounted using a
7% real discount rate. These results can be considered to be a crude aggregate valuation
for the generation fleet. Using this measure, the Reference case implies the total coal fleet is
valued at around $25.5 billion. Large-scale wind and solar projects (both existing and
projected to be developed) would be valued at around $19.5 billion. Results for gas-fired
technologies and hydro are relatively small by comparison, but this analysis ignores the
additional revenues these technologies earn through hedge products, which for peakers and
hydro stations may account for a large proportion of total revenues.15
14 The modelling methodology employed for this exercise is less detailed than market simulation modelling and these
calculations ignore contract premiums and revenues derived through hedge products so will tend to understate values.
15 PowerMark LT, as a long-term least cost planning model also does not characterise the top portion of the price duration as
well as simulations models do and hence will tend to underestimate generation and revenues from peaking generation.
20
Figure 15 Electricity generation sector profitability by fuel type: Reference case

EBITDA by fuel type NPV (7%)
9,000
Geothermal 0.0
8,000
7,000 Solar 2.0
6,000
Real 2014 $m
Wind 17.5
5,000
4,000 Hydro 2.0
3,000
Baseload Gas 1.5
2,000
1,000 Peaking Gas 0.0
0
Brown coal 7.1
-1,000
Black coal 18.4
Black coal Brown coal Peaking Gas Baseload Gas -5 0 5 10 15 20

Hydro Wind Solar Geothermal NPV ($ billion)
Note: EBITDA measure calculated as modelled pool revenues (energy and LGCs) less fixed operating and maintenance costs and variable
generating costs. Excludes revenues from contract premium or hedge products; NPV calculated over the period 2015-2040.
Source: ACIL Allen
Projected greenhouse gas emissions under the Reference case rise initially from
171 Mt CO2-e in 2014 (an emission intensity of 0.86 tonnes CO2-e/MWh sent-out) up to
181 Mt CO2-e by 2017 (0.87 tonnes CO2-e/MWh sent-out).16 This is largely a result of the
withdrawal of gas plant from service due to rising wholesale gas prices, although the repeal
of the carbon tax in mid-2014 also has some impact on near-term outcomes. Both of these
influences tend to increase the competitiveness of coal-fired generation in the short-term
and result in higher emissions from the sector. As shown in Figure 16, black and brown coal
account for the vast majority of emissions from the generation sector, with other fuels only
accounting for around 7% of the total in 2014.
The introduction of wind under the LRET reduces emissions later in the decade down to
165 Mt CO2-e in 2020 (0.78 tonnes CO2-e/MWh sent-out), before rising with demand
growth. The two noticeable single year drops in 2031 and 2035 relate to major brown coal
refurbishments which occur in these years, with the model withdrawing units from service
while the generator overhauls are undertaken.
16 These figures do not represent total emissions from electricity generation as they exclude non-scheduled generation in NEM
regions, own-generation and off-grid generation components which have not been explicitly modelled for this process.
21
Figure 16 Electricity generation sector emissions and emissions intensity: Reference case
Greenhouse gas emissions Emissions intensity
200 0.90
180 0.88
160 0.86
Tonnes CO2-e/MWh sent-out

140
0.84
120
Mt CO2-e
0.82
100
0.80
80
0.78
60
0.76
40
0.74
20
0.72
0
0.70
Black coal Brown coal Peaking Gas Baseload Gas Other
Note: Excludes non-scheduled generation in NEM regions, own-generation in the SWIS and off-grid generation. ‘Other’ category includes
Source: ACIL Allen
2.2.3 Retail price outcomes

Retail prices are modelled for four customer types in each major grid/region as a ‘cost stack’
of all key retail cost components: wholesale, network, green schemes (RET and other),
losses and retailing costs (costs and margins).
Stylised average Australian retail price series for each customer class have been calculated
by weighting retail price series by demand in each region. These series are shown in Figure
17. However, readers should note that the series presented will only be broadly reflective of
the prices paid by any individual customer due to variation between regions and differences
in the consumption patterns of individual customers. Assumed annual consumption for each
customer group is as follows:
 Residential: 5.9-8.3 MWh per year (varies across networks)
 Commercial: 2,900 MWh per year
 Industrial: 6-10 GWh per year
 Emissions Intensive Trade Exposed Industry: Generally very large.
The broad trend is of a slight decline in retail prices from 2014 to 2015 and then relatively
flat prices until the middle of the next decade. The residential series rises slightly over the
long-term, while commercial and industrial series fall slightly in real terms. These trends are
principally driven by:
 The assumed repeal of carbon pricing from mid-2014 and an associated fall in
wholesale costs
 A flattening of network costs driven by an expected move towards a lower regulated
return on capital in future determination periods (an 8% nominal return on capital is
assumed for future regulatory periods, reflecting recent decisions in NSW and South
Australia, as compared to rates of around 9% to 10% under many existing
determinations) and a flattening in capital expenditure trends
 An assumed gradual rebalance in tariff structures between fixed and variable
components (fixed, demand and capacity charges were assumed to comprise 50% of
residential and commercial network charges by 2030)
22
 Growth in wholesale costs as demand growth results in a tightening supply-demand

balance in the wholesale market in the absence of ongoing new renewable entry.
Direct RET costs for consumers are dominated by LRET costs, due to high LGC prices and
the rising Renewable Power Percentage (RPP) as the fixed annual GWh renewable target
increases. Direct LRET costs increase to 2020, peaking at around $13/MWh for non-exempt
consumers. Direct costs to partially exempt consumers (on the highest level of exemption)
are on average around one third of those to non-exempt consumers on a per MWh basis.
SRES costs drop in 2015 as modelled STC creation in 2014 is lower than that implied by the
(pre-determined) Small-scale Technology Percentage (STP), and the difference is netted off
the implied SRES target for 2015. SRES costs decline consistently from 2017 as the
deeming period for SGUs is progressively reduced to reflect the approaching end date of the
RET in 2030.
Figure 17 Weighted average Australian retail prices and direct RET costs: Reference case
Average retail cost Non-exempt consumers
20
30 LRET SRES
Real 2014 $/MWh

15
25 10
Real 2014 c/kWh
5
20
0
2017
2029
2014
2015
2016
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2030
15
10 Partially-exempt consumers
5
LRET SRES
Real 2014 $/MWh
4
5 3
2
0
1
2016
2020
2014
2018
2022
2024
2026
2028
2030
2032
2034
2036
2038
2040
0
2019
2014
2015
2016
2017
2018
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Residential Commercial Industrial EITE
Note: Partially exempt customers are presented for those coming under the ‘Highly’ emission intensive category ( e.g.. aluminium).
Source: ACIL Allen
Figure 18 shows a breakdown of the modelled retail price components for average
residential consumers throughout Australia. Network costs remain by far the largest cost
component, accounting for 50-55%, followed by wholesale energy costs at 20-25%. The
RET currently comprises around 3.7% of total costs, with this projected to rise to around
6.6% by 2020.After 2020, RET costs decline as a proportion of total retail prices, with
compliance costs accounting for 2.3% in the scheme’s final year, 2030. As the scheme ends
in 2030, no direct costs are incurred after 2030.
23
Figure 18 National average residential retail tariff components: Reference case

Losses Retailing costs
Residential tariff breakdown Other
3.2% 13.0% 2014
green
30 schem
es
5.0%
SRES
1.6%
Network
25 51.7%
LRET
2.1% Wholesale
energy
23.4%
20
Real 2014 c/kWh
15
10
Other green Retailing costs
schemes Losses 12.3% 2020
2.6% 2.8%
SRES
5 0.9%
LRET
5.7%
0 Wholesale
Network
56.3%
energy
19.4%
Network Wholesale energy

LRET SRES
Other green schemes Losses
Retailing costs
Note: Wholesale energy component includes simulated NSLP load-weighted price plus hedging costs. Other green schemes include energy
efficiency schemes and feed-in-tariffs. Retail cost component includes retail cost to serve and retail margins.
Source: ACIL Allen
24
3 Repeal case

Under the Repeal case it is assumed that the SRES and LRET schemes cease to operate
from 1 January 2015 with 2014 being the last compliance year. Projects that have already
reached financial close and/or begun construction are assumed to be completed and come
into operation in this scenario. This scenario assumes that any mechanism introduced to
compensate investments made under the RET (if any) does not affect wholesale or retail
price outcomes.
3.2 Results
Under the Repeal case wholesale electricity prices do not fall in the period 2017-2020 due to
there being no new renewable plant entry through this period as shown in Figure 19. Prices
rise very gradually in NEM states, with increased demand in Queensland a key driver. The
lack of wind entry prompts a return to service of mothballed coal capacity over time as
market conditions warrant and this limits the extent to which prices rise towards new entry
levels. Even in the absence of renewable capacity though, no new baseload capacity is
required in NEM regions within the modelled period. This is due to the muted demand
growth in the NEM, resulting in more expensive new entrant capacity generally not being
required.
As demand growth is higher in the regional grids, prices rise much faster than in the NEM
under this scenario, with new gas-fired entry projected to occur from around 2023 (primarily
operating as ‘intermediate’ plant at mid-level capacity factors) followed by conventional coal
plant from 2025 onwards. From this point onward, wholesale prices in non-NEM grids
maintain a price level of around $85-$90 in real terms as input costs stabilise and the
markets remain at their long-run equilibrium price level.
While the price differential between the scenarios peaks early next decade, prices tend to
converge over the longer-term as the wholesale markets slowly recover to long-run
equilibrium new entrant levels.
Figure 19 Wholesale price outcomes: Repeal case

100
80
Real 2014 $/MWh
60
40
20
0
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
Reference case NEM Repeal case NEM

Reference case Other grids Repeal case Other grids
prices for grids (NEM = NSW, QLD, SA, TAS and VIC; Other = SWIS, NWIS, DKIS and Mt Isa).
Source: ACIL Allen
25
The generation mix under the Repeal case is virtually static at current shares. Existing coal-
fired generation increases output to meet most of the load growth over the period to 2040. In
the absence of the RET, very little renewable development occurs due to the wholesale
price outcomes being below those required to stimulate renewable entry in the absence of
subsidies. As shown in Figure 20, energy which would have been supplied by new wind and
solar under the Reference case is met mainly by black and brown coal. Aggregate gas-fired
generation remains relatively stable at around 15-16 TWh per annum. This is up to 6.3 TWh
higher relative to the Reference case in the period 2020 to 2026. By 2040 though, gas-fired
generation is largely unchanged from the Reference case.
Dispatch from wind and solar technologies is 567 TWh lower in aggregate to 2040 under the
Repeal case, which is offset by 485 TWh of coal and 86 TWh of gas-fired generation.
Figure 20 Dispatch by fuel type: Repeal case

Dispatch outcomes Change (Repeal case - Reference case)
300,000 40,000
250,000 30,000
20,000
200,000
GWh sent-out
GWh sent-out
10,000
150,000
0
100,000
-10,000
50,000
-20,000
0 -30,000
Source: ACIL Allen
Figure 21 shows the new entrant capacity developed by the model over the period to 2040
under the Repeal scenario. No new large-scale renewable capacity is developed in the NEM
under this scenario. Indeed, very little capacity is developed throughout Australia over the
period to 2025, which reflects the market’s current level of oversupply and limited growth in
demand.
Conventional fossil-fuelled capacity is projected to enter markets from around 2025 in the
SWIS and from 2032 in the NEM, with capacity largely being gas-based (900 MW of
baseload gas; 2,300 MW of peaking gas). A total of 1,280 MW of new coal capacity is
projected to be developed in the SWIS. This is the only region to see new coal being
developed in the Repeal case.17
Utility scale solar PV installations still occur in the regional markets (NWIS, DKIS and Mt
Isa), but much later than in the Reference case, from around 2035 onwards. Subsidies
through the LRET are necessary to make these investments commercially viable before that
time.
17 As per the scenario settings, new coal entry is permitted to occur in the modelling however ACIL Allen recognise that there
are a range of financial and other constraints which may create obstacles to such developments occurring in practice.
26
The fact that the modelling replaces the almost 9,000 MW of capacity built by 2030 in the
Reference case with only 640 MW of additional new coal capacity in the Repeal case
indicates that the vast majority of the generating capacity brought on by the LRET is surplus
to market needs.
Figure 21 New entry capacity: Repeal case

New entrant capacity developed Change (Repeal case - Reference case)
7,000 2,000
6,000 0
5,000
-2,000
4,000
MW
MW
-4,000
3,000
-6,000
2,000
1,000 -8,000
0 -10,000
Source: ACIL Allen
Figure 22 presents the capital expenditure of new entrant capacity under the Repeal case
(left) and the change relative to the Reference case (right). Capital expenditure on wind is
$16.6 billion lower; large-scale solar $1.1 billion lower; peaking gas-fired generation $660
million lower and coal-fired generation $2.6 billion higher. Under this scenario, capital
expenditure on new generation capacity is deferred until around mid-next decade for the
SWIS and 2032 for the NEM.
Figure 22 New entrant capital expenditure: Repeal case

New entrant capital expenditure Change (Repeal case - Reference case)
1,400 2,000
1,200 1,000
0
1,000
Real 2014 $m
Real 2014 $m
-1,000
800
-2,000
600
-3,000
400
-4,000
200 -5,000
0 -6,000
Source: ACIL Allen
27
Under the Repeal scenario overall generation capacity is largely unchanged from current
levels as shown in Figure 23. The figure on the right shows the change in capacity relative
to the Reference case. In the absence of the wind and solar capacity mothballed capacity is
reintroduced to the market earlier (brown coal from 2017) and a black coal plant is
refurbished in 2028 which did not get refurbished in the Reference case.
Figure 23 Generation capacity by fuel type: Repeal case

Installed capacity Change (Repeal case - Reference case)
60,000 4,000
50,000 2,000
0
40,000
-2,000
MW
MW
30,000
-4,000
20,000
-6,000
10,000 -8,000
0 -10,000
Source: ACIL Allen
3.2.1 RET outcomes

With the repeal of the scheme, the LRET ceases to function and the market value of LGCs
falls to zero immediately upon announcement. There are no further surrender obligations on
liable entities beyond the 2014 compliance year. For compliance year 2014, we have
assumed the LGC price is the same as within the Reference case as retailers would have
already locked in costs in contracts to customers. Therefore there is no modelled change to
direct LRET costs within the Repeal case for calendar year 2014.
Similarly for the SRES, there are no future liabilities beyond the 2014 compliance year under
the Repeal case.
Development of small-scale SGU and SWH technologies still occur albeit at a slightly lower
uptake rate than under the Reference case as shown in Figure 24. Initial install rates for
solar PV are around 30% below rates under the Reference case, however, the higher retail
prices under the repeal case result in a slightly faster install rate from around 2024. As a
result, the cumulative installations under a Repeal scenario begin to narrow the gap
between the two series from that point on.
Throughout the modelling process it was revealed that the level of uptake from solar PV is
more dependent upon the structure of retail electricity tariffs (whether these remain largely
variable or move to more fixed/demand based charging for network components) rather than
subsidies provided through the SRES. ACIL Allen’s retail model makes some assumptions
regarding a slow transition to more cost reflective tariff structures (see section A.6.2 for
more detail) which may explain why solar PV uptake rates are slower than in some other
studies.
28
Figure 24 Energy generation/displacement from SRES: Repeal case

Energy generation/displacement Change (Repeal case - Reference case)
25,000 0
-200
20,000 -400
-600
15,000
GWh
GWh
-800
-1,000
10,000
-1,200
5,000 -1,400
-1,600
0 -1,800
2015
2014
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Solar PV output SWH energy displacement Solar PV output SWH energy displacement
Total (Reference case)
Note: Energy values presented do not include network losses that would otherwise arise.
Source: ACIL Allen
There are no future compliance costs for the RET under the Repeal case as shown in
Figure 25. The scenario results in a reduction of subsidies to renewables in the order of
$37.8 billion over the period to 2030.
Figure 25 RET compliance costs: Repeal case

RET Compliance costs Change in cumulative RET compliance costs
3,500 0.0
3,000 -5.0 -11.5
2,500 -10.0
Real 2014 $m
Real 2014 $ billion
-2.8
2,000 -15.0
-33.3 -33.3
-20.0
1,500
-25.0
1,000
-30.0
500
-35.0 -4.5 -4.5
0
-40.0
2015-20 2015-30 2015-40
LRET SRES Reference case LRET SRES
Source: ACIL Allen

Figure 26 presents the change in electricity generation resource costs broken down by
category for modelled grids. The figure on the left shows the annual changes relative to the
Reference case, while the figure on the right shows the calculated NPV of the changes over
the period to 2040 using a discount rate of 7% pre-tax real.
The year-on-year resource cost differences between the two scenarios highlights the
additional expenditure, predominantly capital expenditure on new renewable capacity,
prompted by the LRET under the Reference case. The savings associated with the lower
capacity under the Repeal case also result in savings to fixed O&M costs associated with
29
servicing this equipment. Offsetting these savings are higher variable O&M costs (principally
fuel costs) and refurbishment costs associated with increased utilisation of the existing
generating fleet.
In aggregate resource costs for the Repeal case total $107.9 billion in present value terms.
Relative to the Reference case, resource costs associated with large-scale energy under
this scenario are $11 billion lower in the Repeal case, while resource costs associated with
small-scale systems are $3 billion lower. This combines to a total saving of $14 billion in
NPV terms under the Repeal scenario.18
Figure 26 Change in electricity generation sector resource costs: Repeal case

Change (Repeal case - Reference case) NPV of change (7%)
2,000
Total -14.0
1,000
Total Small-scale -3.0
0
SRES: SWHs -1.1
-1,000
Real 2014 $m
SRES: SGUs -1.9

-2,000
-3,000 Total Large-scale -11.0
-4,000 Unserved Energy 0.0
-5,000 Variable O&M 4.5
-6,000 Fixed O&M -3.9

-7,000 Refurbishment 0.8
New build -12.4

-20 -10 0 10
NPV ($ billion)
SRES: SWHs
Source: ACIL Allen
Figure 27 presents the changes in calculated generator profitability by fuel type. The Repeal
of the RET is clearly negative for large-scale wind and solar developers, with around $2
billion less earnings each year from 2019 to 2030. Reductions in earnings continue post-
2030 as less renewable energy is developed in aggregate.
Incumbent generators are projected to be more profitable under the Repeal scenario. This is
due to both the higher wholesale prices which result and the larger generation volumes
which occur. Due to coal-plant facing the largest decline in output from renewable
displacement, these technologies improve the most under the Repeal. Interestingly, existing
hydro is also better off in net terms from repeal of the scheme with higher wholesale prices
more than compensating for the forgone LGC revenue (the majority of hydro generation
comes under the pre-RET baseline and is therefore excluded from the LRET).
In NPV terms (as shown in the figure on the right) renewables are $16 billion worse off,
while coal-fired plant is $16.3 billion better off from repeal of the scheme.
18 Note that results presented at the preliminary modelling workshop showed a difference between the scenarios of $12.8
billion using a discount rate of 10% pre-tax real. The 7% discount rate is more reflective of a social discount rate whereas
10% is more suited for commercial project evaluation.
30
Figure 27 Electricity generation sector profitability by fuel type: Repeal case

Change (Repeal case - Reference case) NPV of change (7%)
4,000
Geothermal 0.0
3,000
Solar -1.8
2,000
Real 2014 $m
Wind -14.2
1,000
Hydro 1.0
0
Baseload Gas 1.3
-1,000
Peaking Gas 0.1
-2,000
Brown coal 4.2
-3,000
Black coal 12.1
Black coal Brown coal Peaking Gas Baseload Gas -20 -15 -10 -5 0 5 10 15
generating costs
Source: ACIL Allen
The key economic benefit from wind entry under the LRET is the emissions abatement it
provides. In 2020 emissions are projected to be 23.7 Mt CO2-e higher, with aggregate
emissions over the period 2015-20 being 58.2 Mt CO2-e higher. Under the Repeal case,
emissions continue an upward trajectory increasing to 210 Mt CO2-e by 2040 (a 23%
increase over projected 2014 levels for modelled grids).
Figure 28 shows the change in emissions and intensity resulting from the Repeal of the
scheme. From 2020 to 2030 emissions from the sector are around 24 Mt CO2-e higher per
annum. Aggregate emissions are 299 Mt CO2-e higher in the period to 2030 and
520 Mt CO2-e higher in the period to 2040.
Under the Repeal scenario, the sector’s emission intensity remains relatively static,
averaging 0.88 tonnes CO2-e/MWh sent-out.
Figure 28 Change in modelled electricity generation sector emissions: Repeal case

Change (Repeal case - Reference case) Emissions intensity
30.0 0.90
0.88
25.0
0.86
20.0
0.84
Mt CO2-e
15.0 0.82
10.0 0.80
0.78
5.0
0.76
0.0
0.74
Reference case
-5.0 0.72
Repeal case
0.70
Source: ACIL Allen
31

Initially the repeal of the RET results in a reduction in retail prices to consumers (the
reduction in direct RET costs outweighs any wholesale impact) until around 2020, when this
reverses (the wholesale price suppression outweighs direct RET costs) for the remainder of
the period. For EITE industries, which are partially exempted from the direct RET costs, the
switch occurs earlier (around 2018), although most EITE users operate under long-term
electricity contracts and are generally insulated from annual changes in pool prices.
Figure 29 provides the price components of the national average residential tariff. Repeal of
the RET results in a decrease in direct compliance costs associated with retailers having to
acquire and surrender certificates. This is illustrated by the reduction in costs to consumers
from the LRET and SRES components in the chart.19 However, this benefit is offset by the
projected higher wholesale electricity prices which prevail under this scenario, shown by the
blue bars. This also has flow on costs for network losses and retail costs/margins as these
are assumed to be a percentage of non-network costs within the retail model.
The chart on the right of Figure 29 shows the NPV of these changes over various time
frames. In the period 2015-20 the net effect is for a reduction of $247 for a typical household
electricity costs. Over the period to 2030, households are no better or worse off from the
repeal of the scheme, however over the very long-term households are projected to see
some small net cost of around $115 in present value terms. It should be noted that this
represents a very small change in aggregate household expenditure on electricity over the
period to 2040 (a 0.6% increase in present value terms).
Figure 29 Change in residential retail tariff components and NPV of annual bill: Repeal case
Change (Repeal case - Reference case) NPV of change in residential retail bills
2.50 (Repeal case - Reference case)
$1,000
2.00 $793
$800 $677
1.50
$600
1.00
Real 2014 c/kWh
$400
0.50
NPV ($)
0.00 $200 $121 $115
-0.50 $0
($1)
-1.00 ($200)
-1.50 ($247)
($400)
($367)
-2.00
($600)
($800) ($678) ($678)

Network Wholesale energy
LRET SRES 2015-20 2015-30 2015-40
Other green schemes Losses Direct RET costs Other costs Net cost
Retailing costs Total cost
Source: ACIL Allen
19 Wholesale energy price outcomes for 2017 are slightly lower under the Repeal case as shown in Figure 19. This is due to
differences in mothballed capacity more than offsetting the impact of wind entry in this year. This also explains the retail
price outcome in this year as shown in Figure 29.
32
4 Closed to new entrants

In this scenario it is assumed that the LRET scheme continues to operate, but is closed to
new installations from 1 January 2015.
The SRES, which operates under a deeming arrangement whereby installations receive
certificates upfront, does not continue to operate and is closed from 1 January 2015.
Under the LRET, installations receive certificates annually based on generation. Closure of
the scheme to new entrants (no new accredited generators) would result in the supply of
LGCs being limited to generators already accredited under the scheme.20
Wholesale market impacts

A feature of renewable generation is that virtually all technologies have a very low marginal
cost of generation.21 Therefore, regardless of LGC market prices, operation of these
facilities is expected to be unchanged.
Therefore, the modelled outcomes within the wholesale market are the same as the Repeal
case. ACIL Allen has verified this by modelling a scenario in which new entrants were
restricted from creating LCGs, with various demand levels set for the scheme. It was verified
that LGC prices were either zero (in the case where supply exceeds demand) or the penalty
price (where supply is less than demand), with no changes to the underlying wholesale
energy market from the Repeal case. However, it should be noted that these binary
outcomes result in part from the ‘perfect foresight’ nature of the model used. In reality,
market outcomes under these policy settings would tend to oscillate between the two
extremes due to variation in weather and strategic actions by retailers based on their LGC
holdings.
Table 3 provides the estimated annual targets required to clear LGCs projected to be
created from existing and committed projects. These are calculated as the sum of
incumbent LGC creation, plus an allowance to absorb banked LGCs, minus projected
voluntary demand for LGCs. These figures decline slightly over time as the output from
generators with finite resource bases decline (e.g. landfill gas).
20 An allowance for projects that have already reached financial close and/or begun construction as at a nominated date to
become accredited would seem reasonable and has been included in this scenario. ACIL Allen’s analysis suggest that at
the time of writing there is at least 480 MW of known projects committed but have not yet sought or gained accreditation
from the Clean Energy Regulator.
21 Exceptions include some biomass technologies which involve marginal costs associated with the physical collection task for
fuel. However, these technologies account for a very small proportion (<1%) of the existing supply mix.
33
Table 3 Estimated LRET to clear existing supply (LGCs ‘000)
Calendar Incumbent Allowance for Total LGC Voluntary LGC Adjusted LRET
year LGC Creation banked LGCs supply demand demand
2015 16,503 1,401 17,903 1,670 16,233

2016 16,784 1,401 18,184 1,670 16,514
2017 16,767 1,401 18,168 1,670 16,498
2018 16,738 1,401 18,138 1,670 16,468
2019 16,710 1,401 18,111 1,670 16,441
2020 16,684 1,401 18,085 1,670 16,415
2021 15,905 1,401 17,306 1,670 15,636
2022 15,883 1,401 17,283 1,670 15,613
2023 15,862 1,401 17,262 1,670 15,592
2024 15,842 1,401 17,242 1,670 15,572
2025 15,823 1,401 17,224 1,670 15,554
2026 15,805 1,401 17,206 1,670 15,536
2027 15,789 1,401 17,189 1,670 15,519
2028 15,773 1,401 17,173 1,670 15,503
2029 15,758 1,401 17,159 1,670 15,489
2030 15,744 1,401 17,145 1,670 15,475
Note: Allowance for banked LCGs calculated as 1/16th of projected banked certificates at the end of
2014 compliance year of 22.4 million. Estimated LGC creation from waste coal mine gas generators are
included within these figures until 2020.
Source: ACIL Allen analysis
LGC pricing
The intention of maintaining the scheme (but restricting new entrants) is to provide a level of
ongoing revenue for investments that have already been made (grandfather existing
investments). Two primary design options arise, either:
 Setting a level of demand for LGCs which would clear the market in an orderly fashion
and allow the market to determine the price
 Setting a fixed price for LGCs.
For a range of reasons, option 1 may prove difficult due to the uncertain nature of LGC
creation from some sources.
Alternatively, setting a fixed price would need to take into account:
 Political exposure
 Regulatory risk
 Potential for windfall gains or losses to projects/retailers.
These issues arise because any chosen level of fixed LGC price will result in wealth
transfers between industry players, and will determine whether wind farms or electricity
consumers face a higher share of the burden from the policy change. Due to the high level
of uncertainty of capital costs, black energy prices and broader market settings over the life
of a renewable project, any LGC price determined will inevitably be ‘wrong’ from a range of
stakeholders’ points of view.
For this scenario, ACIL Allen was instructed by the Expert Panel to examine the impact
based on a fixed $40/LGC price point (constant in nominal dollars) on retail price.
34
4.2 Results
Under the Closed to New Entrants scenario the wholesale market outcomes are the same
as for the Repeal case. Whilst there is a slight change to the uptake of small-scale systems
(as discussed in the next section), these are considered to be immaterial to wholesale
market outcomes.
4.2.1 RET outcomes

Under this policy scenario, LGC prices are set at the fixed $40/MWh (nominal) price level.
All LGCs created from existing accredited stations (and those considered committed)
continue to receive subsidies at this level through until 2030.
With the effective closure of the SRES, no new STCs are created after 2014 as shown in
Figure 30. Without the subsidy provided through the SRES, uptake of small scale systems is
lower when compared with the Reference case (see Figure 31). Uptake of solar PV and
SWHs are very similar to those under the Repeal case – the only difference being a slight
change to retail prices which flow through to improved paybacks for these technologies.
Figure 30 STC creation under SRES: Closed to New Entrants case

STC creation Change (Closed to New Entrants case - Reference case)
18,000 0
16,000 -2,000
14,000
-4,000
12,000
-6,000
STCs ('000)
STCs ('000)
10,000
-8,000
8,000
-10,000
6,000
-12,000
4,000
2,000 -14,000
0 -16,000
2015
2014
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Solar PV SWH Total (Reference case) Solar PV SWH
Note: STC creation based on generation (install) year.

Source: ACIL Allen
35
Figure 31 Energy generation/displacement from SRES: Closed to New Entrants case

Energy generation/displacement Change (Closed to New Entrants case - Reference case)
25,000 0
-200
20,000
-400
-600
15,000
GWh
GWh
-800
10,000
-1,000
-1,200
5,000
-1,400
0 -1,600
2015
2014
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Source: ACIL Allen
Under this scenario LRET compliance costs continue into the future based on LGCs created
from existing generators at the assumed fixed $40/MWh price. Annual LRET costs in 2015
fall by around 50% from $1.1 billion under the Reference case down to approximately $630
million under this scenario due to the lower prices and volumes. These costs continue to
decline in real terms over the period to 2030. Aggregate LRET compliance costs over the
period equates to $8.3 billion in real 2014 dollars.
In cumulative terms the changed arrangements result in a decrease in direct compliance
costs of $29.3 billion ($25 billion reduction in LRET; $4.5 billion reduction in SRES) as
shown in Figure 32.
Figure 32 RET compliance costs: Closed to New Entrants case

3,500 0.0
3,000 -7.9
-5.0
2,500 -10.0 -2.8

Real 2014 $m
Real 2014 $ billion
-25.0 -25.0
2,000
-15.0
1,500
-20.0
1,000
-25.0
500 -4.5 -4.5
-30.0
0
-35.0
2015-20 2015-30 2015-40
Source: ACIL Allen

Consumers face slightly higher retail bills under this scenario compared with the Repeal
case due to the requirement to continue to pay for LGCs at the fixed $40/LGC price level. In
36
present value terms, these additional costs equate to around $108/household in the period
to 2020 or $186/household in the period to 2030 for residential consumers.
As shown in the figure on the right hand side of Figure 33, this results in the change in costs
to households in present value terms being:
 $138 (1.8%) lower in the period to 2020
 $185 (1.2%) higher in the period to 2030
 $302 (1.6%) higher in the period to 2040.
Figure 33 Change in residential retail tariff components and NPV of annual bill: Closed to New Entrants
case
Change (Closed to New Entrants case - Reference case) NPV of change in residential retail bills (Closed
2.50 to New Entrants case - Reference case)
$1,000
2.00 $811
$800 $694
1.50
$600
Real 2014 c/kWh
1.00
0.50 NPV ($) $400 $302

$185
$200 $131
0.00
-0.50 $0
-1.00 ($200) ($138)
-1.50 ($269)
($400)
Network Wholesale energy ($600) ($509) ($509)

LRET SRES 2015-20 2015-30 2015-40
Source: ACIL Allen
37
5 Real 20% by 2020

The Real 20% scenario involves two significant changes to the current policy:
 Reducing the LRET 2020 target level from 41,000 GWh down to a level which
represents a ‘Real 20%’ of expected demand in this year
 Closing the SRES after 2020 and reducing the period of deeming for solar PV from 15
years down to 10 years from 1 January 2015 (deeming period is constant at 10 years
through to the end of 2020).
The ‘Real 20%’ level is calculated in accordance with the following formula which has been
provided by the Expert Panel:
RELRET ELIGIBLE + RESRES PV + REPRE−RET
% RE (ACTUAL 2020) =
ENEM + ESWIS + PVNEM&SWIS + EOTHER & OFF GRID
Where:
 RE(LRET eligible): Projected sent-out generation in 2020 from LRET-accredited renewable
power stations.
 RE(SRES PV): Projected electricity generation from SRES-eligible small-scale solar PV. For
consistency with the other components, this generation includes an allowance for
distribution and transmission losses so that energy produced at customer meters is
grossed up to a grid equivalent. The figure will be sensitive to any changes in the design
of the SRES.
 RE(Pre-RET): Power stations pre-dating the RET can only create LGCs for annual
generation (mainly hydro) above historical baselines set under the RET Regulations.
This component forecasts the ineligible (below-baseline) sent-out generation based on
historical levels and long-term forecasts of hydro resource availability.
 ENEM: This component represents ‘native electricity’ demand in the NEM as reported by
the Australian Energy Market Operator, with a small adjustment as explained below.
AEMO supplied the Panel with updated projections of native demand and demand
expected to be met by small-scale solar generation (subsequently published in June
2014). Native demand includes electricity demand that is met by all scheduled
generation (large power stations bidding into the NEM), semi-scheduled generators
(wind farms and other “intermittent” generators) and non-scheduled generators. Notably,
it does not include demand met by generation from small-scale solar, so ACIL Allen’s
modelled forecast of small-scale solar generation will be added separately (see the term
PVNEM&SWIS below). As ACIL’s modelled forecast is slightly lower than AEMO’s, E NEM
includes a small adjustment so that overall demand (native plus solar PV) will remain
consistent with AEMO.
 ESWIS: This component represents electricity demand in the SWIS as reported by WA’s
Independent Market Operator, with a small adjustment similar to that made in E NEM
above. IMO supplied the Panel with updated projections of electricity demand
(subsequently published in June 2014) and, separately, demand expected to be met by
small-scale solar. IMO’s projections are for sent-out demand, covering all significant
38
generation except self-generation, and small-scale solar PV (reported separately). As

such, for consistency this component will include an adjustment to reflect the difference
between IMO’s and ACIL Allen’s forecasts for solar PV.
 PV(NEM&SWIS): Projected generation from small-scale solar PV for both the NEM and
SWIS. As explained above, these projections will replace the projections AEMO and
IMO have provided and the demand numbers for each grid will be adjusted to maintain
consistency with AEMO’s and IMO’s overall demand forecasts.
 E(other & off grid): This component is ACIL’s estimate of electricity supplied on smaller grids
(including solar PV, other embedded generation) and self-generation. It also includes
self-generation (including LNG related) in the NEM states and WA, and off-grid energy
based on a report ACIL Allen provided to BREE in 2013.
The calculated proportion of renewable energy in 2020 is estimated to be:
41,000 + 9,673 + 16,148
26.2% =
190,844 + 20,734 + 9,513 + 34,207
Solving for a LRET 2020 target to yield a ‘Real 20%’ gives:

RELRET ELIGIBLE = 20% × (255,298) − (9,673 + 16,148)
RELRET ELIGIBLE = 25,238 GWH
For the modelled Real 20% scenario we have rounded this value to 25,500 GWh. Figure 34
and Table 4 provide the annual LRET targets under the Real 20% by 2020 scenario relative
to the Reference case. The annual targets follow a linear trajectory from 2015 to 2020.
As noted earlier, displacement by SRES-eligible solar water heaters has not been included
as part of the renewable energy component. ACIL Allen estimates that SWH systems will
displace approximately 2,870 GWh in 2014, rising to just under 3,500 GWh by 2020 under
this scenario. If displacement from SWH was to be added to both the renewable energy
component (the numerator), and to aggregate electricity demand (the denominator), the
Real 20% 2020 target would fall to around 22,400 GWh.
Figure 34 Real 20% by 2020 annual LRET targets
45,000
40,000
35,000
30,000
25,000
GWh
20,000
15,000
10,000
Reference case Real 20% case
5,000
0
2022
2029
2014
2015
2016
2017
2018
2019
2020
2021
2023
2024
2025
2026
2027
2028
2030
Note: Values exclude 850 GWh allowance for WCMG

Source: ACIL Allen
39
Table 4 Real 20% by 2020 annual LRET targets

Calendar year Reference case Real 20% by 2020 Change
2014 16,100 16,100 0

2015 18,000 17,667 -333
2016 20,581 19,233 -1,348
2017 25,181 20,800 -4,381
2018 29,781 22,367 -7,414
2019 34,381 23,933 -10,448
2020 41,000 25,500 -15,500
2021-30 41,000 25,500 -15,500
Source: ACIL Allen
Aside from the changes to the LRET and SRES described above, all other assumptions
remain unchanged.
5.2 Results
Figure 35 shows the projected change in wholesale prices under this scenario and relative
to the Reference case. The NEM price trace diverges from 2018, with the gap widening to
around $9/MWh (or 32%) above the Reference case in 2020. The price curves converge
again over time with only a small long-term differential of around $2/MWh.
The price trace for the ‘Other grids’ which is dominated by the SWIS is virtually unchanged
aside from a three year period 2027-29. This relates to altered timing of new entrant plant in
the SWIS.
Figure 35 Wholesale price outcomes: Real 20% case

100
80
Real 2014 $/MWh
60
40
20
0
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
Reference case NEM Real 20% case NEM

Reference case Other grids Real 20% case Other grids
Source: ACIL Allen
Dispatch outcomes sit roughly half way between the Reference and Repeal cases. The Real
20% has approximately 15,000 GWh less wind output, which is replaced by increased
output from incumbent generation (mainly black and brown coal-fired output) as shown in
Figure 36.
Changes in new entrant capacity developed are shown in Figure 37. Some 5,000 MW less
wind is built under this scenario. There are also some slight changes to the timing of utility
scale solar PV in regional grids, with the overall level of development largely unchanged.
The reduction in wind development is also associated with a small reduction in the
installation of gas-fired peaking capacity toward the end of the modelling period. Importantly,
as with the Repeal case, this reduction in wind capacity developed is not replaced with a
40
material amount of alternative generation capacity, with the market having sufficient capacity
from incumbent generators to meet its requirements.
Figure 36 Dispatch by fuel type: Real 20% case

Dispatch outcomes Change (Real 20% case - Reference case)
300,000 20,000
250,000 15,000
10,000
200,000
GWh sent-out
GWh sent-out
5,000
150,000
0
100,000 -5,000
-10,000
50,000
-15,000
0
-20,000
Source: ACIL Allen
Figure 37 New entry capacity: Real 20% case

New entrant capacity developed Change (Real 20% case - Reference case)
9,000 1,000
8,000
0
7,000
-1,000
6,000
5,000 -2,000
MW
MW
4,000 -3,000
3,000
-4,000
2,000
-5,000
1,000
0 -6,000
Source: ACIL Allen
Figure 38 shows the resulting changes in capital expenditure over the period to 2040. Under
the Reference case around $15.8 billion is spent on new capacity. Under the Real 20%
case, this drops to $7.7 billion (a 52% reduction). Expenditure on wind capacity accounts for
the vast majority of this, dropping by two thirds from $12 billion down to $4 billion.
Investment in new fossil fuelled capacity is virtually unchanged under this scenario.
Despite the scaling back of the LRET, market scheduled wind capacity more than doubles
under this scenario from around 2,370 MW in 2014 to 5,420 MW by 2020 as shown in
Figure 39. This remains around 5,000 MW below levels reached under the Reference case.
41
Over the period 2014-28, some mothballed brown coal capacity is re-introduced to the
market (approximately 500 MW more capacity relative to the Reference case) as a result of
the lower wind build. The market also sees an additional 1,300 MW of black coal-fired
capacity, mainly returning mothballed capacity, which includes some new incremental
capacity in the SWIS and also the refurbishment of a plant which was projected to retire
under the Reference case.
Figure 38 New entrant capital expenditure: Real 20% case

New entrant capital expenditure Change (Real 20% case - Reference case)
4,000 1,000
3,500 0
3,000
-1,000
Real 2014 $m
Real 2014 $m
2,500
-2,000
2,000
-3,000
1,500
-4,000
1,000
500 -5,000
0 -6,000
Source: ACIL Allen
Figure 39 Generation capacity by fuel type: Real 20% case

Installed capacity Change (Real 20% case - Reference case)
60,000 3,000
2,000
50,000
1,000
40,000 0
-1,000
MW
MW
30,000
-2,000
20,000 -3,000
-4,000
10,000
-5,000
0 -6,000
Source: ACIL Allen
5.2.1 RET outcomes

balance and the amount of LGCs banked. Under this scenario the LRET is fully met with just
enough certificates created to ensure the target is subscribed over the period to 2030.
Banked LGCs are exhausted after the 2020 surrenders occur and a shortfall occurs in the
42
following year (2021), however this only accounts for 2% of aggregate liabilities. This
shortfall is carried forward to future years and hence no penalties are payable. From the
model’s perspective this is a cheaper option compared with bringing forward capital
expenditure on a renewable project.
The LGC price for 2014 is approximately $42, which is around $14/LGC (26%) lower than
the Reference case. The modelling assumes that any changes to the policy are instantly
reflected in market prices. Due to the anticipated fall in LGC price, a policy change to a Real
20% level may have financial implications for existing renewable plants where LGC creation
is sold on a merchant basis.
trend down, reaching $30/LGC in 2030.
Figure 40 LRET supply-demand balance and price outcomes: Real 20% case
30,000 100
90
25,000
80
Real 2014 $/LGC
LGCs ('000)
20,000 70
60
15,000
50
10,000 40
30
5,000
20 Spot price
0 Tax-adjusted Penalty price
10
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Spot price (Reference case)

0
demand (including WCMG volumes) plus voluntary demand. Banked LGCs are as at end of calendar year. S pot LGC prices are those sufficient
Source: ACIL Allen
The lower LGC requirement under the Real 20% results in less wind being developed, with
wind producing around 16 million less LGCs annually from 2020 onwards as shown in
Figure 41. In aggregate wind remains the primary source of LGCs, accounting for over 75%
of total LGCs created to 2030, with the next largest fuel being large-scale solar at 7.4%.
43
Figure 41 LGC creation by source: Real 20% case

LGC creation by source Change (Real 20% case - Reference case)
30,000 2,000
0
25,000
-2,000
-4,000
20,000
LGCs ('000)
LGCs ('000)
-6,000
15,000 -8,000
-10,000
10,000
-12,000
-14,000
5,000
-16,000
0 -18,000
Bagasse Geothermal Hydro Landfill gas Bagasse Geothermal Hydro Landfill gas
Sewerage gas Solar WCMG Wind Sewerage gas Solar WCMG Wind
Wood Other Wood Other
Note: LGC creation by generation year. Other category includes food, crop and agricultural and municipal solid wastes.
Source: ACIL Allen
Figure 42 shows the projected STC creation from SGUs and SWHs over the period to 2030
under the Real 20% case. STC creation is well below the Reference case from 2015 to 2019
due to the reduction in the deeming period for solar PV from 15 years down to 10 years. The
outcome for 2020 is almost identical due to the deeming period being the same under the
two scenarios for this year. As the SRES is closed after 2020, no STCs are created from
2021 onwards under this scenario.
Figure 42 STC creation under SRES: Real 20% case

Change (Real 20% case - Reference case)
0
STC creation
18,000
16,000 -2,000
14,000
-4,000
12,000
STCs ('000)
-6,000
STCs ('000)
10,000
8,000
-8,000
6,000
-10,000
4,000
2,000 -12,000
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Solar PV SWH
Solar PV SWH Reference case (Total)

Source: ACIL Allen
The reduction to the upfront deeming period and early closure of the scheme only has a
relatively small impact upon projected uptake though, with aggregate generation
displacement volumes from these technologies being only around 750 GWh (or 4%) lower in
2030. Table 5 summarises LRET and SRES market outcomes under the Real 20%
scenario.
44
Figure 43 Energy generation/displacement from SRES: Real 20% case

Energy generation/displacement Change (Real 20% case - Reference case)
25,000 0
-100
20,000
-200
-300
15,000
GWh
GWh
-400
10,000
-500
-600
5,000
-700
0 -800
2015
2014
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Source: ACIL Allen
45
Table 5 LRET and SRES market outcomes: Real 20% case
Units 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030

Bank Balance Year Start ('000 LGCs) 26,144 22,576 19,137 15,246 10,704 5,487 1,148 0 0 0 0 0 0 0 0 0 0
Created ('000 LGCs) 15,051 16,748 17,863 18,778 19,670 22,114 26,872 26,647 26,883 26,994 27,056 27,126 27,209 27,305 27,399 27,499 27,582
Net Banking ('000 LGCs) -3,569 -3,439 -3,890 -4,542 -5,217 -4,340 -1,148 0 0 0 0 0 0 0 0 0 0
Bank Balance Year End ('000 LGCs) 22,576 19,137 15,246 10,704 5,487 1,148 0 0 0 0 0 0 0 0 0 0 0
LGC Shortfall ('000 LGCs) 0 0 0 0 0 0 0 523 287 176 636 331 138 502 102 0 0
LGC Shortfall% % 0% 0% 0% 0% 0% 0% 0% 2% 1% 1% 2% 1% 1% 2% 0% 0% 0%
LGC Shortfall Refunds ('000 LGCs) 0 0 0 0 0 0 0 0 0 0 523 287 176 636 331 329 412
STC creation from SGU ('000 STCs) 14,013 7,520 7,385 7,641 8,142 8,198 8,234 0 0 0 0 0 0 0 0 0 0
STC creation from SWH ('000 STCs) 1,410 1,599 1,733 2,006 2,415 3,570 3,421 0 0 0 0 0 0 0 0 0 0
Total STC creation ('000 STCs) 15,424 9,118 9,118 9,648 10,557 11,768 11,655 0 0 0 0 0 0 0 0 0 0
RET summary
PECs GWh 27,200 20,912 23,716 24,895 25,904 26,846 27,387 23,091 23,255 23,406 23,566 23,757 23,969 24,186 24,421 24,674 24,878
RPP % 9.87% 9.96% 10.65% 11.33% 12.10% 12.89% 13.64% 12.80% 12.71% 12.63% 12.54% 12.44% 12.33% 12.22% 12.10% 11.98% 11.88%
STP % 10.48% 3.17% 4.83% 5.05% 5.50% 6.12% 6.03% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
price x STP.PECs = Partial Exemption Certificates.
Source: ACIL Allen

RET compliance costs are significantly reduced under a Real 20% scenario as shown in
Figure 44. Annual costs are on average around 40% lower under this scenario, with
aggregate costs over the period to 2030 $17.5 billion lower. This is a result of the lower LGC
price and RPP each year. Costs under SRES cease after 2020 due to the revised end date
for the scheme.
Figure 44 RET compliance costs: Real 20% case

RET Compliance costs Cumulative RET compliance costs
3,500 0.0
-2.0 -4.7
3,000
-4.0
2,500 -0.6
Real 2014 $m
Real 2014 $ billion

-6.0
2,000 -8.0 -15.2 -15.2
-10.0
1,500
-12.0
1,000
-14.0
500 -16.0 -2.3 -2.3
0 -18.0
-20.0
2015-20 2015-30 2015-40
LRET SRES Reference case
LRET SRES
Source: ACIL Allen

Figure 45 presents the annual system resource costs broken down by category for modelled
grids. In aggregate resource costs for the Real 20% case total $112.6 billion in present
value terms over the period to 2040 using a discount rate of 7% pre-tax real ($121.9 billion
in the Reference case). Relative to the Reference case, resource costs associated with
large-scale energy under this scenario are $7.9 billion lower, while resource costs
associated with small-scale systems are $1.3 billion lower. This combines to a total saving
of $9.3 billion in NPV terms.
As with the Repeal case, the largest saving is associated with the reduction in capital
expenditure on new capacity (predominantly wind). The savings associated with the lower
capacity also result in savings to fixed O&M costs associated with servicing this equipment.
Offsetting these savings are higher variable O&M costs (principally fuel costs) and
refurbishment costs associated with increased utilisation of the existing generating fleet. The
large refurbishment cost differential in 2028 relates to the life extension of a large black coal-
fired generator in NSW which occurs under this scenario but was projected not to occur
under the Reference case due to lower wholesale price outlook.
47
Figure 45 Change in electricity generation sector resource costs: Real 20% case
Change (Real 20% case - Reference case) NPV of change (7%)
2,000
Total -9.3
1,000
0
SRES: SWHs -0.6
Real 2014 $m
-1,000
SRES: SGUs -0.7
-2,000
Total Large-scale -7.9
-5,000 Fixed O&M -2.1
New build -8.0

-12 -7 -2 3
NPV ($ billion)
SRES: SWHs
Source: ACIL Allen
The Real 20% scenario improves the profitability of fossil fuel generation across Australia
due to the higher wholesale prices which prevail under this scenario but not to the same
extent as complete repeal of the policy. Figure 46 shows the present value of calculated
EBITDA values by fuel type. The shift to the lower mandated LRET improves coal-fired
generators values by around $9.3 billion present values terms ($6.6 billion for black coal;
$2.7 billion for brown coal). Much of this improved profitability occurs between 2020 and
2030 due to the higher wholesale prices, but also continues out to 2040 due to the
increased generation volumes.
The aggregate value of wind declines by around $7 billion, noting most of this relates to the
lower volume of wind deployed. For existing wind farms the only decline they see is due to
any reduction in LGC prices which occur. However this may be offset by higher wholesale
electricity prices.
Figure 46 Electricity generation sector profitability by fuel type: Real 20% case
2,000
Geothermal 0.0
1,500
Solar -0.3
1,000
Real 2014 $m
Wind -7.0
500
Hydro 0.4
0
Baseload Gas 0.2
-500
Peaking Gas 0.0
-1,000
Brown coal 2.7
-1,500
Black coal 6.6
Black coal Brown coal Peaking Gas Baseload Gas -8 -6 -4 -2 0 2 4 6 8

generating costs
Source: ACIL Allen
48
Figure 47 shows emissions and intensity outcomes for the modelled grids. Under the Real
20% case emissions are higher, averaging a 15 Mt increase over the Reference case
between 2020 and 2030, declining slightly to average 13 Mt higher from 2030 to 2040.
Aggregate emissions over the period to 2040 are 322 Mt CO2-e higher under the Real 20%
case. This represents a 6.8% increase over the Reference case.
Under this scenario, grid emission intensity falls only slightly from current levels of
0.86 tonnes CO2-e/MWh sent-out, levelling out at around 0.84 tonnes CO2-e/MWh sent-out.
Figure 47 Change in electricity generation sector emissions: Real 20% case

Emissions intensity
18.0 0.90
16.0 0.88
14.0 0.86

12.0 0.84
10.0
Mt CO2-e
0.82
8.0
0.80
6.0
0.78
4.0
0.76
2.0
0.0 0.74
Reference case
-2.0 0.72
Real 20% case
0.70
Source: ACIL Allen

Retail prices to consumers are initially lower under the Real 20% scenario as a result of the
reduction in LGC prices and the lower RPP. The projected increase in wholesale electricity
prices more than offsets this reduction though from around 2018 onwards. Figure 48 shows
the change in residential retail bill components. The reduction in direct RET costs resulting
from the shift to a Real 20%, can be seen as the purple and yellow in the chart on the left.
The NPV of these reductions is shown as the green negative bar in the figure on the right.
These cost reductions are offset by the higher wholesale electricity costs. Higher wholesale
electricity costs also have flow-on effects for loss and retail cost components as these are
estimated as a percentage of non-network costs.
These offsetting cost increases are illustrated by the positive orange bars in the NPV chart
on the right. The net effect for retail consumers in NPV terms is shown as the blue bars.
Residential retail costs increase under this scenario (relative to the Reference case) by
around $178/household in NPV terms over the period to 2040. This equates to a 0.9%
increase in total expenditure on electricity over this period.
49
Figure 48 Change in residential retail tariff components and NPV of annual bill: Real 20% case
Change (Real 20% case - Reference case) NPV of change in residential retail bills (Real
1.50 20% case - Reference case)
$600
$484
$500
1.00 $424
$400
Real 2014 c/kWh
0.50 $300
NPV ($)
$178
$200
$114 $118
0.00 $100
$0
-0.50 ($24)
($100)
($200) ($139)
-1.00
($300)
($306) ($306)
Network Wholesale energy ($400)
LRET SRES 2015-20 2015-30 2015-40
Source: ACIL Allen
50
6 Real 30% by 2030

The Real 30% by 2030 scenario involves the following changes:
 Modifying the LRET 2030 target level to 30% of anticipated demand and extension of the
large-scale scheme to 2040. Annual targets from 2015 to 2030 follow a linear trajectory.
 No change to the current SRES with the scheme ending in 2030.
The ‘Real 30%’ level is calculated in accordance with the following formula which has been
determined by the Expert Panel22:
RELRET ELIGIBLE + RESRES PV + REPRE−RET
% RE (ACTUAL 2030) =
ENEM + ESWIS + PVNEM&SWIS + EOTHER & OFF GRID
The calculated proportion of renewable energy in 2030 (under the current RET) is estimated
to be:
41,000 + 15,966 + 16,148
26.0% =
201,199 + 24,351 + 15,736 + 39,809
Solving for a LRET 2030 target to yield a ‘Real 30%’ gives:

RELRET ELIGIBLE = 30% × (281,095) − (15,966 + 16,148)
RELRET ELIGIBLE = 52,214
For the Real 30% scenario we have rounded this value to 52,500 GWh.
As noted earlier, displacement by SRES-eligible solar water heaters has not been included
as part of the renewable energy component. ACIL Allen estimates that SWH systems will
displace approximately 2,870 GWh in 2014, rising to around 4,660 GWh by 2030 under this
scenario. If displacement from SWH was to be added to both the renewable energy
component (the numerator), and to aggregate electricity demand (the denominator), the
Real 30% 2020 target would fall to around 49,000 GWh.
22 Definitions of the terms in this equation are described in section 5.1.
51
Figure 49 Real 30% by 2030 annual LRET targets
60,000
50,000
40,000
GWh
30,000
20,000
10,000 Reference case Real 30% case
2027
2040
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
Source: ACIL Allen
Table 6 Real 30% by 2030 annual LRET targets
Calendar year Reference case Real 30% by 2030 Change
2014 16,100 16,100 0

2015 18,000 18,375 375
2016 20,581 20,650 69
2017 25,181 22,925 -2,256
2018 29,781 25,200 -4,581
2019 34,381 27,475 -6,906
2020 41,000 29,750 -11,250
2021 41,000 32,025 -8,975
2022 41,000 34,300 -6,700
2023 41,000 36,575 -4,425
2024 41,000 38,850 -2,150
2025 41,000 41,125 125
2026 41,000 43,400 2,400
2027 41,000 45,675 4,675
2028 41,000 47,950 6,950
2029 41,000 50,225 9,225
2030 41,000 52,500 11,500
2031-40 0 52,500 52,500
Source: ACIL Allen
6.2 Results
to the Reference case. The NEM price trace is higher from 2018 due to the lower level of
wind built in the period to 2020 under this scenario, but then falls below the Reference case
from around 2022 onwards, converging toward the end of the modelling horizon. This is due
to the overall higher level of renewables built in the period to 2030.
The price trace for the ‘Other grids’ which is dominated by the SWIS, is marginally higher
over the period to 2024, but then averaging around $13 below the Reference case to 2040.
52
Figure 50 Wholesale price outcomes: Real 30% case

100
80
Real 2014 $/MWh

60
40
20
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
Reference case NEM Real 30% case NEM
Reference case Other grids Real 30% case Other grids
prices for grids (NEM = NSW , QLD, SA, TAS and VIC; Other = SWIS, NWIS, DKIS and Mt Isa).
Source: ACIL Allen
As designed, the Real 30% results in a larger proportion of renewable generation in the
system as shown in Figure 51. Wind dispatch reaches around 37 TWh annually from 2029
onwards. Large scale solar grows throughout, approaching 8 TWh by the end of the
projection. Wind dispatch is lower relative to the Reference case through the period 2017-
2025 due to the lower LRET targets. This results in less brown coal capacity being
mothballed, explaining the increase in its output over this same period.
In the long-term the Real 30% target prompts additional wind and solar generation of up to
10.5 TWh by 2040. The additional output under the ‘Other’ category relates to the
development of some geothermal capacity with annual output of around 2,700 GWh. The
additional renewable output displaces energy primarily from black coal-fired generation.
Figure 51 Dispatch by fuel type: Real 30% case

Dispatch outcomes Change (Real 30% case - Reference case)
300,000 20,000
250,000 15,000
10,000
200,000
GWh sent-out
GWh sent-out
5,000
150,000
0
100,000
-5,000
50,000
-10,000
0 -15,000
Source: ACIL Allen
Changes in new entrant capacity developed are shown in Figure 52. The wind build early in
the period is deferred, but over time results in around 1,300 MW of additional wind, with the
increase occurring just prior to 2030. Around 2,200 MW of additional utility scale solar also
occurs under this scenario.
53
Around 4,000 MW of new gas-fired capacity occurs under this scenario, with a slight overall
increase in peaking capacity required. No new coal-fired plant is constructed under this
scenario.
The model also projects the development of around 350 MW of geothermal plant in the
SWIS in the period leading up to 2030. This capacity comes under the ‘Other’ category in
Figure 52.
Figure 52 New entry capacity: Real 30% case

New entrant capacity developed Change (Real 30% case - Reference case)
18,000 5,000
16,000 4,000
14,000 3,000
12,000 2,000
10,000 1,000
MW
MW
8,000 0
6,000 -1,000
4,000 -2,000
2,000 -3,000
0 -4,000
Source: ACIL Allen
the Reference case around $15.8 billion is spent on new capacity. Under the Real 30%
case, this increases to $17.6 billion (an 11% increase). Despite the increased wind capacity,
expenditure on wind actually declines slightly from $12 billion to $11.8 billion due to the
declining capital costs and the deferred build under this scenario. Solar see a large increase
from $1.8 billion to $3.1 billion. Around $1.4 billion is spent on the 350 MW of geothermal
capacity.
Renewable capacity accounts for around 38% of total installed capacity by 2040 as shown
in Figure 54. Most of the existing generation fleet remains in service but no new fossil
fuelled entry occurs under this scenario.
54
Figure 53 New entrant capital expenditure: Real 30% case

New entrant capital expenditure Change (Real 30% case - Reference case)
6,000 3,000
2,000
5,000
1,000
4,000 0
Real 2014 $m
Real 2014 $m
-1,000
3,000
-2,000
2,000 -3,000
-4,000
1,000
-5,000
0 -6,000
Source: ACIL Allen
Figure 54 Generation capacity by fuel type: Real 30% case

Installed capacity Change (Real 30% case - Reference case)
70,000 6,000
5,000
60,000
4,000
50,000 3,000
2,000
40,000
MW
MW
1,000
30,000
0
20,000 -1,000
-2,000
10,000
-3,000
0 -4,000
Source: ACIL Allen
6.2.1 RET outcomes

Under this scenario the LRET is extended to 2040. Whilst the target itself would be easily
modified through the legislation, there are a number of transitional issues which would need
to be resolved. Key amongst them is whether existing accredited generators would be able
to participate in the extended scheme as revenues post 2030 would represent windfall gains
to such generators.23 If new baselines for existing generators were to come into play post
2030, then the renewable build required to meet the targets would be greatly increased. For
this scenario it was assumed that no new baselines would be set and existing generators
could continue to create LGCs beyond 2030.
23 The same issue applied when the original MRET scheme was extended to 2030, however no adjustment was made.
55
Figure 55 provides the LRET supply-demand balance (left) and the projected LGC prices
(right) under the Real 30% scenario. The expanded target is fully met, although a large
shortfall occurs in 2018 and 2019 (5.1 million LGCs short in aggregate across these years).
As this amount of shortfall exceeds the 10% leeway provision, penalties are paid, however
an excess of LGCs created in subsequent years are surrendered to offset this shortfall and
penalties are fully refunded. The model deals with this situation with perfect foresight and
does not allow prices to rise toward the penalty level.
LGC prices are projected to be initially lower compared with the Reference case,
commencing at around $50, rising to $60 by 2018, before declining over the remainder of
the period, reaching a low of just above $10 by 2040. Due to the low wholesale electricity
prices throughout, renewables remain uncompetitive for entry in the absence of subsidies.
Figure 55 LRET supply-demand balance and price outcomes: Real 30% case
60,000 100
Spot price
50,000 Spot price (Reference case)
80
Real 2014 $/LGC

LGCs ('000)
40,000 70
60
30,000
50
20,000 40
30
10,000
20
0 10
0
Source: ACIL Allen
Figure 56 shows the LGC creation by fuel under this scenario. Wind remains dominant with
over 80% of LGCs created from 2014-2040. Solar accounts for 6.4% of the total, with other
technologies accounting for small proportions.
LGCs continue to be created post 2030 under this scenario which explains the large
increase in the last 10 years when compared with the Reference case.
56
Figure 56 LGC creation by source: Real 30% case

LGC creation by source Change (Real 30% case - Reference case)
60,000 60,000
50,000
50,000
40,000
40,000
30,000
LGCs ('000)
LGCs ('000)
30,000 20,000
10,000
20,000
0
10,000
-10,000
0 -20,000
Source: ACIL Allen
Under the Real 30% scenario, there is no change to SRES settings, and therefore outcomes
are virtually identical to those under the Reference case as shown in Figure 57 and Figure
58. Minor variations between the scenarios arise due to the slightly different retail price
outcomes which flow through to financial returns for small-scale systems.
The slightly lower retail prices in the longer-term under this case, results in a slightly lower
up-take of solar PV systems.
Figure 57 STC creation under SRES: Real 30% case

STC creation Change (Real 30% case - Reference case)
18,000 1,000
16,000 800
14,000 600
400
12,000
200
STCs ('000)
STCs ('000)
10,000
0
8,000
-200
6,000
-400
4,000 -600
2,000 -800
0 -1,000
2015
2014
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030

Source: ACIL Allen
57
Figure 58 Energy generation/displacement from SRES: Real 30% case

Energy generation/displacement Change (Real 30% case - Reference case)
25,000 200
0
20,000
-200
15,000
GWh
GWh
-400
10,000
-600
5,000
-800
0 -1,000
2015
2014
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Source: ACIL Allen
RET compliance costs are lower each year to 2028 under this scenario relative to the
Reference case as shown in Figure 59. This is due to the less steep trajectory of the LRET
(see Figure 49), with the Real 30% only exceeding the 41,000 GWh from 2025 onwards.
Also the lower LGC prices act to reduce direct compliance costs. SRES costs are
unchanged under this scenario with only minor differences in STC creation projected. In
aggregate, compliance costs are still higher when considering the period to 2040 by around
$6.2 billion.
Figure 59 RET compliance costs: Real 30% case

3,500 8.0
3,000 6.0
2,500 4.0
Real 2014 $m
Real 2014 $ billion
6.2
2,000 2.0
0.0 0.0 0.0

1,500
-3.1
-2.0
1,000
-6.8
-4.0
500
-6.0
0 0.0
-8.0
2015-20 2015-30 2015-40
Source: ACIL Allen
Table 7 provides the key figures for the LRET and SRES market outcomes in tabular form
out to 2030 (note that the LRET extends to 2040 under this scenario). Line items include
LGCs created and surrendered, banked certificates, shortfalls (if any) and other values
required for the calculation of RPP and STP values each year.
58
The Real 30% case sees shortfalls emerge in 2018-19, with the shortfall in 2019 being large
enough to case shortfall penalties to be paid. With excess certificates created in subsequent
years, these penalties are refunded to liable entities.
59
Table 7 LRET and SRES market outcomes: Real 30% case
Units 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030

Bank Balance Year Start ('000 LGCs) 26,144 22,576 18,428 13,253 6,714 0 0 0 0 0 0 0 0 0 0 596 0
Created ('000 LGCs) 15,051 16,748 17,995 18,906 19,839 25,939 33,655 36,254 37,248 38,245 40,520 42,795 45,070 47,345 50,216 51,299 51,435
Net Banking ('000 LGCs) -3,569 -4,147 -5,175 -6,539 -6,714 0 0 0 0 0 0 0 0 0 596 -596 0
Bank Balance Year End ('000 LGCs) 22,576 18,428 13,253 6,714 0 0 0 0 0 0 0 0 0 0 596 0 0
LGC Shortfall ('000 LGCs) 0 0 0 0 1,167 4,056 0 0 0 0 0 0 0 0 0 0 2,735
LGC Shortfall% % 0% 0% 0% 0% 4% 14% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 5%
LGC Shortfall Refunds ('000 LGCs) 0 0 0 0 0 0 1,385 2,559 1,278 0 0 0 0 0 0 0 0
STC creation from SGU ('000 STCs) 14,052 12,664 12,079 11,381 10,929 9,586 8,297 7,217 6,334 5,430 4,598 3,736 2,988 2,249 1,626 979 368
STC creation from SWH ('000 STCs) 1,410 1,599 1,733 2,006 2,415 3,570 3,421 3,469 3,702 2,793 2,921 2,116 1,571 1,119 815 543 231
Total STC creation ('000 STCs) 15,462 14,263 13,813 13,387 13,344 13,155 11,718 10,686 10,037 8,223 7,519 5,853 4,559 3,368 2,441 1,522 599
RET summary
PECs GWh 27,200 23,495 25,675 26,576 27,311 27,923 28,278 28,572 29,069 29,332 29,798 30,139 30,570 31,023 31,538 32,061 32,518
RPP % 9.87% 10.49% 11.53% 12.57% 13.70% 14.84% 15.94% 16.56% 17.64% 18.70% 19.76% 20.76% 21.73% 22.68% 23.60% 24.48% 25.40%
STP % 10.48% 6.04% 7.41% 7.08% 7.02% 6.89% 6.10% 5.53% 5.16% 4.21% 3.82% 2.95% 2.28% 1.67% 1.20% 0.74% 0.29%
price x STP.PECs = Partial Exemption Certificates. Table only shows results to 2030, even though scheme extends to 2040.
Source: ACIL Allen


Overall sector resource costs are virtually unchanged under this scenario, despite the larger
renewable build that occurs. This is due to the deferral of the build task through the slower
LRET ramp, allowing the market to take advantage of the capital cost reductions which are
projected to occur. A slower ramp under the LRET allows the market to determine the
optimal time to build new renewable capacity, with the trade-offs being deferral to take
advantage of lower capital costs and technology improvements in the future versus earlier
build to create more LGCs over the plant’s life. The largest component change in resource
costs is the variable O&M reductions which occur in the long-term due to reduced fossil fuel
consumption.
Figure 60 Change in electricity generation sector resource costs: Real 30% case
3,000
Total -0.1
2,000
1,000
SRES: SWHs 0.0
0
Real 2014 $m
SRES: SGUs -0.2

-1,000
-2,000 Total Large-scale 0.1
-4,000 Variable O&M -1.1
-5,000 Fixed O&M 0.6

-6,000 Refurbishment -0.1
New build 0.7

-2 -1 -1 0 1 1
NPV ($ billion)
SRES: SWHs
Source: ACIL Allen
Calculated NPVs of EBITDA measures are lower for all technologies excepting solar, which
sees a modest gain of $0.6 billion due to the larger amount of solar developed. The lower
values for wind are a result of the deferred developments and the lower LGC revenue
stream in present value terms.
61
Figure 61 Electricity generation sector profitability by fuel type: Real 30% case
2,000
Geothermal 0.7
1,500
1,000 Solar 0.6
500
Real 2014 $m
Wind -2.4
0
Hydro -0.8
-500
-1,000 Baseload Gas -0.2
-1,500 Peaking Gas -0.1

-2,000
Brown coal -1.2
-2,500
Black coal -1.9
Black coal Brown coal Peaking Gas Baseload Gas -3 -3 -2 -2 -1 -1 0 1 1

generating costs
Source: ACIL Allen
Emissions outcomes match the level of renewable development, being higher than the
Reference case in the period to 2026, then lower thereafter. Moving to a Real 30% results in
lower annual emissions in the long-term of around 10 Mt CO2-e less.
Figure 62 Change in electricity generation sector emissions: Real 30% case

Change (Real 30% case - Reference case) Emissions intensity
15.0 0.90
10.0
0.85
5.0
Mt CO2-e
0.80
0.0
-5.0 0.75
-10.0
0.70
Reference case
-15.0
Real 30% case
0.65
Source: ACIL Allen

Figure 63 provides the price components of the national average residential tariff. Moving to
the Real 30% results in a decrease in direct compliance costs due to the lower LGC prices
and RPPs in the period to 2025. This is illustrated by the reduction in costs to consumers
from the LRET components in the chart (there is no material change in SRES costs).
However, this benefit is reinforced by the projected lower wholesale electricity costs which
prevail under this scenario, shown by the blue bars. This also has flow on costs for network
losses and retail costs/margins as these are assumed to be a percentage of non-network
62
costs within the retail model. The net effect is lower overall tariffs through the period
modelled.
The chart on the right of Figure 63 shows the NPV of these changes over various time
frames. In the period 2015-20 the net effect is for a negligible change in retail bills for a
typical household. Over the period to 2030 and 2040, households are better off under this
scenario relative to the Reference case. This is despite the inclusion of direct LRET costs
after 2030 in the retail price stack. The changes in costs to households in present value
terms are:
 $297 (1.6%) lower in the period to 2040.
Figure 63 Change in residential retail tariff components and NPV of annual bill: Real 30% case
Change (Real 30% case - Reference case) NPV of change in residential retail bills (Real
1.00 30% case - Reference case)
$100
$57
0.50 $50
$0
Real 2014 c/kWh
0.00 ($50) ($17)

NPV ($)
($100) ($74) ($72)

-0.50 ($96)
($150)
($137)
-1.00 ($200)
($250) ($233) ($225)
-1.50
($300)
($297)
LRET SRES 2015-20 2015-30 2015-40
Source: ACIL Allen
63
7 50% of growth case

This scenario, which was modelled after the stakeholder workshop held in June, involves
moving the LRET away from fixed annual targets to floating targets, reset each year based
on forecast demand growth for the year ahead. Specifically, the LRET target would be
increased each year based on 50% of the anticipated growth in market-facing demand —
that is, demand growth net of that absorbed by behind the meter solar PV.24
The scenario also implements modifications to the SRES as follows:
 A reduction in deeming for SGUs to 10 years from 1 January 2015, with the deeming
period declining for both SGUs and SWHs by one year each year and the scheme
terminating at the end of 2020
 A reduction in the size eligibility of small generating units for inclusion under SRES
down from the current 100 kW to 20 kW (systems above 20 kW would be eligible for
the LRET).
Figure 64 shows the resulting annual targets under the core demand assumptions. Targets
rise each year, reaching 25,200 GWh by 2020 which are then held flat until 2030. Under
these demand assumptions, the scenario appears very similar to the Real 20% case, with
the only material difference being the changes to the SRES. The annual targets in practice
will be sensitive to the demand growth each year, with targets being lower under low
demand growth conditions and higher under high demand growth conditions.
Figure 64 50% Growth case annual LRET targets
45,000
40,000
35,000
30,000
25,000
GWh
20,000
15,000
10,000
Reference case 50% Growth case
5,000
0
2017
2014
2015
2016
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030

Source: ACIL Allen
Figure 65 presents the historical and projected solar PV capacity under the SRES broken
down by various size categories. Of the 4,220 MW of solar PV capacity installed up to
quarter 2 2014, only 126 MW (3% of the total installs) have been from systems above
20 kW. This is a growing install category for solar PV. ACIL Allen’s SRES projection model
24 Under this scenario LRET targets are not reduced if demand is expected to decline. That is the LRET target for the following
year is the maximum of the previous year’s target or a base year’s target plus 50% of the growth in energy (net of solar PV
energy increase).
64
shows this category growing although still only comprising a relatively small proportion of
overall installs into the future. It is difficult to assess the economics of such large systems as
the financial paybacks are very site specific and are highly dependent on the network tariffs
the host is on and the level to which these network costs can be avoided through the
consumption of energy from a rooftop PV system. In excluding these systems from the
SRES ACIL Allen has made the assumption that 50% of this capacity is still developed but
creates LGCs under the LRET. This additional supply source grows over time as
installations create LGCs annually and reachese around 0.5 million LGCs by 2030.
Figure 65 Historical and projected SGU install by size category
500
450
400
350
Capacity (MW)
20kW - 100kW
300
10kW - 20kW
250 8kW - 10kW
200 5kW - 8kW

2kW - 5kW
150
0kW - 2kW
100
50
0
2019
2010
2011
2012
2013
2014
2015
2016
2017
2018
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Calendar year
Source: ACIL Allen based analysis of CER data
7.2 Results
This section provides the results of the 50% Growth case. For the most part, wholesale
results are very similar (almost identical) to those under the Real 20% case, due to the
calculated targets based on the demand projection matching almost exactly the calculated
real 20% level by 2020.
to the Reference case. The NEM price trace diverges from 2018, with the gap widening to
around $9/MWh in 2020 (32% above the Reference case). The price curves converge
together again over time with only a small long-term differential of around $2/MWh.
The price trace for the ‘Other grids’ which is dominated by the SWIS is virtually unchanged
aside from a three year period 2027-29. This relates to altered timing of new entrant plant in
the SWIS.
As Figure 67 shows, dispatch outcomes sit roughly half way between the Reference and
Repeal cases and are virtually identical to those under the Real 20% scenario. The scenario
sees approximately 15,000 GWh of less wind output, which is replaced by increased output
from incumbent generation (mainly black and brown coal-fired output).
Changes in new entrant capacity developed are shown in Figure 68. Some 5,000 MW less
wind is built under this scenario. There are also some slight changes to the timing of utility
scale solar PV in regional grids, with the overall level of development largely unchanged.
The reduction in wind development is also accompanied by a small reduction in gas-fired
65
peaking capacity toward the end of the modelling period. Importantly, as with the Repeal
case, this reduction in wind capacity developed is not replaced with alternative generation
capacity, with the market having sufficient capacity from incumbent generators to meet its
requirements.
Figure 66 Wholesale price outcomes: 50% Growth case

100
80
Real 2014 $/MWh

60
40
20
0
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
Reference case NEM 50% Growth case NEM
Reference case Other grids 50% Growth case Other grids
Source: ACIL Allen
Figure 67 Dispatch by fuel type: 50% Growth case

Dispatch outcomes Change (50% Growth case - Reference case)
300,000 20,000
250,000 15,000
10,000
200,000
GWh sent-out
GWh sent-out
5,000
150,000 0
-5,000
100,000
-10,000
50,000
-15,000
0 -20,000
Source: ACIL Allen
66
Figure 68 New entry capacity: 50% Growth case

New entrant capacity developed Change (50% Growth case - Reference case)
9,000 1,000
8,000
0
7,000
-1,000
6,000
5,000 -2,000
MW
MW
4,000 -3,000
3,000
-4,000
2,000
-5,000
1,000
0 -6,000
Source: ACIL Allen
the Reference case around $15.8 billion is spent on new capacity. Under this scenario, this
drops to $7.7 billion (a 52% reduction). Expenditure on wind capacity accounts for the vast
majority of this, dropping by two thirds from $12 billion down to $4.1 billion. Investment in
new fossil fuelled capacity is virtually unchanged under this scenario.
Despite the scaling back of the LRET, market scheduled wind capacity more than doubles
under this scenario from around 2,370 MW in 2014 to 5,420 MW by 2020 as shown in
Figure 70. This remains around 5,000 MW below levels reached under the Reference case
though.
Over the period 2014-28, some mothballed brown coal capacity is re-introduced to the
market (approximately 500 MW increase relative to the Reference case) as a result of the
lower wind build. The market also sees an additional 1,300 MW of black coal-fired capacity
(mainly returning mothballed capacity), which includes some new incremental capacity in
the SWIS and also the refurbishment of a plant which was projected to retire under the
Reference case.
67
Figure 69 New entrant capital expenditure: 50% Growth case

New entrant capital expenditure Change (50% Growth case - Reference case)
4,000 1,000
3,500
0
3,000
-1,000
Real 2014 $m
Real 2014 $m
2,500
2,000 -2,000
1,500
-3,000
1,000
-4,000
500
0 -5,000
Source: ACIL Allen
Figure 70 Generation capacity by fuel type: 50% Growth case

Installed capacity Change (50% Growth case - Reference case)
60,000 3,000
2,000
50,000
1,000
40,000 0
-1,000
MW
MW
30,000
-2,000
20,000 -3,000
-4,000
10,000
-5,000
0 -6,000
Source: ACIL Allen
7.2.1 RET outcomes

Despite the slightly different annual targets to 2020, LRET outcomes are also virtually the
same as the Real 20% scenario.
balance and the amount of LGCs banked. Under this scenario the LRET is fully met with just
enough certificates created to ensure the target is subscribed over the period to 2030.
Banked LGCs are exhausted after the 2020 surrenders occur and a shortfall occurs in the
following year (2021), however this only accounts for 2% of aggregate liabilities. This
shortfall is carried forward to future years and hence no penalties are payable. From the
model’s perspective this is a cheaper option compared with bringing forward capital
expenditure on a renewable project.
68
The LGC price for 2014 is approximately $42, which is around $14/LGC (26%) lower than
the Reference case. The modelling assumes that any changes to the policy are instantly
reflected in market prices. Due to the anticipated fall in LGC price, a policy change to a 50%
Growth approach may have financial implications for existing renewable plants where LGC
creation is sold on a merchant basis.
trend down, reaching $30/LGC in 2030.
Figure 71 LRET supply-demand balance and price outcomes: 50% Growth case
30,000 100
90
25,000
80
Real 2014 $/LGC

LGCs ('000)
20,000 70
60
15,000
50
10,000 40
30
5,000
20 Spot price
10
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Spot price (Reference case)

0
Source: ACIL Allen
The lower LGC requirement under this scenario results in less wind being developed, with
wind producing around 16 million less LGCs annually from 2020 onwards as shown in
Figure 72. In aggregate, wind remains the primary source of LGCs though, accounting for
over 75% of total LGCs created to 2030, with the next largest fuel being large-scale solar at
7.4%.
69
Figure 72 LGC creation by source: 50% Growth case

LGC creation by source Change (50% Growth case - Reference case)
30,000 2,000
0
25,000
-2,000
-4,000
20,000
LGCs ('000)
LGCs ('000)
-6,000
15,000 -8,000
-10,000
10,000
-12,000
-14,000
5,000
-16,000
0 -18,000
Source: ACIL Allen
Owing to the changes in deeming and eligibility, SRES outcomes for this case are materially
different to the Real 20% case when compared against the Reference case, as shown in
Figure 73 and Figure 74 (compare also with Figure 42 and Figure 43). STC creation is
significantly lower from 2015 as a result of reduction in deeming available to both SGUs and
SWHs. Exclusion of above 20 kW systems also has some small impact on creation rates.
The scenario results in lower levels of STC creation in the period to 2015 to 2020 compared
with the Real 20% scenario (61.8 milion under the Real 20%; 39.9 million under the 50%
Growth case).
Despite these changes, uptake of these technologies is not significantly changed, with
aggregate generation displacement being only 2,000 GWh less by 2040. When compared
against the Real 20% scenario, energy displaced from small-scale technologies is lower by
around 1,400 GWh by 2030 and 1,800 GWh by 2040.
Figure 73 STC creation under SRES: 50% Growth case

STC creation Change (50% Growth case - Reference case)
18,000 0
16,000
-2,000
14,000
12,000 -4,000
STCs ('000)
STCs ('000)
10,000
-6,000
8,000
6,000 -8,000
4,000
-10,000
2,000
0 -12,000
2016
2023
2030
2014
2015
2017
2018
2019
2020
2021
2022
2024
2025
2026
2027
2028
2029
2015
2014
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030

Source: ACIL Allen
70
Figure 74 Energy generation/displacement from SRES: 50% Growth case

Energy generation/displacement Change (50% Growth case - Reference case)
25,000 0
20,000 -500
15,000 -1,000
GWh
GWh
10,000 -1,500
5,000 -2,000
0 -2,500
2016
2023
2014
2015
2017
2018
2019
2020
2021
2022
2024
2025
2026
2027
2028
2029
2030
2015
2014
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
Source: ACIL Allen
RET compliance costs are significantly reduced under the 50% Growth scenario as shown
in Figure 75. Annual costs are on average around 45% lower under this scenario, with
aggregate costs over the period to 2030 $17.3 billion lower. This is a result of the lower LGC
price and RPPs each year. Costs under the SRES cease earlier, in 2020, due to the
scenario’s revised end date In comparison, the compliance costs under the Real 20%
scenario were 40% lower than the Reference case with a reduction of $17.5 billion in the
period to 2030 as shown in Figure 44.
Figure 75 RET compliance costs: 50% Growth case

3,500 0.0
-2.0 -3.3
3,000
-4.0 -1.4
2,500
Real 2014 $m
Real 2014 $ billion
-6.0
-14.3 -14.3
2,000 -8.0
-10.0
1,500
-12.0
1,000
-14.0
500 -16.0 -3.1 -3.1
0 -18.0
-20.0
2015-20 2015-30 2015-40
Source: ACIL Allen
Table 8 provides the key figures for the LRET and SRES market outcomes in tabular form.
Line items include LGCs created and surrendered, banked certificates, shortfalls (if any) and
other values required for the calculation of RPP and STP values each year. Note that the
LRET continues to 2040, while the table only shows values to 2030.
71
Table 8 LRET and SRES market outcomes: 50% Growth case
Units 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030

Bank Balance Year Start ('000 LGCs) 26,144 22,576 19,397 14,194 7,894 1,630 0 0 0 0 0 0 0 0 0 0 0
Created ('000 LGCs) 15,051 16,800 18,051 18,971 19,967 24,947 26,888 26,546 26,764 26,817 26,877 26,943 27,023 27,103 27,196 27,293 27,373
Net Banking ('000 LGCs) -3,569 -3,179 -5,203 -6,300 -6,264 -1,630 0 0 0 0 0 0 0 0 0 0 0
Bank Balance Year End ('000 LGCs) 22,576 19,397 14,194 7,894 1,630 0 0 0 0 0 0 0 0 0 0 0 0
LGC Shortfall ('000 LGCs) 0 0 0 0 0 388 833 325 496 887 320 424 736 88 100 314 0
LGC Shortfall% % 0% 0% 0% 0% 0% 1% 3% 1% 2% 3% 1% 2% 3% 0% 0% 1% 0%
LGC Shortfall Refunds ('000 LGCs) 0 0 0 0 0 0 0 0 388 833 325 496 887 320 424 736 502
STC creation from SGU ('000 STCs) 14,040 6,797 5,813 5,187 4,684 3,887 3,234 0 0 0 0 0 0 0 0 0 0
STC creation from SWH ('000 STCs) 1,410 1,599 1,560 1,605 1,691 2,142 1,711 0 0 0 0 0 0 0 0 0 0
Total STC creation ('000 STCs) 15,450 8,396 7,373 6,792 6,375 6,029 4,944 0 0 0 0 0 0 0 0 0 0
RET summary
PECs GWh 27,200 20,399 23,620 24,572 24,974 25,273 25,323 22,928 23,092 23,241 23,400 23,590 23,801 24,016 24,249 24,500 24,703
RPP % 9.87% 9.82% 11.44% 12.33% 12.74% 13.05% 13.34% 12.64% 12.55% 12.47% 12.38% 12.29% 12.18% 12.07% 11.95% 11.83% 11.73%
STP % 10.48% 2.79% 3.91% 3.55% 3.31% 3.11% 2.53% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
price x STP.PECs = Partial Exemption Certificates.
Source: ACIL Allen


Figure 76 presents the annual system resource costs broken down by category for modelled
grids. In aggregate, resource costs for the 50% Growth case total $111.4 billion in present
value terms over the period to 2040 using a discount rate of 7% pre-tax real ($121.9 billion
in the Reference case). Relative to the Reference case, resource costs associated with
large-scale energy under this scenario are $7.8 billion lower, while resource costs
associated with small-scale systems are $2.7 billion lower. This combines to a total saving
of $10.5 billion in NPV terms.
The largest saving is associated with the reduction in capital expenditure on new capacity
(predominantly wind). The savings associated with the lower capacity also result in savings
to fixed O&M costs associated with servicing this equipment. Offsetting these savings are
higher variable O&M costs (principally fuel costs) and refurbishment costs associated with
increased utilisation of the existing generating fleet. The large refurbishment cost differential
in 2028 relates to the life extension of a large black coal-fired generator in NSW which
occurs under this scenario but was projected not to occur under the Reference case due to
lower wholesale price outlook.
Resource cost savings are higher under this scenario relative to the Real 20% scenario.
Virtually all of this difference relates to the lower expenditure on small-scale systems.
Figure 76 Change in electricity generation sector resource costs: 50% Growth case
Change (50% Growth case - Reference case) NPV of change (7%)
2,000
Total -10.5
1,000
0
SRES: SWHs -0.5
Real 2014 $m
-1,000
SRES: SGUs -2.2
-2,000
-5,000 Fixed O&M -2.1
New build -7.9

-15 -10 -5 0 5
NPV ($ billion)
SRES: SWHs
Source: ACIL Allen
The 50% Growth scenario improves the profitability of fossil fuel generation across Australia
due to the higher wholesale prices which prevail under this scenario but not to the same
extent as complete repeal of the policy. Figure 77 shows the present value of calculated
EBITDA values by fuel type. The shift to the lower mandated LRET improves coal-fired
generators values by around $9.1 billion present values terms ($6.5 billion for black coal;
$2.6 billion for brown coal). Much of this improved profitability occurs between 2020 and
2030 due to the higher wholesale prices, but also continues out to 2040 due to the
increased generation volumes.
The aggregate value of wind declines by around $6.5 billion, noting most of this relates to
the lower volume of wind deployed. For existing wind farms the only decline they see is due
73
to any reduction in LGC prices which occur. However this may be offset by higher wholesale
electricity prices.
Figure 77 Electricity generation sector profitability by fuel type: 50% Growth case
Change (50% Growth case - Reference case) NPV of change (7%)
2,000
Geothermal 0.0
1,500
Solar -0.3
1,000
Real 2014 $m
Wind -6.5
500
Hydro 0.4
0
Baseload Gas 0.2
-500
Peaking Gas 0.0
-1,000
Brown coal 2.6
-1,500
Black coal 6.5
Black coal Brown coal Peaking Gas Baseload Gas -8 -6 -4 -2 0 2 4 6 8

generating costs
Source: ACIL Allen
Figure 78 shows emissions and intensity outcomes for the modelled grids. Under the 50%
Growth case emissions are higher, averaging a 15.2 Mt increase over the Reference case
between 2020 and 2030, declining slightly to average 13.4 Mt higher from 2030 to 2040.
Aggregate emissions over the period to 2040 are 323 Mt CO2-e higher under the 50%
Growth case. This represents a 6.8% increase over the Reference case.
Under this scenario, grid emission intensity falls only slightly from current levels of
0.86 tonnes CO2-e/MWh sent-out, levelling out at around 0.84 tonnes CO2-e/MWh sent-out.
Figure 78 Change in electricity generation sector emissions: 50% Growth case

Change (50% Growth case - Reference case) Emissions intensity
18.0 0.90
16.0 0.88
14.0 0.86
12.0 0.84
10.0
Mt CO2-e
0.82
8.0
0.80
6.0
0.78
4.0
0.76
2.0
0.0 0.74
Reference case
-2.0 0.72
50% Growth case
0.70
Source: ACIL Allen
74

Retail prices to consumers are initially lower under the 50% Growth scenario as a result of
the reduction in LGC prices and the lower RPP. The projected increase in wholesale
electricity prices more than offsets this reduction though, from around 2018 onwards. Figure
79 shows the change in residential retail bill components. The reduction in direct RET costs
resulting from the shift to the 50% Growth scenario can be seen as the purple and yellow in
the chart on the left. The NPV of these reductions is shown as the green negative bar in the
figure on the right.
These cost reductions are offset by the higher wholesale electricity costs. Higher wholesale
electricity costs also have flow-on effects for loss and retail cost components as these are
estimated as a percentage of non-network costs.
These offsetting cost increases are illustrated by the positive orange bars in the NPV chart
on the right. The net effect for retail consumers in NPV terms is shown as the blue bars.
Residential retail costs increase under this scenario (relative to the Reference case) by
around $179/household in NPV terms over the period to 2040. This equates to a 0.9%
increase in total expenditure on electricity over this period.
Figure 79 Change in residential retail tariff components and NPV of annual bill: 50% Growth case
Change (50% Growth case - Reference case) NPV of change in residential retail bills (50%
1.50 Growth case - Reference case)
$600
$475
$500
1.00 $415
$400
Real 2014 c/kWh
0.50 $300
NPV ($)
$179
$200
$102 $119
0.00 $100
$0
-0.50 ($20)
($100)
($123)
($200)
-1.00
($300)
($296) ($296)
LRET SRES 2015-20 2015-30 2015-40
Source: ACIL Allen
75
8 Sensitivities
This chapter provides assumptions and results for the various sensitivities modelled for the
Reference and other policy cases.
8.1 Low demand

The Low demand sensitivities draw upon the energy and demand forecasts under the low
economic growth forecasts as published by AEMO/IMO. At the request of the Expert Panel,
ACIL Allen has also held energy constant in regional grids and for off-grid components.
Figure 80 shows a comparison of aggregate Australian electricity demand under the Low
sensitivities relative to the Reference case. Peak demands have been scaled down
accordingly in line with the reduction in energy (system load factors remain unchanged from
the Reference case).
Under the low demand sensitivity demand remains relatively flat over the period to 2040
(228.6 TWh 2014 declining to 226.4 TWh in 2040). This compares with the aggregate
growth under the Reference case to 2040 of 68.7 TWh (1% cumulative average growth
rate). Total Australian electricity demand is around 11.7% lower in 2020 and almost 25%
lower by 2040 under this scenario.
Under this demand outlook, the Reference case policy settings would result in renewables
accounting for 29.7% of Australia’s generation mix (26.3% under the core demand
assumptions).
Figure 80 Low energy growth comparison: Australia
Energy (TWh) Energy (TWh)

350 350
Reference Low case
300 300
301.3
281.0
250 250
Sent-out TWh
-1.7% -24.9%
Sent-out TWh
255.3 -11.7% -21.9%

232.6 228.6 225.5 226.4
200 200 219.5
150 150
100 100
Reference Low case
50 50
0 0
2014 2020 2030 2040
-50
Calendar year Calendar year
Note: Includes scheduled, non-scheduled, embedded, self-generation and off-grid elements

Source: ACIL Allen based on AEMO/IMO forecasts
The low demand sensitivity has been run against the Reference case; Repeal case; Real
20% case; Real 30% case and 50% Growth policy cases.
76
8.1.1 Reference case results

This section presents the results of the Reference case: Low demand sensitivity and
compares them against the core Reference case. Figure 81 provides a summary of the key
outputs from this case.
Wholesale pool prices are lower under this sensitivity. Prices do recover over time but at a
slower rate. For much of the period through to 2030 fossil fuelled plants are running with
minimal profit margins and significant plant withdrawal has occurred. This scenario
effectively illustrates the floor on pool price outcomes. If demand was to fall further than that
modelled, additional plant would be retired with virtually no change to price outcomes.
Dispatch outcomes under this sensitivity are materially different. Coal volumes are
significantly lower — a reduction in the order of 30% for both black and brown coal by 2030.
The impact of the RET displacing incumbent generation is magnified under these demand
conditions.
With the lower demand, less new entrant capacity is projected to enter. Utility-scale solar
development in regional grids is lower, owing to the lower wholesale prices and reduced
ability to accommodate as much intermittent generation. In order to meet the fixed LRET
targets however this is offset by slightly more wind development in the larger grids. Overall
fossil fuel capacity development is lower as the existing fleet is sufficient to meet demand for
the foreseeable future under this demand outlook. Aside from the initial wind and solar build,
only very small investments in gas-fired capacity occur throughout the remainder of the
period modelled.
The low demand sensitivity is a very different outlook for the industry. Under this scenario,
not only does the market not require new entry, but a significant supply-side adjustment is
required with up to 10,200 MW less fossil fuelled capacity in the system by 2040. The
majority of this occurs through permanent retirement of incumbent capacity (there was very
little new entry in the core Reference case), with 4,500 MW of this occurring before 2020.
RET outcomes
Despite the major upheaval in the industry for fossil fuelled generators, LRET outcomes are
not materially different under this sensitivity. The LRET is fully met and modelled LGC price
outcomes are around $5-$6 higher from around 2020 onwards, which offsets the slightly
lower wholesale energy prices.
LGC creation is similar to the core Reference case, with slightly higher creation from wind
offsetting the lower solar development. Under this scenario, the small amount of geothermal
capacity developed in the SWIS under the Reference case does not occur which explains
the reduction in LCG creation from this fuel type. Aggregate LRET compliance costs are
projected to be 6.3% higher under low demand conditions due to the higher LGC prices.
SRES outcomes are not materially different under the low demand sensitivity. Lower
wholesale electricity prices are offset by higher network cost components in retail prices,
with the net overall effect of retail prices being slightly higher. However this does not result
in a significantly higher uptake of solar PV when compared against the Reference case
(solar PV capacity is only 0.7% higher than the core assumptions by 2030). The reason for
this is much of the higher network costs incurred are levied in the model through fixed and
demand charges which do not affect solar PV paybacks as installation of PV systems will
not avoid these additional charges. Aggregate SRES compliance costs are around 1%
higher under these assumptions.
77
While LRET and SRES market outcomes are not significantly different, the lower demand
results in a large increase in RPP/STP values and therefore direct compliance costs for
users. This is due to the fixed GWh under the LRET and similar levels of STC creation
under SRES being spread across fewer MWh of electricity consumption by end users.
Resource costs and emissions

Resource costs are naturally lower under this sensitivity, with cost savings from fuel and
variable O&M costs resulting from the lower amount of electricity delivered to consumers.
The lower level of aggregate operational capacity also results in savings associated with
fixed O&M costs. Refurbishments which occur in the early 2030’s do not occur under this
sensitivity with around a $1.1 billion saving in NPV terms. Overall resource costs are $16
billion lower over the period to 2040.
The figure also shows the impact of the lower demand conditions upon generator EBITDA
outcomes and valuations. Coal-fired plant bear the brunt of the lower market demand
conditions, with the black coal fleet worth around $8.3 billion less (a 45% reduction) and
brown coal falling by $3.0 billion (a 42% reduction). The lower demand also affects other
technologies negatively. To some extent renewables are insulated from value reductions as
their volumes through LRET are guaranteed and to the extent that lower energy prices
prevail, higher LGC prices will tend to offset this (energy prices and LGC prices are
inversely related).
Emissions are significantly lower under this sensitivity. By 2020 emissions are 30.2 Mt CO2-
e lower (18% lower than the Reference case); by 2040 emissions are 60 Mt CO2-e lower
(32% lower than the Reference case). Modelled emissions intensity also falls due to the
closure of some high emission intensity stations by between 0.05 and 0.07 tonnes CO2-
e/MWh sent-out. Notably the change in emission reductions compared to the Reference
case are driven largely by less demand for electricity rather than renewable generation
displacing carbon emitting generation.
Retail prices
Retail price outcomes under the low demand sensitivity are higher by up to 1.5 cents per
kWh. Lower wholesale prices are more than offset by higher per unit network costs. Direct
RET costs are also higher on a per kWh basis due to the higher RPP and STP values which
result under a low demand scenario.
78
Figure 81 Sensitivity results: Reference case: Low demand (change from Reference case)
100
Wholesale price outcomes Change in dispatch
10,000
90
80 0
70
Real 2014 $/MWh
-10,000
60
GWh sent-out
-20,000
50
40 -30,000
30 -40,000
20
-50,000
10
0 -60,000
-70,000
Reference case NEM
Reference case: Low demand NEM
Reference case Other grids Black coal Brown coal Peaking Gas Baseload Gas
Reference case: Low demand Other grids Hydro Wind Solar Other
NPV of change in resource costs Change in emissions

10.0
Total -16.0
0.0
-10.0
SRES: SWHs 0.0
SRES: SGUs 0.1 -20.0

Mt CO2-e
Total Large-scale -16.1 -30.0
Unserved Energy 0.0 -40.0
Variable O&M -9.2 -50.0

Fixed O&M -5.0
-60.0
Refurbishment -1.1
-70.0
New build -0.7
-20 -15 -10 -5 0 5

NPV ($ billion)
Change in LRET outcomes Change in SRES outcomes

Aggregate SRES compliance
Aggregate LRET compliance 1.0%
cost to 2040 (real 2014 $)
6.3%
costs to 2040 (real 2014 $)
STC creation (SWH) 0.0%

Change in present value of LGC
3.9%
prices
STC creation (SGU) 1.4%
Change in utility scale solar PV

-24.1%
capacity developed by 2030 Capacity of new Solar PV
0.7%
installations to 2030
Change in wind capacity Capacity of new Solar PV

1.3% 1.3%
developed by 2030 installations to 2020
-30% -20% -10% 0% 10% 0.0%0.2%0.4%0.6%0.8%1.0%1.2%1.4%1.6%

Percentage change Percentage change
79
NPV of change in generator EBITDA (7%) NPV of change in residential retail bills
Geothermal 0.0 $700

$631
$605
Solar -0.5 $600
$513
Wind -0.6 $488
$500
NPV ($)
Hydro -1.4
$400
Baseload Gas -0.8

$300 $255
$221
Peaking Gas -0.2
$200
Brown coal -3.0 $117 $117
$100
$35
Black coal -8.3
$0
-10 -8 -6 -4 -2 0 2015-20 2015-30 2015-40
NPV ($ billion) Direct RET costs Other costs Net cost
Note: All NPV values calculated using a 7% real discount rate over the period 2015-40
Source: ACIL Allen
8.1.2 Other policy cases

Figure 82 through to Figure 85 summarise the outcomes for the other policy cases when run
using the low demand input assumptions. These results are compared against the
Reference case: Low demand results. Notable results are summarised as follows:
 Repeal case (Figure 82): wholesale price increases from repeal of the RET are
smaller due to the overall supressed market. Total resource cost saving from repeal
totals $15.3 billion ($14 billion under core assumptions). Emissions savings are slightly
lower due to the lower emission intensity under the Reference case: Low demand
outcomes. Consumers do not see any future LGC and STC compliance costs under
Repeal of the schemes. Small-scale solar PV uptake is projected to continue, albeit at
lower installation rates. New solar PV capacity developed is around 24% lower by
2020 and 15% lower by 2030 when compared with the Reference case under low
demand conditions. Retail consumers are better off from repeal under these
assumptions due to the much smaller wholesale price increase and the higher STP
and RPP values under the Reference case: Low demand scenario.
 Real 20% case (Figure 83): wholesale prices for the SWIS and other regional grids are
virtually unchanged from the Reference case; NEM regions see only a slight uplift in
wholesale energy prices as a result of scaling back the RET (the market responds by
retiring slightly less capacity). Resource costs are $10 billion lower from scaling the
RET back to a real 20% level. Emissions are initially up to 18 Mt higher (2021) as a
result of the deferral of some brown coal retirements, with this stabilising at around
12 Mt over the longer-term. EBITDA results suggest that while wind suffers
significantly from the Real 20% adjustment to targets, very little of this loss is
transferred to fossil fuelled generators through higher valuations. LGC prices are
around 13% lower when compared against the Reference case, with wind
development around 65% lower due to the lower targets. The modifications to the
SRES result in large reductions in direct compliance costs (down around 54% in
aggregate) whilst only having a relatively minor impact upon installations (down around
9% in aggregate). Retail prices to consumers are lower as a result of scaling targets
back with the net effect being an estimated $276/household over the long-term on a
NPV basis. This is in contrast to the standard assumptions set.
80
 Real 30% case (Figure 84): wholesale price outcomes are generally slightly lower for
NEM regions, whilst being initially higher for other grids and lower in the long-term. The
Real 30% targets result in the deferral of wind build, with more wind capacity from
around 2026 onwards. Moving to a Real 30% has little effect on generator values;
emissions are up 12 Mt higher in 2018-19 due to a deferral of brown coal withdrawal,
with the additional wind entry in the long term resulting in lower emissions (by around
10 Mt) from the displacement of black coal. Calculated NPVs of EBITDA’s are
generally negative for all fuel types (wind due to the deferred build). Both wind and
utility scale solar see increases in development with capacity 27% and 45% higher
respectively. Despite LGC prices being lower (down 4.8% in NPV terms), the extension
of the scheme results in aggregate compliance costs being around 19% higher in
aggregate over the period to 2040. Retail prices are projected to be lower by an
estimated $285/household over the long-term on a NPV basis. This results in a slightly
lower level of solar PV uptake (down around 2%).
 50% Growth case (Figure 85): Under the 50% growth case, LRET annual targets do
not exceed the 2014 level of 16,100 GWh (excluding WCMG volumes). This is due to
the negative demand growth overall experienced throughout Australia. As a result, no
new wind developments occur, with wholesale results being much the same as the
Repeal case (shown in Figure 82). Regional grids still see some utility scale solar PV
capacity developed however this is later than under the Reference case. Under this
scenario, the modelled LGC prices fall to zero as the existing renewable fleet plus
banked LGCs is sufficient to meet LCG demand. Whilst modelled in this manner, this
clearly would have implications for transitional arrangements which would need to
provide ‘a reasonable LGC revenue stream’ for existing accredited power stations. For
this reason, the retail price outcomes under this scenario have included a fixed
$40/MWh (nominal) LGC price stream – similar to that employed under the Closed to
New Entrants scenario. This results in the NPV of LGC prices being around 44% lower
when compared against the Reference case. The modifications to SRES result in
lower levels of solar PV uptake (down 21% by 2020 and 19% by 2030) and large
reductions in SRES compliance costs (down 70%). Retail consumers are projected to
see lower bills as a result with the lower compliance costs more than offsetting the
increase in wholesale energy costs.
Figure 82 Sensitivity results: Repeal case: Low demand (change from Reference case: Low demand)
100
40,000
90
80 30,000
70
Real 2014 $/MWh
20,000
60
GWh sent-out
50 10,000
40
0
30
20 -10,000
10
0 -20,000
-30,000
Repeal case: Low demand NEM
Reference case: Low demand Other grids Black coal Brown coal Peaking Gas Baseload Gas
Repeal case: Low demand Other grids Hydro Wind Solar Other
81

30.0
Total -15.3
Total Small-scale -3.5 25.0
SRES: SWHs -1.1 20.0
SRES: SGUs -2.4
Mt CO2-e
15.0
10.0
Unserved Energy 0.0
Variable O&M 3.8 5.0
Fixed O&M -3.2 0.0

Refurbishment 0.4
-5.0
New build -12.7
-20 -15 -10 -5 0 5

NPV ($ billion)
Aggregate SRES compliance -100.0%

Aggregate LRET compliance
-100.0% cost to 2040 (real 2014 $)
STC creation (SWH) -100.0%

-100.0%
prices
STC creation (SGU) -100.0%

-83.1%
-15.1%
Change in wind capacity

-100.0% Capacity of new Solar PV
developed by 2030 -24.2%
-120%-100% -80% -60% -40% -20% 0% -120% -100% -80% -60% -40% -20% 0%
NPV of change in generator EBITDA NPV of change in residential retail bills
Geothermal 0.0 $600

$375
Solar -1.4 $400
$239
Wind -14.3 $200

NPV ($)
$0
Hydro 0.2
($200) ($69)
Baseload Gas 0.8
($400)
Peaking Gas -0.1
($402) ($420)
($600) ($471)
Brown coal 1.6 ($556)
($800)
Black coal 5.0 ($795) ($795)
($1,000)
-20 -15 -10 -5 0 5 10 2015-20 2015-30 2015-40
Source: ACIL Allen
82
Figure 83 Sensitivity results: Real 20% case: Low demand (change from Reference case: Low demand)

100
20,000
90
80 15,000
70
Real 2014 $/MWh
10,000
60
GWh sent-out
5,000
50
40 0
30 -5,000
20
-10,000
10
0 -15,000
-20,000
Real 20% case: Low demand NEM
Real 20% case: Low demand Other grids Hydro Wind Solar Other

20.0
Total -10.0

15.0
SRES: SWHs -0.6
SRES: SGUs -1.2 10.0

Mt CO2-e
Unserved Energy 0.0 5.0
Variable O&M 1.2

0.0
Fixed O&M -1.4
Refurbishment 0.1
-5.0
New build -8.0
-15 -10 -5 0 5
NPV ($ billion)

Aggregate LRET compliance costs -53.8%
-41.4% cost to 2040 (real 2014 $)
to 2040 (real 2014 $)

-13.1%
prices

0.0%
capacity developed by 2030
Capacity of new Solar PV
-9.4%

-70%-60%-50%-40%-30%-20%-10% 0% -60% -50% -40% -30% -20% -10% 0%

83
Geothermal 0.0 $100

$60
$50
Solar -0.2
$0
Wind -6.9
($50) ($16)
NPV ($)
Hydro -0.2 ($100)
($105)
($150)
Baseload Gas 0.1 ($137)
($200)
Peaking Gas -0.1
($250)
($242)
Brown coal 0.2 ($300) ($276)
Black coal 0.4 ($350)

($337) ($337)
($352)
($400)
-8 -6 -4 -2 0 2 2015-20 2015-30 2015-40
Source: ACIL Allen
Figure 84 Sensitivity results: Real 30% case: Low demand (change from Reference case: Low demand)
100
20,000
90
80 15,000
70
Real 2014 $/MWh
10,000
60
GWh sent-out
50 5,000
40
0
30
20 -5,000
10
0 -10,000
-15,000
Real 30% case: Low demand NEM
Real 30% case: Low demand Other grids Hydro Wind Solar Other

15.0
Total 0.5
SRES: SWHs 0.0

5.0
SRES: SGUs -0.2
Mt CO2-e
Total Large-scale 0.7 0.0
Unserved Energy 0.0

-5.0
Variable O&M -0.4
Fixed O&M 0.6 -10.0
Refurbishment -0.2
-15.0
New build 0.7
-1.0 -0.5 0.0 0.5 1.0

NPV ($ billion)
84

Aggregate LRET compliance -1.6%
18.9% cost to 2040 (real 2014 $)

-4.8%
prices

45.2%
-2.2%

27.7% Capacity of new Solar PV
-10% 0% 10% 20% 30% 40% 50% -2.5% -2.0% -1.5% -1.0% -0.5% 0.0%
Geothermal 0.0 $0
Solar 0.6
($50) ($37)
Wind -2.0
($76) ($75)
($100)
NPV ($)
Hydro -0.2 ($100)

($113)
($150)
Baseload Gas -0.1
($167)
Peaking Gas 0.0 ($200)
($211)
Brown coal -0.9
($250)
Black coal -1.7
($267)
($300) ($285)
-3 -2 -2 -1 -1 0 1 1 2015-20 2015-30 2015-40
Source: ACIL Allen
Figure 85 Sensitivity results: 50% Growth case: Low demand (change from Reference case: Low demand)
100
30,000
90
80 20,000
70
Real 2014 $/MWh
60 10,000
GWh sent-out
50
40 0
30
-10,000
20
10
-20,000
0
-30,000
50% Growth case: Low demand NEM
50% Growth case: Low demand Other grids Hydro Wind Solar Other
85

30.0
Total -15.1
SRES: SGUs -2.5
Mt CO2-e
15.0
10.0
Unserved Energy 0.0
Fixed O&M -3.3 0.0

Refurbishment 0.3
-5.0
New build -12.7
-20 -15 -10 -5 0 5

NPV ($ billion)

-75.7% cost to 2040 (real 2014 $)

-45.6%
prices

-73.1%
-18.8%

-120%-100% -80% -60% -40% -20% 0% -80%-70%-60%-50%-40%-30%-20%-10% 0%

Geothermal 0.0 $600

$401
Solar -1.4 $400
$259
Wind -14.1 $200
NPV ($)
Hydro 0.2
$0
Baseload Gas 0.8 ($74)

($200)
($180)
Peaking Gas -0.1 ($256)
($400) ($330) ($323)
Brown coal 1.4
($600)
Black coal 4.7 ($582) ($582)
($800)
-20 -15 -10 -5 0 5 10 2015-20 2015-30 2015-40
Source: ACIL Allen
86
8.2 High demand

The High demand sensitivities draw upon the energy and demand forecasts under the high
economic growth forecasts as published by AEMO/IMO. As instructed by the Expert Panel,
ACIL Allen has also increased energy growth in regional grids and for off-grid components
by one percentage point above growth rates in the Reference case. Figure 86 shows a
comparison of aggregate Australian electricity demand under the High demand sensitivities
relative to the Reference case. Peak demands have also been scaled up accordingly in line
with the increase in energy (system load factors remain unchanged from the Reference
case).
Under the high demand sensitivity, demand grows strongly over the period to 2040, growing
by an aggregate of 144.7 TWh, which equates to a cumulative average growth rate over the
period of 1.9% per annum (1.0% under the Reference case). In the current market
environment it is difficult to see how such strong electricity demand growth could eventuate
and would require a significant reversal of trends over the last few years. Developments
such as faster uptake of electric vehicles could be a potential source of significant demand
growth. Under this scenario total Australian electricity demand is estimated to be 7.6%
higher in 2020 and over 26% higher by 2040.
Under this demand outlook, the Reference case policy settings would results in renewables
accounting for 24.4% of Australia’s generation mix (26.3% under the core demand
assumptions).
Figure 86 High energy growth comparison: Australia
Energy (TWh) Energy (TWh)

400 450
Reference High case
350 400 26.1%
350 379.9
300 15.6%
Sent-out TWh
Sent-out TWh
300 7.6% 324.9

250 301.3
250 1.1% 281.0
274.6
200 255.3
200 232.6 235.2
150
150
100
100
Reference High case
50
50
0 0
2014 2020 2030 2040
-50
Calendar year Calendar year
Note: Includes scheduled, non-scheduled, embedded, self-generation and off-grid elements

Source: ACIL Allen based on AEMO/IMO forecasts

This section presents the results of the Reference case: High demand sensitivity and
compares them against the core Reference case.
As one would expect, wholesale pool prices are higher under this sensitivity as shown in
Figure 87. After the renewable capacity induced by the RET is absorbed, prices trend
towards new entry levels. For the SWIS and regional grids this occurs by around 2025 with
prices remaining relatively flat in real terms at around $85/MWh. This occurs later in the
NEM (around 2030) with prices lower at around $60/MWh reflecting lower new entrant
costs.
87
Under the high demand growth conditions, output from coal-fired plant increases over time.
An additional 57 TWh of coal-fired generation occurs relative to the core assumptions. Grid
based demand exceeds 300 TWh in aggregate by 2040 under this scenario.
An additional 7,000 MW of capacity is developed over the core Reference case in this
sensitivity, with around 3,800 MW of this being black coal. Around 3,500 MW of additional
gas-fired capacity is also developed primarily to meet the increased peak demand
requirements from the system. Resulting capital expenditure on new entrant capacity is
therefore higher. There is virtually no change in the level or timing of renewable deployment
under this scenario.
With the additional coal volumes, emissions rise at a relatively fast rate (17 Mt higher in
2020; 53 Mt higher by 2040), with much of this a result of the development of new coal-fired
capacity. These higher wholesale prices, coupled with higher volumes results in higher
valuations for all fuel types with the exception of utility solar PV which sees slightly less
development occur (higher wholesale prices in NEM regions induces more wind).
The higher wholesale electricity prices result in lower LGC prices (11% lower in NPV terms)
and a reduction in LRET compliance costs (14% lower). Slightly less wind and solar PV
capacity is developed, with this being offset by a small amount of geothermal capacity
modelled to occur in the SWIS (130 MW). Conversely, SRES compliance costs are higher
(+5.5%) due to the higher level of solar PV uptake (+4.4% by 2020; +8.5% by 2030). This is
due to the higher retail prices making solar PV more financially attractive for consumers.
Retail prices are higher under this sensitivity. Lower network costs (costs spread over more
MWh), is offset by the higher wholesale prices relative to the core Reference case. Direct
RET costs are slightly lower, which is the result of two elements: lower LGC prices and also
lower RPP/STP values due to the higher energy growth.
Figure 87 Sensitivity results: Reference case: High demand (change from Reference case)
100
80,000
90
70,000
80
70 60,000
Real 2014 $/MWh
60 50,000
GWh sent-out
50 40,000
40
30,000
30
20,000
20
10 10,000
0 0
-10,000
Reference case NEM
Reference case: High demand NEM
Reference case: High demand Other grids Hydro Wind Solar Other
88

60.0
Total 16.1
Total Small-scale 0.9 50.0
SRES: SWHs 0.0

40.0
SRES: SGUs 0.9
Mt CO2-e
Total Large-scale 15.2 30.0
Unserved Energy 0.0

20.0
Variable O&M 8.8
Fixed O&M 1.9 10.0
Refurbishment 1.0
0.0
New build 3.5
0 5 10 15 20
NPV ($ billion)

Aggregate LRET compliance costs 5.5%
-14.3% cost to 2040 (real 2014 $)
to 2040 (real 2014 $)

-11.3%
prices

-8.0%
8.5%

-2.8% 4.4%
-16%-14%-12%-10%-8% -6% -4% -2% 0% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0%
Geothermal 0.7 $1,000

$850
Solar -0.3 $751
$800
Wind 1.1 $600

$600
$501
NPV ($)
Hydro 2.7
$400
Baseload Gas 1.5
Peaking Gas 1.1 $200 $154

$125
Brown coal 6.2

$0
($99) ($99)
($200)
-5 0 5 10 15 20 25 2015-20 2015-30 2015-40
Source: ACIL Allen
89

Figure 88 and Figure 89 present results from applying the High demand sensitivity to the
Repeal and 50% Growth policy scenarios respectively. These figures compare results
against the Reference: High demand scenario. Notable results are summarised as follows:
 Repeal case (Figure 88): wholesale price increases from repeal of the RET are larger
due to the higher level of demand growth in the market. Prices move quickly to new
entry levels from early next decade which is a significant increase from current price
levels. Resource cost saving from repeal totals $11.4 billion ($14 billion under core
assumptions). Emissions savings are slightly lower at around 20 Mt per annum due to
the lower emission intensity under the Reference case: High demand outcomes.
Despite the higher black energy prices, no new wind developments occur, with utility
scale solar PV also seeing a large reduction in development in the absence of
subsidies. Small-scale solar PV is projected to be around 17% lower by 2020 and
5.5% lower by 2030. Retail consumers are worse off from repeal under these
assumptions due to the large wholesale price increase which result. Also, the STP and
RPP values are lower under this scenario as costs are spread across more MWh.
 50% Growth case (Figure 89): Under the 50% growth case LRET targets increase to
around 33,500 GWh by 2020. Under the scenario settings, targets are held flat at this
level until 2030. As a result around 2,000 to 2,500 MW less wind capacity is developed
in order to reach the targets. Wholesale prices are projected to be slightly higher in
NEM regions throughout most of the period. Prices in other grids rise faster from 2020
but return to the same price path from 2025. Under this scenario aggregate resource
cost savings total $5.2 billion; emissions are around 6 Mt per year higher. Coal-fired
generation is around $9.1 billion better off relative to the Reference case. LGC prices
are around 9% lower in NPV terms and less wind and utility scale solar PV capacity is
developed, down 25% and 21% respectively. LRET compliance costs are 23% lower
when compared against the Reference case. New small-scale solar PV capacity
installed is around 16% lower by 2030, with SRES compliance costs around 69%
lower. Consumer retail bills are projected to be around $282/household higher in NPV
terms over the period to 2040.
Figure 88 Sensitivity results: Repeal case: High demand (change from Reference case: High demand)
100 Wholesale price outcomes Change in dispatch

40,000
90
80 30,000
70
Real 2014 $/MWh
20,000
60
GWh sent-out
50 10,000
40
0
30
20 -10,000
10
0 -20,000
-30,000
Repeal case: High demand NEM
Reference case: High demand Other grids Black coal Brown coal Peaking Gas Baseload Gas
Repeal case: High demand Other grids Hydro Wind Solar Other
90

25.0
Total -11.4
SRES: SWHs -1.1

15.0
SRES: SGUs -1.4
Mt CO2-e
Total Large-scale -8.8 10.0
Unserved Energy 0.0

5.0
Variable O&M 5.5
Fixed O&M -4.5 0.0
Refurbishment 0.0
-5.0
New build -9.8
-15 -10 -5 0 5 10
NPV ($ billion)

-100.0% cost to 2040 (real 2014 $)

Change in present value of LGC -100.0%
prices

-86.1%
-5.5%

-105%-100% -95% -90% -85% -80% -75% -120% -100% -80% -60% -40% -20% 0%
Geothermal -0.7 $1,500

$1,276
Solar -1.4 $1,064

$1,000
Wind -13.9 $697
$485
NPV ($)
Hydro 1.6 $500 $340
Baseload Gas 1.7 $2

$0
Peaking Gas 0.3
Brown coal 8.2 ($500) ($338)
($579) ($579)
Black coal 19.5
($1,000)
-20 -15 -10 -5 0 5 10 15 20 25 2015-20 2015-30 2015-40
Source: ACIL Allen
91
Figure 89 Sensitivity results: 50% Growth case: High demand (change from Reference case: High demand)

10,000
90
8,000
80
6,000
70
Real 2014 $/MWh
60 4,000
GWh sent-out
50 2,000
40 0
30 -2,000
20 -4,000
10 -6,000
0 -8,000
-10,000
50% Growth case: High demand NEM
Reference case: High demand Other grids Black coal Brown coal Peaking Gas Baseload Gas
50% Growth case: High demand Other grids Hydro Wind Solar Other

8.0
Total -5.2
7.0
6.0
SRES: SWHs -0.5
5.0
SRES: SGUs -2.3
Mt CO2-e
4.0
3.0
Unserved Energy 0.0
2.0
Variable O&M 1.8
1.0
Fixed O&M -1.3
0.0
Refurbishment 0.1
-1.0
New build -3.1
-6 -4 -2 0 2 4
NPV ($ billion)

Aggregate LRET compliance costs
-23.2% cost to 2040 (real 2014 $)
to 2040 (real 2014 $)

-9.1%
prices

-21.3%
-16.0%

-30% -25% -20% -15% -10% -5% 0% -80%-70%-60%-50%-40%-30%-20%-10% 0%

92
Geothermal -0.7 $500

$442
Solar -0.3 $400 $352
Wind -2.1 $282

$300
NPV ($)
Hydro 0.6 $192
$200
Baseload Gas 0.4 $75

$100
Peaking Gas 0.1 $13
$0
Brown coal 3.0
($100) ($63)
Black coal 6.1
($200) ($160) ($160)
-4 -2 0 2 4 6 8 2015-20 2015-30 2015-40
Source: ACIL Allen
8.3 Shadow carbon price

The modelling in the reference and other scenarios has been undertaken on a basis that
reflects the Government’s policy to repeal the carbon tax legislation and meet Australia’s
2020 emissions reduction target of 5 per cent compared to the level in 2000 through other
means. However, a number of stakeholders have expressed interest in understanding the
effect of policies that could potentially be introduced after 2020 to meet future emissions
reduction goals. While the nature of any such policies is unknown, they can be modelled
using a shadow carbon price.
For the carbon price sensitivity, ACIL Allen has adopted a carbon price which takes as a
starting point the current European emission permit prices, projecting them forward at an
assumed cost of carry. The shadow carbon price sensitivity will represent the impacts of
policies to reduce emissions beyond 2020, with prices linked to projected European prices.
Figure 90 shows historical permit spot prices in Europe over the last four and a half years.
Prices have declined over this period due to abatement targets being relatively easily
reached as a result of the downturn in economic activity in Europe. Whilst there are
prospects for Government intervention in the market which may increase permit prices, we
assume that the prospects of this are already factored into the current spot price.
93
Figure 90 Historical European emissions permit spot prices
18
16
14
Euro/tonne CO2
12
10
8
6
4
2
0
2010 2011 2012 2013 2014
Note: Spot prices determined as weighted average of trades during each trading day
Source: ACIL Allen based on data from the Intercontinental Exchange
Figure 91 shows the average prices for future year delivery of EU permits. Forward prices
escalate at the cost of carry which is influenced by available interest rates in Europe. These
prices imply current interest rates of 4-5% (nominal).
Figure 91 Average European emissions permit forward prices
6.60
6.44
6.40
Euro/tonne CO2
6.20 6.13
6.00
5.85
5.80
5.63
5.60
5.40
5.20
Dec-14 Dec-15 Dec-16 Dec-17
Note: Average of prices for forward delivery over Jan-May 2014

Source: ACIL Allen based on data from the Intercontinental Exchange
ACIL Allen has projected these prices forward beyond 2017 at an assumed cost of carry of
5.5% nominal (approximately 3% real assuming European inflation of 2.5%). This price
series is then converted into Australian dollars and forms the basis for shadow carbon prices
under this sensitivity. These are shown in Figure 92. Using this methodology, the projected
shadow carbon price in 2021 is around $11.30/tonne CO2-e in nominal dollars ($9.50/tonne
in Real 2014 dollars). Prices in subsequent years increase by approximately 5.5% nominal
(3% real).
94
Figure 92 Assumed domestic shadow carbon prices
35
30
25
A$/tonne CO2
20
15
10
5
0
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
Nominal Real 2014 $
Source: ACIL Allen

This section presents the results of the Reference case: Carbon sensitivity and compares
them against the core Reference case. Figure 93 summarises the results which shows the
impact of introducing a shadow carbon price from 2021 onwards.
The introduction of a shadow carbon price increases wholesale prices in NEM regions by an
average of $10/MWh in 2021, with this rising to around $16/MWh by 2040. Price increases
in the ‘Other grids’ are around half this amount, which is a result of gas-fired generation
playing a much bigger role in these regions. Some degree of fuel switching occurs, largely
from brown coal to lower emission intensity technologies as shown in the change in dispatch
chart within Figure 93.
The wind build under this scenario is slightly delayed, with the model projecting better
profitability later due to the carbon price. LGC prices are lower (down 8.3% in NPV terms)
resulting in lower LRET compliance costs (down 10.8%). Small-scale solar PV installations
are unchanged in the period to 2020, but around 6% higher in the period to 2030. A small
increase in SRES compliance costs occurs as a result.
Resource costs under this sensitivity are around $1 billion higher which is due to the
utilisation of generation which has higher marginal operating costs (gas-fired generation and
lower emission intensity, higher cost coal). Overall emissions are around 2 Mt per year lower
over that achieved in the Reference case.
Coal-fired profitability is lower, with EBITDA NPVs increasing for renewable plant by around
the same amount. Retail prices are higher due to the wholesale cost increase.
Retail prices are higher due to the pass through of carbon through to wholesale electricity
prices. In NPV terms, household see prices rise by $468/household in the period to 2030
and by $884/household in the period to 2040. The lower compliance costs from the RET are
only a small offsetting factor to these price increases.
95
Figure 93 Sensitivity results: Reference case: Carbon (change from Reference case)

120
4,000
100 3,000
2,000
Real 2014 $/MWh
80
1,000
GWh sent-out
60 0
40 -1,000
-2,000
20
-3,000
0 -4,000
-5,000
Reference case NEM
Reference case: Carbon NEM
Reference case: Carbon Other grids Hydro Wind Solar Other

3.0
Total 0.9
2.0
1.0
SRES: SWHs 0.0
0.0
SRES: SGUs 0.5
Mt CO2-e
-1.0
-2.0
Unserved Energy 0.0
-3.0
Variable O&M 0.4
-4.0
Fixed O&M 0.0
-5.0
Refurbishment 0.0
-6.0
New build 0.0
-0.5 0.0 0.5 1.0

NPV ($ billion)

-10.8% cost to 2040 (real 2014 $)
to 2040 (real 2014 $)

-8.3%
prices

-0.5%
5.9%

1.2% Capacity of new Solar PV
developed by 2030 0.0%
-12%-10% -8% -6% -4% -2% 0% 2% -1% 0% 1% 2% 3% 4% 5% 6% 7%

96
Geothermal 0.0 $1,000 $931

$884
Solar 0.0
$800
Wind 1.2
$600 $515
NPV ($)
Hydro 1.0 $468
$400
Baseload Gas 0.2
Peaking Gas 0.0 $200
Brown coal-1.8 $7
$0
Black coal -0.6 ($14) ($7)
($47) ($47)
($200)
-2 -2 -1 -1 0 1 1 2 2015-20 2015-30 2015-40
Source: ACIL Allen

The carbon sensitivity was also applied to the Repeal and Closed to new entrants policy
cases, with results for the former shown in Figure 94. Results are compared against the
Reference: Carbon scenario. Notable results are summarised as follows:
 Repeal case (Figure 94): wholesale price increases from repeal of the RET are around
$11/MWh over the period to 2030 for NEM regions and $15/MWh for other grids. Over
the longer-term prices tend to converge however. Resource cost saving from repeal
totals $14.2 billion ($14 billion under core assumptions). Emissions savings are only
very marginally higher compared with the core assumptions. Over the period to 2040
this scenario results in an additional 4.8 Mt in abatement. This indicates that the RET
and the shadow carbon price modelled working in tandem would yield higher
abatement than either acting alone, but in cost terms would be less efficient than either
operating in isolation. NPVs of EBITDA values are only slightly different to those under
the core assumptions. No new large-scale renewable developments occur, aside from
solar PV in regional grids. New small-scale solar PV capacity is around 20% lower by
2020, however the higher retail prices results in installations ‘catching up’ relative to
the Reference case and being only 9% lower by 2030. Consumer retail bill impacts are
slightly more negatively affected from repeal of the RET due to the larger wholesale
price impacts.
 Closed to new entrants case: Wholesale market results under this scenario are
identical to the Repeal case above and hence are not shown. 25 The fixed $40 LGC
price modelled to support existing large-scale renewable generators results in an
increase to household retail bill NPV of $108/household in the period to 2020 and $186
in the period to 2030.
25 A small change to the up-take of small scale technologies is observed but considered immaterial to wholesale market
outcomes.
97
Figure 94 Sensitivity results: Repeal case: Carbon (change from Reference case: Carbon)

120
40,000
100 30,000
Real 2014 $/MWh
80 20,000
GWh sent-out
60 10,000
40 0
20 -10,000
0 -20,000
-30,000
Reference case: Carbon NEM
Repeal case: Carbon NEM
Reference case: Carbon Other grids Black coal Brown coal Peaking Gas Baseload Gas
Repeal case: Carbon Other grids Hydro Wind Solar Other

30.0
Total -14.2
SRES: SGUs -1.8

Mt CO2-e
15.0
10.0
Unserved Energy 0.0
Fixed O&M -4.0 0.0

Refurbishment 0.6
-5.0
New build -12.3
-20 -15 -10 -5 0 5 10

NPV ($ billion)

-100.0% cost to 2040 (real 2014 $)

prices

-87.1%
-9.3%

-105% -100% -95% -90% -85% -80% -120% -100% -80% -60% -40% -20% 0%
98
Geothermal 0.0 $1,000

$817
$800 $703
Solar -1.6
$600
Wind -14.7
$400
NPV ($)
Hydro 1.2 $152 $186
$200 $73
Baseload Gas 1.3
$0
Peaking Gas 0.1 ($200)

($201)
($353)
($600)
Black coal 11.9
($631) ($631)
($800)
-20 -15 -10 -5 0 5 10 15 2015-20 2015-30 2015-40
Source: ACIL Allen
8.4 High capital costs

Wind and solar capital costs are key inputs into the RET modelling and subject to significant
uncertainty. In recent years, wind capital costs have actually increased when examined in
real $/kW installed terms; however improvements in size, blade design and yield have
resulted in cost reductions on a levelised cost of energy or LRMC basis. Solar PV has seen
dramatic cost declines over the last 10 years as production scaled up from a very low base.
The Reference case already includes a continuation of this trend into the future (see section
A.4 for discussion on assumed capital costs).
This High capital cost sensitivity considers a situation in which the expected cost declines
built into the core assumptions set do not eventuate to the same degree or are delayed.
Wind
Wind capital costs have been changed through simple modifications against the Reference
case of +10% initially, rising to +15% by 2016-17. The final capital costs used within the
High sensitivity is shown in Figure 95. Note that the costs in the Reference case are the final
price taking into account foreign and local learning rates, exchange rate movements and
wage costs.
Figure 95 High wind capital costs
3,000
Reference High
2,396
2,500 2,284 2,211 2,237
2,164
Real 2014 A$/kW installed
1,882 1,923 1,945

2,000
1,500
1,000
500
0
2014 2020 2030 2040
Calendar year
99
Solar
As of March 2014 the national average cost of a residential solar installation is estimated to
be $2,560/kW excluding GST and excluding the STC rebate. The cost of solar installations
consists of a local technology component, a foreign technology component and a local
labour component. For future cost projections each of these components is indexed by the
projected cost reduction factors sourced from the 2013 Australian Energy Technology
Assessment (AETA). While the cost indices of foreign and local equipment are assumed to
decline, the local labour cost component is assumed to increase in these projections as
wages are projected to increase in real terms. The Reference case capital costs are
provided in Figure 96 for residential and commercial rooftop systems as well as for fixed flat
plate utility-scale solar PV.
It should be noted that utility scale PV attracts a higher capital cost initially due to increased
costs associated with transformer and network connection, which offset the economies of
scale.26 This capital cost differential is projected to reduce over time due to the higher
proportion of imported materials in the overall build relative to small-scale installations which
conversely have a high domestic labour component.
Other solar technologies such as solar thermal (parabolic trough, central receiver and
compact linear Fresnel) follow a similar decline path but are omitted from the chart for
clarity.
Figure 96 Solar PV capital costs: Reference case
3,000
2,524 2,602 Residential roof top solar
2,500 Commercial roof top solar

2,181
2,026 Utility scale PV fixed
1,908
2,000 1,750 1,747
1,509 1,540
1,500
1,164
1,005 1,002
1,000
500
0
2014 2020 2030 2040
Calendar year
Source: ACIL Allen based on AETA learning rates
In the high system cost sensitivity we assume the cost reductions projected in the central
scenario are achieved later (up to a 10 year delay in achieving the same cost level as the
Reference case). In the high system cost sensitivity the installed cost of residential solar is
assumed to be $2,471/kW in 2020 and $1,747/kW by 2040 in real terms.
26 The most recent large-scale solar PV development is the AGL Nyngan/Broken Hill project which is expected to cost around
$450 million for a 155 MW (AC) project (approximately $2,900/kW) and is due for completion in mid-2015.
100
Figure 97 Solar PV capital costs: High case

Residential roof top solar
3,000
2,524 2,602 2,527 Commercial roof top solar
2,471

2,500 2,181 Utility scale PV fixed
2,135
2,026
1,908
2,000 1,750 1,747
1,509 1,540
1,500
1,000
500
0
2014 2020 2030 2040
Calendar year
Source: ACIL Allen based on AETA learning rates

This section presents the results of the Reference case: High capital cost sensitivity and
compares them against the core Reference case. Figure 98 presents these results.
Most results are broadly unchanged from the Reference case core assumptions. Wholesale
prices are almost identical for NEM regions, whilst there is reduction in energy prices in the
‘Other grids’. This is due to the development of geothermal capacity in the SWIS (320 MW in
2020) which, being baseload, supresses wholesale prices significantly more than the
corresponding amount of energy from wind. The geothermal capacity displaces an amount
of utility scale solar PV in regional locations (–480 MW) and also wind capacity in the
SWIS (–615 MW).
Owing to the higher capital costs for wind and solar, resource costs are higher by around
$1.9 billion in NPV terms. Although there are some year-on-year changes, aggregate
emissions outcomes are largely unchanged over the period.
Despite the higher capital costs for the primary renewable technologies, the LRET remains
fully satisfied, although LGC prices are around $7/MWh higher on average. The present
value of projected LGC prices is 11.8% higher which is slightly less than the change in wind
capital costs within the assumptions. Aggregate LRET compliance costs are 13.7% higher,
reflecting the higher LGC price series.
High capital costs for small scale solar PV also results in a fall in capacity installed, with
around 10% less capacity installed in the period to 2020 and just under 16% less by 2030.
This results in less STCs being created (13.2% less than under core assumptions), and as a
result, lower compliance costs for consumers by around 9.7% ($0.5 billion less).
EBITDA outcomes are mostly positive for most technologies. Wind sees are an increase of
$0.7 billion, with much of this due to the higher LGC prices projected flowing through to
existing wind (new wind will largely be unchanged due to the higher LGC revenues offsetting
the higher capital costs). Solar declines due to the reduction in development which occurs.
Other technologies are a result from slight changes in back prices within individual regions.
NPV of retail bills are projected to be around $80-$90 higher. The higher LGC prices are the
principle mechanism by which retail prices rise in this sensitivity.
101
Figure 98 Sensitivity results: Reference case: High capital cost (change from Reference case)

5,000
100 4,000
3,000
Real 2014 $/MWh
80
2,000
GWh sent-out
60 1,000
0
40
-1,000
-2,000
20
-3,000
0 -4,000
-5,000
Reference case NEM
Reference case: High capital costs NEM
Reference case: High capital costs Other grids Hydro Wind Solar Other

1.5
Total 1.9

1.0
SRES: SWHs 0.0
SRES: SGUs 0.1 0.5

Mt CO2-e
Variable O&M -0.2

-0.5
Fixed O&M 0.0
Refurbishment -0.1
-1.0
New build 2.1
-1 0 1 2 3
NPV ($ billion)

13.7% cost to 2040 (real 2014 $)

11.8%
prices

-36.8%
-15.8%

-40% -30% -20% -10% 0% 10% 20% -20% -15% -10% -5% 0%
102
Geothermal 1.6 $100

$89
$90
Solar -0.2 $81
$80
Wind 0.7
$70
NPV ($)
Hydro 0.2 $60 $56 $56
$50
Baseload Gas 0.1
$38
$40 $33
Peaking Gas -0.1
$30 $25
$19 $19
Brown coal 0.4 $20
Black coal -0.1 $10
$0
-0.5 0.0 0.5 1.0 1.5 2.0 2015-20 2015-30 2015-40
Source: ACIL Allen

Figure 99 and Figure 100 present results from applying the High capital cost sensitivity to
the Repeal and 50% Growth policy scenarios respectively. These figures compare results
against the Reference: High capital cost scenario. Notable results are summarised as
follows:
 Repeal case (Figure 99): As the change to capital costs only affects wind and solar
technologies, the wholesale market outcomes are the same as those under the core
Repeal case (in the absence of the RET no new large-scale renewables are
developed).27
 50% Growth case (Figure 100): Wholesale market results under this scenario are
similar to those under the core assumptions. As a sensitivity analysis (rather than a full
scenario), the LRET targets have not been modified, although the higher capital costs
would result in a slightly lower uptake of solar PV (new capacity installed is 16.3%
lower by 2030), which in turn would increase the LRET 2020 slightly. Resource costs
are slightly higher at $11.4 billion ($10.5 billion under the core assumptions). NPV of
residential bills are slightly lower compared with the core scenario, as a result of the
lower level of uptake of small-scale solar PV, particularly as above 20 kW systems
have been excluded from SRES under these scenario settings.
27 A small change to the up-take of small scale technologies is observed but considered immaterial to wholesale market
outcomes.
103
Figure 99 Sensitivity results: Repeal case: High capital cost (change from Reference case: High capital
cost)
120
40,000
100
30,000
Real 2014 $/MWh
80 20,000
GWh sent-out
60 10,000
40 0
20 -10,000
0 -20,000
-30,000
Repeal case: High capital NEM
Reference case: High capital costs Other grids Black coal Brown coal Peaking Gas Baseload Gas
Repeal case: High capital Other grids Hydro Wind Solar Other

30.0
Total -15.8
SRES: SGUs -1.9

Mt CO2-e
15.0
10.0
Unserved Energy 0.0
Fixed O&M -3.9 0.0

Refurbishment 0.8
-5.0
New build -14.6
-20 -15 -10 -5 0 5 10

NPV ($ billion)

-100.0% cost to 2040 (real 2014 $)

prices

-79.7%
-10.9%

-120%-100% -80% -60% -40% -20% 0% -120% -100% -80% -60% -40% -20% 0%
104
Geothermal -1.6 $1,000

$760
$800 $652
Solar -1.7
$600
Wind -14.9
$400
NPV ($)
Hydro 0.8 $200 $101
$26
$0
Baseload Gas 1.3
($200) ($81)
Peaking Gas 0.1
($400) ($285)
($386)

($734) ($734)
($1,000)
-20 -15 -10 -5 0 5 10 15 2015-20 2015-30 2015-40
Source: ACIL Allen
Figure 100 Sensitivity results: 50% Growth case: High capital cost (change from Reference case: High
capital cost)
120
20,000
100 15,000
Real 2014 $/MWh
80 10,000
GWh sent-out
5,000
60
0
40
-5,000
20 -10,000
0 -15,000
-20,000
50% Growth case: High capital NEM
Reference case: High capital costs Other grids Black coal Brown coal Peaking Gas Baseload Gas
50% Growth case: High capital Other grids Hydro Wind Solar Other

18.0
Total -11.4
16.0
SRES: SGUs 10.0

-2.1
Mt CO2-e
8.0
6.0
Fixed O&M -2.1 0.0

-2.0
Refurbishment 0.5
-4.0
New build -9.5
-15 -10 -5 0 5
NPV ($ billion)
105

-42.1% cost to 2040 (real 2014 $)
to 2040 (real 2014 $)

-13.9%
prices

-46.0%
-16.3%

-70%-60%-50%-40%-30%-20%-10% 0% -80%-70%-60%-50%-40%-30%-20%-10% 0%
Geothermal -1.1 $500

$418
$400 $368
Solar -0.9
$300
Wind -6.2
$200
NPV ($)
Hydro 0.2 $81 $101

$100 $51
Baseload Gas 0.6
$0
Peaking Gas 0.0 ($100) ($54)
($200) ($135)
Brown coal 2.1
($300)
Black coal 6.1
($317) ($317)
($400)
-8 -6 -4 -2 0 2 4 6 8 2015-20 2015-30 2015-40
Source: ACIL Allen
8.5 Permanent retirements

In response to new renewable capacity entering under the RET, it is expected that some
incumbent fossil fuelled plant will withdraw capacity from the market due to market revenues
not covering avoidable costs. Withdrawal of capacity can take a number of forms:
 Partial unit withdrawal or de-rating
 Weekend/overnight de-commitment
 Seasonal mothballing during off-peak months
 Temporary mothballing of units (rotating operational units in a multi-unit station)
 Long-term mothballing (cold storage)
 Permanent retirement of units/power stations.
Under the scenarios in which significant volumes of renewable energy capacity enters the
market, ACIL Allen has withdrawn plant from the market so as to ensure that all remaining
plant are operating on a sustainable commercial basis (net revenues for operating plant are
covering avoidable costs). Over time as market conditions improve, this capacity is
reintroduced to service. This could be considered the minimum market response expected
106
by incumbent generators. This outcome reflects the expected outcome in a workably

competitive generator sector, which historically the NEM has shown itself to be.
There is potential for the incumbent market response to renewable entry to be larger. This is
particularly the case where the supressed profitability of incumbents results in further
consolidation amongst generator portfolios (a larger portfolio has greater incentive for
capacity withdrawal, as it has more remaining capacity still operating to benefit from the
resulting higher pool prices). Under conditions of decreased competition, oligopolistic
behaviour within the generation sector could result in a larger market response.
This sensitivity tests the possibility of more capacity being withdrawn (relative to the
Reference case) and for that capacity not to be reintroduced to the market in later years
based on engineering and technical constraints around long-term mothballing.

This section presents the results of the Reference case: Permanent retirement sensitivity.
Under this sensitivity a total of around 2,400 MW of additional capacity is withdrawn over the
period 2017-2021 (roughly 6% of the fleet), with this capacity not returned to service for the
remainder of the modelling period.28
Figure 101 presents these results and compares them against the core Reference case.
With the greater capacity withdrawal wholesale price outcomes are well above the core
Reference case, averaging around $8/MWh higher for the NEM and $5/MWh for the ‘Other
grids’. Dispatch outcomes show the lower dispatch from the additional coal retirements
being offset by gas-fired generation. With the lower level of firm fossil fuelled capacity
operating, some wind capacity is displaced by geothermal capacity in the SWIS.
Relative to the core Reference case, resource costs are higher due to the higher new build
component ($0.7 billion higher) and higher variable O&M costs ($1.2 billion higher). This is
offset by lower fixed O&M due the retired capacity not being required to be maintained ($1.2
billion lower). Higher retail prices also results in a greater level of uptake of small scale solar
PV, with resource costs $0.7 billion higher. In net terms resource costs are $1.3 billion
higher in aggregate.
Overall emissions are around 5 Mt/year lower from 2020 onwards due to the withdrawal of
the additional coal-fired capacity relative to the core assumptions.
Net change in generator profitability indicates that the coal-fired sector is materially more
profitable under this scenario which provides some justification for the withdrawal. However
this doesn’t necessarily mean that portfolios which withdrew capacity themselves are
financially better off under this scenario.
The higher wholesale prices results in lower LGC prices (down 8.6% in NPV terms) and
slightly higher uptake of small-scale solar PV (7.3% higher in the period to 2030).
Compliance costs are 11.3% lower for the LRET, but 4.6% higher for the SRES.
Under this scenario, household retail electricity bills are higher relative to the core Reference
case, with present values being $584 higher in the period to 2030 and $768 higher in the
period to 2040. Direct RET costs are slightly lower due to the lower LGC price outcomes
offsetting increases from SRES.
28 With the lower fossil fuelled capacity in place, the model chooses to refurbish additional incumbent fossil fuelled capacity
relative to the Reference case, making the net differential much smaller from around 2030 onwards.
107
Figure 101 Sensitivity results: Reference case: Permanent retirements (change from Reference case)

100
8,000
90
6,000
80
Real 2014 $/MWh
70 4,000
60
GWh sent-out
2,000
50
40 0
30 -2,000
20
-4,000
10
0 -6,000
-8,000
Reference case NEM
Reference case: Permanent retirements NEM
Reference case: Permanent retirements Other grids Hydro Wind Solar Other

2.0
Total 1.3
1.0
0.0
SRES: SWHs 0.0 -1.0
SRES: SGUs 0.7 -2.0
Mt CO2-e
Total Large-scale 0.6 -3.0
Unserved Energy 0.0 -4.0

-5.0
Variable O&M 1.2
-6.0
Fixed O&M -1.2
-7.0
Refurbishment -0.2
-8.0
New build 0.7
-2 -1 0 1 2
NPV ($ billion)

cost to 2040 (real 2014 $)
-11.3%
to 2040 (real 2014 $)

-8.6%
prices

-2.8%
7.3%

-6.4% 3.6%
-12% -10% -8% -6% -4% -2% 0% 0.0%1.0%2.0%3.0%4.0%5.0%6.0%7.0%8.0%

108
Geothermal 1.1 $900

$815
$800 $768
Solar -0.2
$700 $631
Wind 0.4 $584
$600
NPV ($)
Hydro 1.8 $500
$400
Baseload Gas 0.7
$300
Peaking Gas 0.5
$200 $152 $140
Brown coal 3.4 $100
Black coal 7.5 $0

($12)
($100) ($47) ($47)
-2 0 2 4 6 8 2015-20 2015-30 2015-40
Source: ACIL Allen

The permanent retirements sensitivity has also been compared against the Repeal case
(Figure 102) and run under the 50% Growth policy case (Figure 103).
Significant points to note from the results are:
 Repeal case: wholesale price outcomes under the Repeal case when compared with
the Reference case: Permanent retirements sensitivity results do not appear
significantly different for NEM regions. This indicates that the additional withdrawal of
capacity has been successful in negating the price supressing impact of renewable
entry. This results in the Repeal case having slightly higher relative emissions.
Importantly, the impact on retail bills from repeal of the policy is for overall lower prices
to consumers (here the reduction in compliance costs is not outweighed by change in
wholesale price outcomes), with overall present values being $585 lower (in present
value terms) in the period to 2030 and $653 lower in the period to 2040. This is the
opposite outcome to the standard Reference case and highlights the uncertainty
surrounding the RET’s impact on wholesale price outcomes.
 50% Growth case: this sensitivity shows similar outcomes with wholesale prices being
largely unchanged from the Reference case. Projected LGC prices are around 10%
lower in NPV terms, with aggregate LRET compliance costs around 29% lower than
the Reference case. Small-scale solar PV installations are around 20% lower against
the Reference case; aggregate SRES compliance costs are 70% lower than the
Reference case. Household retail costs are lower from the move to the 50% Growth
policy: $243 in the period to 2030 and $308 in the period to 2040. This compares with
the result under the core assumptions in which retail prices were higher by $119 in the
period to 2030 and $179 in the period to 2040.
109
Figure 102 Sensitivity results: Repeal case (change from Reference case: Permanent retirements)
100
40,000
90
80 30,000
70
Real 2014 $/MWh
20,000
60
GWh sent-out
50 10,000
40
0
30
20 -10,000
10
0 -20,000
-30,000
Repeal case NEM
Reference case: Permanent retirements Other grids Black coal Brown coal Peaking Gas Baseload Gas
Repeal case Other grids Hydro Wind Solar Other

35.0
Total -15.3
30.0
25.0
SRES: SWHs -1.1
SRES: SGUs -2.6 20.0

Mt CO2-e
Total Large-scale -11.6 15.0

Fixed O&M -2.7
0.0
Refurbishment 0.9
-5.0
New build -13.1
-20 -15 -10 -5 0 5

NPV ($ billion)

-100.0% cost to 2040 (real 2014 $)

-100.0%
prices

-86.8%
-16.5%

-120%-100% -80% -60% -40% -20% 0% -120% -100% -80% -60% -40% -20% 0%
110
Geothermal -1.1 $100 $46
Solar -1.6 $0
($31) ($21)
Wind -14.7 ($100)
NPV ($)
($200)
Hydro -0.8
($300)
Baseload Gas 0.6
($400) ($356)
Peaking Gas -0.4 ($387)
($500)
Brown coal 0.8
($600)
($631) ($631)
($700) ($653)
-20 -15 -10 -5 0 5 10 2015-20 2015-30 2015-40
Source: ACIL Allen
Figure 103 Sensitivity results: 50% Growth case: Permanent retirements (change from Reference case:
Permanent retirements)

100
20,000
90
80 15,000
70
Real 2014 $/MWh
10,000
60
GWh sent-out
5,000
50
40 0
30 -5,000
20
-10,000
10
0 -15,000
-20,000
50% Growth case: Permanent retirements NEM
Reference case: Permanent retirements Other grids Black coal Brown coal Peaking Gas Baseload Gas
50% Growth case: Permanent retirements Other grids Hydro Wind Solar Other

20.0
Total -11.4
18.0
SRES: SGUs 12.0

-2.7
Mt CO2-e
10.0
8.0
Fixed O&M -1.5 2.0

0.0
Refurbishment 0.6
-2.0
New build -8.5
-15 -10 -5 0 5
NPV ($ billion)
111

-38.7% cost to 2040 (real 2014 $)
to 2040 (real 2014 $)

-9.8%
prices

-15.8%
-20.0%

-70%-60%-50%-40%-30%-20%-10% 0% -80%-70%-60%-50%-40%-30%-20%-10% 0%
Geothermal -1.1 $50 $20
Solar -0.3 $0
($10)
Wind -6.7 ($50)
($44)
NPV ($)
Hydro ($100)
-0.5
($150) ($117)
($127)
Baseload Gas -0.2
($200)
Peaking Gas -0.3
($250)
($263) ($263)
($300)
Black coal 2.0
($308)
($350)
-8 -6 -4 -2 0 2 4 2015-20 2015-30 2015-40
Source: ACIL Allen
112
9 Abatement costs
9.1.1 Methodology for calculation

ACIL Allen has employed two methodologies in calculating the cost of abatement from the
RET. Both use the present value of the change in resource costs (the numerator), while one
method applies a time-based discount factor to the change in emissions (the denominator).
𝑃𝑟𝑒𝑠𝑒𝑛𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑐ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑟𝑒𝑠𝑜𝑢𝑟𝑐𝑒 𝑐𝑜𝑠𝑡𝑠
𝐶𝑜𝑠𝑡 𝑜𝑓 𝐴𝑏𝑎𝑡𝑒𝑚𝑒𝑛𝑡 =
[𝐷𝑖𝑠𝑐𝑜𝑢𝑛𝑡𝑒𝑑] 𝑐ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑒𝑚𝑖𝑠𝑠𝑖𝑜𝑛𝑠
Resource costs
In this context, Resource costs are defined as the aggregate expenditure on electricity
generation29 throughout Australia. That is, the sum of:
 capital costs of new developments
 capital costs associated with refurbishments of existing power stations
 fixed operating and maintenance costs (FOM)
 variable operating and maintenance costs (VOM) 30
 unserved energy (a notional value placed on any energy which is demanded but not
supplied).
The changes in resource costs are calculated for the Policy case and the no Policy case and
a present value of this series is calculated using a discount rate of 7% (real). The choice of
discount rate for resource costs is lower than a commercial discount rate for renewable
energy projects (typically around 10% real currently).
The capital cost elements in the resource costs measure are typically for long-life assets
which may continue operating beyond the modelling period. For this reason, the capital cost
components are converted to annualised equivalent costs using a Present Value Interest
Factor Annuity (PVIFA) calculation which gives a constant annual cost over the life of the
asset which is equivalent in present value terms to the up-front capital cost. The resource
cost calculation then sums the annual equivalent costs for the period modelled (costs are
discounted to return them to present value terms). Therefore, a 25-year life asset which is
constructed in the last year of the modelling only attributes a small proportion of the total up-
front capital cost in the resource cost measure. Other costs such as FOM, VOM and
unserved energy are annual costs and no adjustments are required (apart from discounting
to convert them to present value terms).
Emissions
The first approach uses a discounted stream of emissions abatement achieved by the
Policy. Whilst abatement is a physical rather than a financial stream, discounting future
abatement is done for two primary reasons:
29 For the purposes of this calculation, it is implicitly assumed that there are no consequential changes in other electricity costs
(such as network expenditure) resulting from the policy.
30 This element include fuel costs associated with running power stations
113
 it attempts to mimic the discounting of the financial costs incurred through greenhouse
gas emissions and climate change (estimated through a social carbon cost damage
function)
 it recognises the social rate of time preference for abatement in that earlier abatement
is more valuable than later abatement.
It is not feasible to calculate the economic costs from damage incurred as a result of climate
change within the context of this project. Social carbon cost damage functions are usually
estimated through Integrated Assessment Models (IAM). However, there is significant
ongoing academic debate with respect to this matter and any estimates remain highly
uncertain. If a credible damage function was available to estimate the social carbon cost at
various points in time, a more correct approach would be to discount the value of abatement
at each point in time in accordance with the damage function, as it reflects the value at the
margin of each tonne of abatement. In the absence of this, discounting the abatement
stream is a second best solution.
We note that in evaluating the merit of undertaking abatement activities, if no discount rate
is applied, then it is implied that abatement at any time in the future has the same value as
abatement today. As it is most probable that the costs of undertaking abatement actions will
decline into the future, no discounting implies that the best policy is to defer any abatement
until a long time into the future when costs are very small compared with costs of abatement
today. In ACIL Allen’s view, discounting the abatement stream is therefore a sensible
solution to the lack of greater detail about the carbon damage function. A discount rate of
7% has been used for both the numerator and denominator.
The second approach (method 2) uses an undiscounted emissions value as the
denominator in the calculation. ACIL Allen has used the second method to calculate
abatement costs at the direction of the Expert Panel. This approach recognises that
abatement is generally considered in cumulative form as a budget over a timeframe and the
allocation of abatement in any year over the lifetime of a policy is small relative to the
cumulative abatement goal – it assumes that the timing of the abatement is not critical.
However, ACIL Allen considers that the second method does not appropriately reflect the
costs of emissions abatement on an inter-temporal basis.
9.1.2 Abatement costs across policy cases to 2030

In presenting abatement costs for the RET, ACIL Allen has attempted to split out LRET and
SRES components. Owing to the difficulty in estimating abatement and resource costs for
solar water heating (SWH), this has been excluded from the analysis. Therefore abatement
costs are provided for the LRET, solar PV and a total overall value excluding SWH.
Table 9 presents the calculated abatement costs for the RET in the period to 2030 for all
policy cases and sensitivities using Method 1. These are also presented graphically in
Figure 104. Abatement costs are presented relative to the Repeal case for that assumption
set (for example, base policy cases are compared with the base Repeal scenario; low
demand policy cases are compared with the low demand Repeal scenario etc.).31
The cost of abatement arising from the RET policy is relatively high, ranging from a low of
$49/tonne CO2-e under the 50% Growth Permanent retirements case to as high as
$77/tonne CO2-e under the Reference case with high capital costs scenario. The 50%
31 The Closed to New Entrants scenario has a zero value as this scenario does not provide any policy incentives for new
renewable developments.
114
Growth case under Low demand conditions is somewhat unique in that it effectively only
provides a subsidy to small-scale technologies which have much higher abatement costs.
Abatement costs are significantly lower for the LRET compared with the SRES ($40-
$72/tonne compared with $164-$191/tonne). This is primarily a result of the underlying
levelised cost of energy from large-scale technologies (for example wind at $80-$100/MWh)
compared with small scale solar PV which even at today’s costs is around $180-$200/MWh.
Therefore, the more skewed renewable policy is toward subsidising solar PV relative to
large-scale technologies, the higher the effective abatement cost of the policy.
Costs range between scenarios depending on the underlying grid emissions intensity and
the costs of renewable power relative to fossil-fuel based power.
Table 9 Calculated abatement costs for the RET to 2030: Method 1 (abatement stream discounted)
RET LRET Solar PV
NPV of NPV of NPV of

NPV of Abatement NPV of Abatement NPV of Abatement
change in change in change in
abatement cost abatement cost abatement cost
resource resource resource
(Mt CO2-e) ($/tonne) (Mt CO2-e) ($/tonne) (Mt CO2-e) ($/tonne)
costs ($m) costs ($m) costs ($m)
Reference case 10,430 154 $68 9,015 146 $62 1,414 8 $175
Real 20% case 3,447 56 $62 2,527 50 $51 919 6 $165
Real 30% case 9,134 130 $70 7,784 122 $64 1,351 8 $173
50% Growth case 3,335 57 $59 2,622 53 $50 713 4 $169
Reference case: Carbon 10,668 152 $70 9,279 144 $65 1,389 8 $175
Reference case: High capital costs 11,902 154 $77 10,514 147 $72 1,388 7 $191
50% Growth case: High capital 4,025 59 $68 3,338 55 $60 687 4 $181
Reference case: High demand 8,708 129 $68 7,544 122 $62 1,164 7 $177
50% Growth case: High demand 5,951 90 $66 5,492 88 $63 460 2 $190
50% Growth case: Low Demand 640 4 $164 0 0 n/a 640 4 $164
Reference case: Low demand 10,932 160 $68 9,199 151 $61 1,733 9 $183
Real 20% case: Low demand 3,972 54 $74 3,066 48 $64 906 5 $171
Real 30% case: Low demand 9,821 136 $72 8,237 127 $65 1,584 9 $181
Reference case: Perm retirements 11,010 178 $62 9,229 167 $55 1,781 10 $172
50% Growth case: Perm retirements 3,552 73 $49 2,727 68 $40 825 5 $166
Note: 7% real discount rate used for NPV calculations. Excludes resource costs and abatement outcomes from SWH installations. Capital
expenditure on new-build capacity and refurbishments is pro-rated down (on a NPV equivalent annuity basis) based on the remaining modelling
horizon relative to the full economic life of technologies.
Source: ACIL Allen
115
Figure 104 Calculated abatement costs for the RET to 2030: Method 1 (abatement stream discounted)
$250
RET LRET Solar PV
Abatement cost $/tonne CO2-e
$200 191 190

181 177 183 181
175 173 169 175 171 172
165 164 164 166
$150
$100
77 72 74 72
68 62 70 64 70 65 68 68 62 66 63 68 65
62 59 60 61 64 62
51 50 55 49
$50 40
0
$0
Note: 7% real discount rate used for NPV calculations. Excludes costs and abatement outcomes from SWH installations. Capital expenditure on
new-build capacity and refurbishments is pro-rated down (on a NPV equivalent annuity basis) based on the remaining modelling horizon relative
to the full economic life of technologies.
Source: ACIL Allen
Table 10 and Figure 105 shows the corresponding cost of abatement using method 2. While
the relativities between scenarios are the same, the absolute values are lower due to the
abatement not being discounted.
Table 10 Calculated abatement costs for the RET to 2030: Method 2 (abatement stream undiscounted)
RET LRET Solar PV

Sum of Abatement Sum of Abatement Sum of Abatement
Reference case 10,430 299 $35 9,015 284 $32 1,414 15 $95
Real 20% case 3,447 109 $32 2,527 99 $26 919 10 $91
Real 30% case 9,134 271 $34 7,784 257 $30 1,351 14 $96
50% Growth case 3,335 110 $30 2,622 103 $26 713 7 $96
Real 20% case: Low demand 3,972 106 $38 3,066 96 $32 906 10 $94
50% Growth case: Perm retirements 3,552 142 $25 2,727 133 $20 825 9 $90
Source: ACIL Allen
116
Figure 105 Calculated abatement costs for the RET to 2030: Method 2 (abatement stream undiscounted)
$140 131
RET LRET Solar PV
$120
104 103 101

95 96 96 95 96 94 97
$100 91 90 90 89 90
$80
$60
36 33 40 37 36 38
35 32 32 34 30 35 35 32 34 32 32 32 35 32 28
$40 30 31 31
26 26 25
20
$20
0
$0
Source: ACIL Allen
9.1.3 Abatement costs across policy cases to 2040

While the previous analysis examines the period to 2030, it should be recognised that plant
installed throughout this period will continue to have benefits in terms of abatement beyond
the end of the policy period. As the modelling undertaken extends to 2040, abatement costs
to 2040 have also been calculated, as presented in Table 11 and shown graphically in
Figure 106.32
While resource costs have been pro-rated down (on a NPV equivalent annuity basis) to
2030, abatement is likely to be relatively stable in future years, so using a 2030 cut-off date
will tend to slightly overestimate abatement costs for large-scale technologies. This can be
seen through the slightly lower abatement costs calculated for the LRET in Table 11, which
range from $45-$71/tonne ($49-$77/tonne when calculated to 2030).
However, calculated abatement costs for the SRES are slightly higher when the analysis is
extended to 2040 due to ongoing capital investment which occurs with small-scale systems
(unlike the large-scale which virtually ceases early next decade in most scenarios) and the
declining grid emission intensity in the longer-term which more than offsets capital cost
reductions for small scale systems.
Overall RET abatement costs are slightly lower, averaging around 8-10% lower compared
with the 2030 figures.
32 Extending the analysis further beyond 2040 into perpetuity is difficult as it is beyond the period modelled and due to
approach of discounting both abatement and resources costs, is unlikely to materially change the results.
117
Table 11 Calculated abatement costs for the RET to 2040: Method 1 (abatement stream discounted)
RET LRET Solar PV

NPV of Abatement NPV of Abatement NPV of Abatement
Reference case 12,913 207 $62 11,008 197 $56 1,905 10 $185
Real 20% case 4,291 77 $56 3,087 70 $44 1,204 7 $176
Real 30% case 12,859 205 $63 11,122 196 $57 1,737 9 $189
50% Growth case 4,064 77 $52 3,164 73 $44 900 5 $182
50% Growth case: Perm retirements 4,501 100 $45 3,367 93 $36 1,134 7 $174
Source: ACIL Allen
Figure 106 Calculated abatement costs for the RET to 2040: Method 1 (abatement stream discounted)
$250
RET LRET Solar PV 204
193 192 195 198
$200 185 189 187 185 184

176 182 180
172 172 174
$150
$100
71 65 68 66 71
62 56 63 57 64 58 61 59 53 60 64 58
56 52 53 56 54 55 51
44 44 45
$50 36
0
$0
Source: ACIL Allen
Table 12 and Figure 107 shows the corresponding cost of abatement using method 2. While
the relativities between scenarios are the same, the absolute values are lower due to the
abatement not being discounted.
118
Table 12 Calculated abatement costs for the RET to 2040: Method 2 (abatement stream undiscounted)
RET LRET Solar PV

Sum of Abatement Sum of Abatement Sum of Abatement
Reference case 12,913 520 $25 11,008 496 $22 1,905 24 $79
Real 20% case 4,291 197 $22 3,087 182 $17 1,204 15 $79
Real 30% case 12,859 589 $22 11,122 569 $20 1,737 20 $89
50% Growth case 4,064 196 $21 3,164 186 $17 900 10 $86
50% Growth case: Perm retirements 4,501 255 $18 3,367 239 $14 1,134 16 $73
Source: ACIL Allen
Figure 107 Calculated abatement costs for the RET to 2040: Method 2 (abatement stream undiscounted)
$140
127
RET LRET Solar PV
$120
92 95
$100 89 88
86 83 85
79 79 81 80
$80 73 73 69 73
$60
$40 28 26 28 25
25 22 22 22 20 25 23 24 20 23 21 22 21 25 25 23 23 20
21 17 21 18 14
17
$20
0
$0
Note: 7% real discount rate used for NPV calculations. Excludes costs and abatement outc omes from SWH installations. Capital expenditure on
Source: ACIL Allen
119
10 Summary and conclusions

This section provides a cross scenario comparison of various outputs from the modelling
and conclusions regarding the impact of the RET policy variants.
10.1 Summary of modelling outcomes
10.1.1 Electricity generation sector resource costs

Figure 108 presents a summary of the NPVs of aggregate sector resource costs over the
period 2015-40 across each of the scenarios and sensitivities modelled. Sector resource
costs represent the annual expenditure by the sector on:
 Capital expenditure (on both generating capacity and any interconnector
expansions/augmentations33)
 Refurbishment of existing and new generators for life extension beyond initial economic
life
 Fixed operating costs (fixed costs associated with normal operation and stay in business
capital expenditure associated with existing and new generating capacity)
 Variable operating costs (fuel costs and variable O&M costs for existing and new
generation)
 Unserved energy (which, if occurs, is valued at the equivalent market price cap).34
This measure can be interpreted as the broad costs to society of generating electricity for
consumption by consumers. It gives an indication of the sector’s labour and capital
productivity when viewed on a per MWh basis. Notably, it does not include transmission and
distribution cost elements and does not indicate who bears this cost burden for provision of
power. It also does not take into account the opportunity cost of using labour and capital to
produce electricity versus other alternative uses throughout the economy (which is relevant
when considering policies that subsidise large amounts of renewable generation
investment).
For the Reference case (under Core assumptions), costs total $121.9 billion in present value
terms over the period to 2040 using a discount rate of 7% pre-tax real. Under the core
assumptions, all of the policy variants examined resulted in a reduction in sector resource
costs, indicating increased capital and productivity efficiencies are being gained by the
economy. The Real 30% scenario has almost the same result due to deferral of wind built
being offset by an overall larger amount of renewable development in the longer-term.
The Repeal case has the lowest resource cost result, with the competitive wholesale
electricity markets determining the most efficient, least cost plant mix to meet demand. We
note that this was one of the core intentions in the establishment of the NEM, with its rules
and principles being technology agnostic. In the absence of the RET policy the market is
allowed to determine its own optimal level of generation investment, rather than have
33 Capital expenditure on new-build capacity is pro-rated down (on a NPV equivalent annuity basis) based on the remaining
modelling horizon relative to the full economic life of technologies.
34 Unserved Energy is routinely calculated as part of sector resource costs. It represents the value of any energy which is not
supplied through non-voluntary load shedding.
120
arbitrary constraints imposed upon it. In a market environment where capacity is already
oversupplied and demand may continue to decline it is quite reasonable (and efficient) for
no new investment in capacity to occur.
The sectoral resource costs under the carbon sensitivities are only marginally higher than
for the respective core scenarios. This is due to the relatively low carbon price assumed
which does not materially alter the merit order (note that the carbon permit costs are not
included within the resource cost measure).
Results for the High capital cost assumptions are higher for the Reference and 50% Growth
scenarios, but largely unchanged for the Repeal case due to the lack of large-scale
renewables developed. High and Low demand sensitivities result in higher/lower resource
costs respectively due to the different volume of energy provided and implications for
changes in variable generating costs.
The permanent retirement sensitivity has slightly higher resource costs relative to the
respective core assumptions, owing to the additional requirement to replace withdrawn
capacity (where required). The model also chooses to refurbish some incumbent capacity
which it doesn’t do under the core assumptions.
Figure 108 Aggregate sector resource costs (NPV 2015-2040): All scenarios/sensitivities
132.9
138.1
126.7
123.8
123.2
122.8
121.9
121.8
160
112.7
112.5
111.8
111.4
108.9
108.6
108.2
108.0
107.9
106.4
106.0
140
96.0
90.8
90.7
Real 2014 $ billion
120
100
80
60
40
20
0
Real 20%
Real 30%
Real 20%
Real 30%
Repeal
Repeal
Repeal
Repeal
Repeal

50% Growth
50% Growth
50% Growth
50% Growth
50% Growth
Reference
Reference
Reference
Reference
Reference
Core Carbon High capital costs High demand Low demand Permanent Reference
retirements
Note: Measure includes capital expenditure (on both generating capacity and any interconnector expansions/augmentations); refurbishment of
existing and new generators for life extension beyond initial economic life; fixed operating costs (fixed costs associated with normal operation
and stay in business capital expenditure associated with existing and new generating capacity); variable operating costs (fuel costs and variable
O&M costs for existing and new generation) and unserved energy. NPV calculated using a 7% real discount rate.
Source: ACIL Allen
10.1.2 Greenhouse gas emissions and abatement costs

Aggregate projected greenhouse gas emissions are presented in Figure 109. It should be
noted that this measure excludes some generation (specifically from non-scheduled
generation in NEM regions, own-generation in the SWIS and off-grid generation), so does
not represent a complete view of total electricity sector emissions.
The RET policy delivers emissions abatement, with the Reference case and Real 30%
scenarios consistently resulting in the lowest emission outcomes across assumption sets.
Conversely the Repeal of the RET would increase emissions by between 8% and 14%
relative to the Reference case.
As with any Government expenditure or program, an important consideration is whether the
policy offers value for money relative to alternatives. Figure 110 and Figure 111 present
121
results for the estimated cost of abatement from the policy cases relative to the
corresponding Repeal scenario under two different methods. Method 1 calculates
abatement costs as the NPV of the change in resource costs divided by the discounted
change in abatement. Method 2 is the same except the emissions in the denominator are
not discounted (see section 9 for more detail on the methodology).35
Using method 1, abatement costs range from $49/tonne under the 50% growth case
(permanent retirement sensitivity) to $164/tonne under the 50% growth case (low demand
assumptions). There is also a large difference between calculated abatement costs for the
LRET and SRES, with the abatement costs for the SRES at least 2 to 3 times more
expensive than the LRET. Therefore policy scenarios which tend to reduce incentives for
solar PV will tend to lower the overall RET abatement cost. This is why the result for the
50% growth case (low demand assumptions) appears to be such an outlier. Under this
scenario where the LRET is increased based on demand growth, in a low demand
environment the LRET target remains static at its 2014 level. Under these conditions, the
policy therefore reduces to a SRES-only policy which explains the high abatement cost
outcome.
Abatement costs under the permanent retirement sensitivity results in a lower cost per
tonne. This is due to the impact of renewables resulting in a larger withdrawal of some of the
least profitable incumbent plant which has positive implications for emission outcomes. 36
Figure 109 Aggregate emissions: 2015-2040: All scenarios/sensitivities
5,784
5,481
5,345
7000
5,117
5,114
5,114
5,072
5,072
4,917
4,917
4,909
4,859
4,594
4,593
4,557
4,525
4,494
6000
3,973
3,952
3,781
3,463
3,393
5000
4000
Mt CO2-e
3000
2000
1000
0
Real 30%
Real 20%
Real 30%
Real 20%
Repeal
Repeal
Repeal
Repeal
Repeal

50% Growth
50% Growth
50% Growth
50% Growth
50% Growth
Reference
Reference
Reference
Reference
Reference
Reference
retirements
Source: ACIL Allen
35 As noted in Section 9.1.1, ACIL Allen has used Method 2 at the request of the Expert Panel but considers that it does not
appropriately reflect the costs of emissions abatement on an inter-temporal basis.
36 Note that the plant expected to be withdrawn from the market are the least profitable rather than the highest emitting plant.
122
Figure 110 Calculated abatement cost for RET (Method 1): All scenarios/sensitivities
$164
$180
$160
Abatement cost (Real 2014 $/tonne)
$140
$120
$100
$77
$74
$72
$70
$70
$68
$68
$68
$68
$66
$62
$62
$80
$59
$49
$60
$40
$20
$0
Real 20%
Real 30%
Real 30%
Real 20%
50% Growth
50% Growth
50% Growth
50% Growth
50% Growth
Reference
Reference
Reference
Reference
Reference
Reference
Core Carbon High capital High demand Low demand Permanent
costs retirements
Note: Method 1 uses the change in present value of electricity generation sector resource costs divided by the discounted change in abatement.
Calculation covers the period 2015 to 2040 using a 7% real discount rate. Excludes resource costs and abatement outcomes from SWH
installations
Source: ACIL Allen
Abatement costs calculated using method 2 results in virtually the same relativities between
scenarios, with the absolute cost values being lower (the volume of emissions are not
discounted and so the denominator is larger). However, even under this method, the RET
policy appears to offer abatement at relatively high cost compared with international permit
prices.
Figure 111 Calculated abatement cost for RET (Method 2): All scenarios/sensitivities
$100 $90
$90
Abatement cost (Real 2014 $/tonne)
$80
$70
$60
$40
$50
$38
$36
$36
$35
$35
$35
$35
$34
$34
$32
$32
$30
$40
$25
$30
$20
$10
$0
Real 20%
Real 30%
Real 30%
Real 20%
50% Growth
50% Growth
50% Growth
50% Growth
50% Growth
Reference
Reference
Reference
Reference
Reference
Reference
Core Carbon High capital High demand Low demand Permanent

costs retirements
Note: Method 2 uses the change in present value of electricity generation sector resource costs divided by the undiscounted change in
abatement. Calculation covers the period 2015 to 2040 using a 7% real discount rate. Excludes resource costs and abatement outcomes from
SWH installations
Source: ACIL Allen
10.1.3 Investment in renewable generation

Figure 112 shows that investment in large-scale renewable generation is critically
dependent upon subsidies provided through the LRET. In the absence of these subsidies
(i.e. under Repeal scenarios), virtually no large-scale development occurs within the grids
123
modelled in the period to 2040. Investment drops to as little as $1.8 billion under the core
assumptions and $0.7 billion under low demand assumptions.
Scenarios that scale back the LRET to align with a Real 20% level, reduce investment by
around 60%.
Figure 112 Total investment in Large-scale renewable generation: 2015-2040: All scenarios/sensitivities
30.1
29.1
35
Investment (Real 2014 $ billion)
30
22.6
20.4
20.2
20.1
19.7
18.8
25
14.6
20
15
8.4
8.4
8.3
8.3
7.6
10
2.5
2.5
1.8
1.8
1.8
0.7
0.7
5
0.0
0
Real 20%
Real 30%
Real 20%
Real 30%
Repeal
Repeal
Repeal
Repeal
Repeal

50% Growth
50% Growth
50% Growth
50% Growth
50% Growth
Reference
Reference
Reference
Reference
Reference
Reference
retirements
Note: Excludes non-scheduled generation in NEM regions, own-generation in the SWIS and off-grid generation.
Source: ACIL Allen
The development of small-scale solar PV is much less reliant upon subsidies provided
through the SRES. Figure 113 shows that across the scenarios aggregate investment is at
most 16% lower than the Reference case (low demand assumptions). ACIL Allen’s
modelling indicates that the level of uptake from solar PV is more dependent upon structural
reform to retail electricity tariffs (retention of variable charging versus moving to more
fixed/demand based charging for network components) rather than subsidies provided
through the SRES. Of the policy cases under the core assumptions, the 50% Growth case
results in the lowest level of aggregate investment in small-scale systems – even lower than
that which occurs under the Repeal case. This is due to the 50% Growth scenario scaling
back subsidies through shortened deeming periods and also the lower retail prices which
prevail. As the measure is a simple sum of investment over the period to 2040, the lower
retail prices in the long-term under this scenario results in a lower level of aggregate
investment. This illustrates that more generally, the impact of the SRES declines over time
as the deeming period reduces and the outlook for retail prices becomes a more important
factor for small-scale systems.
124
Figure 113 Total investment in Small-scale renewable generation: 2015-2040: All scenarios/sensitivities
33.2
32.9
32.7
40.0
32.1
31.4
30.9
30.7
30.7
30.6
30.1
30.0
29.7
Investment (Real 2014 $ billion)
29.0
28.8
28.7
28.5
28.2
35.0
27.3
27.2
27.0
26.4
25.9
30.0
25.0
20.0
15.0
10.0
5.0
0.0 Real 20%
Real 30%
Real 20%
Real 30%
Repeal
Repeal
Repeal
Repeal
Repeal
50% Growth
Reference
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
Reference
retirements
Note: Solar PV only (excludes capex on SWHs)

Source: ACIL Allen
10.1.4 RET compliance costs

Figure 114 shows the aggregate compliance costs (Projected LGC price multiplied by the
LRET targets) summed over the period 2015-40. This represents the subsidy provided to
renewable developers under the LRET scheme, the costs of which are borne by consumers
(directly through the RPP) and other incumbent generators (indirectly through lower
earnings). For consumers, changes to other cost elements in electricity charges may or may
not offset this direct cost (see section 10.1.5).
The aggregate subsidy required to bring on renewable development is significant. Under the
Reference case, LRET compliance costs totals $33.3 billion (real 2014 dollars) over the
period. This value is slightly less under carbon and high demand assumptions and slightly
more under high capital costs and low demand assumption sets.
A Repeal of the scheme would reduce future compliance costs to zero and does not provide
any compensation to existing investments. The level of compensation provided under the
Closed to New Entrants scenario is significant in itself at $8.3 billion (real 2014 dollars).
Figure 115 shows that compliance costs associated with the SRES are significantly smaller
than the LRET due to declining deeming periods and smaller uptake on a capacity basis.
Policy scenarios which shorten the policy life to 2020 and reducing upfront deeming tend to
reduce compliance costs significantly (by 50-70%).
125
Figure 114 Aggregate LRET compliance costs (2015-2040): All scenarios/sensitivities
42.1
50
39.5
37.9
45
35.4
33.3
40
29.7
29.6
28.6
Real 2014 $ billion
35
21.9
21.9
30
20.8
19.0
18.1
18.1
25
20
8.6
8.3
8.3
15
10
0.0
0.0
0.0
0.0
0.0
5
0
Real 20%
Real 30%
Real 20%
Real 30%
Repeal
Repeal
Repeal
Repeal
Repeal
50% Growth
Reference
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
Reference
retirements
Note: Sum of projected LGC price multiplied by LRET targets

Source: ACIL Allen
Figure 115 Aggregate SRES compliance costs (2015-2040): All scenarios/sensitivities

4.6
4.7
4.7
4.5
4.5
4.4
4.4
5.0
4.0
4.5
4.0
Real 2014 $ billion
3.5
3.0
2.1
2.1
2.5
1.5
1.4
1.4
1.3
1.3
2.0
1.5
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.5
0.0
Real 20%
Real 30%
Real 20%
Real 30%
Repeal
Repeal
Repeal
Repeal
Repeal

50% Growth
50% Growth
50% Growth
50% Growth
50% Growth
Reference
Reference
Reference
Reference
Reference
Reference
retirements
Note: Sum of projected STC creation multiplied by assumed STC price

Source: ACIL Allen
Figure 116 shows the aggregate compliance costs for the RET (the sum of LRET and SRES
compliance costs). The aggregate subsidy required to bring on renewable development
under the Reference case totals $37.8 billion (real 2014 dollars) over the period to 2030.
This value is slightly less under carbon ($34.3 billion) and high demand assumptions ($33.3
billion) and slightly more under high capital costs ($41.9 billion) and low demand ($39.9
billion) assumption sets.
The Real 30% case results in higher compliance costs relative to the Reference case for
electricity consumers. All other policy cases examined result in a reduction in direct
compliance costs for electricity consumers.
126
Figure 116 Aggregate RET compliance costs (2015-2040): All scenarios/sensitivities
43.9
46.6
41.9
50.0
39.9
37.8
45.0
34.3
34.2
33.3
40.0
Real 2014 $ billion
35.0
23.4
23.2
22.8
30.0
20.4
20.3
19.5
25.0
20.0
9.9
8.3
8.3
15.0
10.0
0.0
0.0
0.0
0.0
0.0
5.0
0.0
Real 20%
Real 30%
Real 20%
Real 30%
Repeal
Repeal
Repeal
Repeal
Repeal
50% Growth
Reference
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
Reference
retirements
Note: Sum of projected LGC price multiplied by LRET targets and projected STC creation by assumed STC price
Source: ACIL Allen
10.1.5 Residential retail prices

Figure 117 shows the projected aggregate cost for an average Australian household on
electricity over the period 2015-40 in NPV terms. In most cases, moving from the Reference
case to the Repeal case (the most extreme policy variant) results in household electricity
costs rising very slightly in net terms (the reduction in direct compliance costs is outweighed
by the modelled increase in wholesale electricity prices). This indicates the LRET policy
causes wealth transfers from existing generators to both renewable proponents and
consumers.
This pattern of price changes does not hold under low demand conditions. This is due to the
inability of new renewable generation to further supress wholesale prices below levels which
are unsustainable for incumbent generators to keep operating. Under these conditions,
removal of the direct compliance costs is not offset by wholesale price movements and
consumers are better off under a Repeal scenario. This is particularly interesting in the
current NEM environment in which demand for electricity has fallen every year since 2008-
2009 and the largest uncertainty is with respect to future demand growth/decline.
127
Figure 117 NPV of average household total expenditure on electricity (2015-2040): All scenarios/sensitivities
20,260
20,451
20,074
20,037
21,000
19,887
19,771
19,754
19,634
19,464
19,453
20,500
19,358
19,349
19,305
19,213
19,193
19,182
19,181
19,119
19,119
19,092
19,003
20,000
NPV of retail bills ($)
18,706
19,500
19,000
18,500
18,000
17,500
Real 30%
Real 20%
Real 30%
Real 20%
Repeal
Repeal
Repeal
Repeal
Repeal

50% Growth
50% Growth
50% Growth
50% Growth
50% Growth
Reference
Reference
Reference
Reference
Reference
Reference
retirements
Note: NPV of annual residential bills for average household over the period 2015-40. Uses a 7% real discount rate
Source: ACIL Allen
10.1.6 Assessing net electricity consumer benefits

The public analysis of the costs and benefits of the RET scheme has been dominated by
views on the net benefits that the RET scheme provides to electricity consumers. These net
electricity consumer benefits are generally calculated by assessing wholesale price (pool)
and RET certificate price changes for the market with and without the RET scheme (i.e. the
modelled impact of the subsidised renewable generation on wholesale market prices).
Assessing the net consumer benefits limited to a specific economic sector cannot be
considered to be either a Social Benefit-Cost analysis (SCBA) or an economy wide
assessment of the RET scheme.
The approach to calculating net electricity consumer benefits that has generally been
applied by various modellers has been to model the differences in RET certificate costs and
pool prices over time between the current RET scheme and the alternative policy case (s);
and then to apply these differences to typical consumer tariffs. In that sense they do not
even represent net electricity consumer benefits because they are not usually aggregated
across all consumers37.
As part of the scope of works for this assignment, ACIL Allen has been asked to assess the
effects of potential policy changes on retail electricity prices for households, SMEs and
larger electricity users. We have used the same approach as set out above in calculating
these retail electricity price effects which is as follows:
 Model the Reference case (obtain energy pool price projections and RET certificate
price projections across all Australian regions)
 Model the alternative policy case (obtain energy pool price projections and RET
certificate price projections across all Australian regions)
 Model any changes in network tariffs for changes in grid based demand
 Model the changes in various tariffs for each region (changes in RET certificate costs,
changes in network tariffs and changes in energy pool prices).
37 Aggregating net consumer benefits is not easily achieved because of the myriad of tariffs and tariff structures, the
variations in usage and usage patterns and the nature of retail contracts which may limit the pass through of any pool
savings that may occur.
128
Uncertainty
The net electricity consumer benefit tariff modelling is subject to uncertainty in the modelled
components. Pool prices are inherently uncertain. This is because many of the drivers of
pool prices are uncertain:
 Weather driving demand is stochastic (random) and highly variable
 Plant performance (outages) is also stochastic
 Fuel prices may vary over time although most fossil fuel fired plant tend to contract over
several years and so these prices tend to be reasonably certain on an annual basis
 Participant behaviour (mothballing, plant retirement, strategic bidding, etc.) may swamp
other effects over time.
The modelling used in our analysis is based on a single set of determined inputs (demand,
outages, fuel prices and participant behaviour). However, as discussed above, future pool
prices are subject to significant uncertainty and expectations about future annual average
prices (which typically feed into tariffs) are uncertain and are distributed across a range of
potential outcomes. Figure 118 shows a typical distribution for expected annual prices.
Figure 118 Representative annual pool price distribution
Distribution
Median
Number of observations
Mean
Time-weighted Av RRP (Nominal $/MWh)
Source: ACIL Allen
RET certificate prices are also uncertain but tend to be inversely correlated with expected
pool prices. This is because when pool prices rise, the subsidy required under the RET
scheme to bank renewable projects falls.
The inverse relationship between pool prices and RET certificate prices might lead one to
conclude that the net effect of these uncertainties is to cancel each other out and provide
relatively stable outcomes for the combination of pool and RET certificate prices. This is the
case for renewable projects where a RET certificate is generated along with each MWh of
renewable energy. However, even in the Reference case, the proportion of RET certificates
to MWh in the retail price is likely to be no more than one certificate in four MWh over the life
of the scheme. Therefore pool price uncertainty will have a much larger influence on retail
prices than the inversely related RET certificate prices.
Basis of policy
The objectives of the RET scheme as set out in the Renewable Energy (Electricity) Act 2000
are:
 to encourage the additional generation of electricity from renewable sources
 to reduce emissions of greenhouse gases in the electricity sector
129
 to ensure that renewable energy sources are ecologically sustainable.

Providing net electricity benefits to consumers is not a fundamental objective of the RET
scheme.
As noted above, net consumer benefits do not represent a full assessment of the economic
impact of a policy or the social benefits and costs of a policy. They represent the benefits
that flow to consumers without taking into account the opportunity costs of the capital and
labour involved and the other welfare effects of the policy.
Implementing the RET policy for the purposes of providing net benefits to electricity
consumers is equivalent to imposing a contract on electricity consumers to fund renewable
energy projects for the retail price benefits that may subsequently occur. Government
intervention in markets for this purpose might be valid where fundamental market failures
prevented electricity consumers being able to access competitive prices or where
fundamental behavioural market failures existed (i.e. myopia, split incentives etc.). However,
there is little evidence that any of these types of market failures exist with most regions of
Australia being considered to have very competitive functioning markets for electricity.
Centrally imposing costs on electricity consumers for potential future benefits does not take
into account the wide range of potential preferences of consumers including a preference to
avoid costly long term obligations where future benefits might not prevail.
A key factor in the uncertainty around future electricity prices is participant behaviour. As
electricity demand has fallen in recent years, an increasing willingness of participants to
mothball or close generation plant has been observed. Closing or mothballing plant can
cause a significant rebound in pool prices and may fully offset any downward pressure from
renewable plant. While we have incorporated some mothballing of plant in the analysis,
participants may have different objectives and take quite different views to mothballing and
plant closure than we have taken. This could substantially change the net benefit to
electricity consumers through net changes in retail prices as demonstrated through the
Permanent retirement sensitivity.
Table 13 shows the NPV of changes for each of the key policy cases with reference to the
Reference case expressed as a percentage of total retail costs over the period. All of the
alternative cases show a benefit to 2020 with some showing benefits to 2030 and beyond.
However, in all cases the benefits or costs are a very small percentage of the total bill and
could easily be swamped by the range of uncertainties in pool prices, especially the
changes in the behaviour of generation participants.
Table 13 Residential retail bills: change in NPV from Reference case

2015-20 2015-30 2015-40
Reference case 0.00% 0.00% 0.00%
Repeal case -3.22% -0.01% 0.61%
Closed to New Entrants case -1.80% 1.23% 1.59%
Real 20% case -0.32% 0.78% 0.94%
Real 30% case -0.22% -1.54% -1.56%
50% Growth case -0.27% 0.79% 0.94%
Note: NPV based on 7% discount rate. Negatives represent consumer savings or benefits
Source: ACIL Allen
130
Appendix A Methodology and assumptions

A.1 Methodology
A.1.1 Modelling process

Figure A1 provides an overview of the modelling process used to undertake the projections
for the Expert Panel under this project.
Figure A1 Modelling framework
Market characteristics
Off-grid database
- existing generator inputs
- new entrant options and costs - detailed breakdown of existing
SRES model generation ~ 10,000 GWh
Coverage:
- projected SWH - current renewable composition
NEM
- projected SGU (solar PV) approximately 2%
SWIS
DKIS
NWIS and Mt Isa
Demand inputs LRET settings

- market facing sent-out energy - targets and banking rules
and peak demand - existing generators
- 100 point sampled load PowerMark LT - new entrant generators
duration curves - certificate multipliers per MWh
Model outputs
- Emissions (CO2-e)
- Generation/capacity mix Retail electricity pricing model
- Fuel use - network costs
- Wholesale electricity prices - retail costs and margin
- LGC prices - hedging costs (based on load shape)
- LRET market surrenders/shortfalls - retail series for households
- New entry technology and build timing - retail series for SMEs/large users
- Retirements and refurbishments
- System resource costs
Source: ACIL Allen
A.1.2 PowerMark LT
PowerMark LT simulates the electricity market across Australia for existing generator
operation, new investment (entry) and retirement decisions (exit). PowerMark LT differs from
our highly detailed, short-term simulation model, PowerMark, but uses similar solving
algorithms to broadly represent the profit-maximising behaviour of energy market
incumbents and potential new entrants, thereby predicting prices, generation patterns and
emissions outcomes.
To aid computation, PowerMark LT uses fewer dispatch periods per model year than
PowerMark (typically 100 for PowerMark LT, compared to 8760, or one per hour, for
PowerMark). Accordingly PowerMark LT solves very quickly and can automatically optimise
generation new entry and dispatch outcomes over long time horizons on an inter-temporal
basis (that is, adjusting outcomes in all periods based on outcomes in all other dispatch
periods). Use of PowerMark LT implies that the market structures that are in place are
efficient and will result in the least cost outcome over the projection period.
PowerMark LT’s ability to solve quickly will ensure that project timeframes are able to be
met and sensitivities on key uncertainties tested. Whilst using PowerMark would provide
greater resolution on price and bidding behaviour, these benefits are generally only present
in modelling periods of less than a decade. Over longer periods either model tends to reach
a stable ‘new entrant’ equilibrium where average prices closely follow the long-run marginal
A1
cost of the most efficient combination of new entrants needed to meet base, intermediate
and peak growth in demand.
The workings of the LRET are fully incorporated into PowerMark LT, with eligible renewable
technologies able to create LGCs as a by-product of their electricity generation. The LRET
scheme’s settings such as the effective penalty price for non-compliance by liable entities
are included and incorporated into the objective function which is to minimise aggregate
resource costs over the period in question.
In terms of geographic scope, PowerMark LT is set up to model all of Australia’s major
electricity grids, namely:
 the National Electricity Market (NEM), covering New South Wales, Queensland,
Victoria, South Australia and Tasmania
 the South-West Interconnected System (SWIS), also known as the Western Australian
Wholesale Electricity Market (WEM), serving south-western Western Australia
including Perth, Geraldton and Albany
 the North-West Interconnected System (NWIS), the grid serving numerous mines and
towns in the Pilbara region of Western Australia
 the Darwin-Katherine Interconnected System (DKIS), the grid serving the more
populous parts of the Northern Territory
 the grid serving the area around Mt Isa in Queensland.
The locations of these electricity grids are shown below.
Figure A2 Geographic coverage of modelled grids
Source: ACIL Allen
This approach gives a detailed picture of renewable development prospects in all the
sizeable electricity grids in Australia. The modelling horizon is calendar years 2014 to 2040.
Generation from off-grid sources are not modelled with PowerMark LT, but rather estimated
using ACIL Allen’s off-grid database.
A2
A.1.3 Demand treatment

First a half-hourly demand trace for each region is developed based on forecast ‘native’
energy and peak summer and winter demands. These traces are weather corrected and
based on recent observed patterns in each market. A half-hourly output trace from the
SRES model is constructed based on projected uptake of each technology. Uptake is a
function of payback periods of solar PV for residential and business customers taking into
consideration the number of suitable dwellings and roof-space and saturation levels. Inputs
for the SRES model consist of system costs, retail electricity prices and government feed-in-
tariffs and upfront subsidies.
The projected output from behind-the-meter systems is deducted from the native demand
series at the half-hourly level. This yields a half-hourly series of market-facing demand for
each region to be modelled for the entire modelling horizon. This series is then rearranged
as load-duration curves and a sampling process is undertaken to represent the load duration
curve as a smaller subset of weighted demand points (typically 100). The purpose of this is
to reduce the LP problem size, but still retain the shape of demand throughout the year for
each region.
A.1.4 Supply
On the supply-side PowerMark LT inputs include:
 Definitions of the markets themselves including price limits; unserved energy and
reserve margin constraints
 Existing interconnector capacities and losses; candidates for interconnector
augmentations including capacity, cost and earliest timing
 A range of assumptions for existing generators including remaining technical life,
outage rates, maximum and minimum capacity factors, thermal efficiency, auxiliaries,
fixed and variable O&M costs, fuel costs, combustion and scope 3 emission intensities,
capacity, refurbishment costs
 Carbon prices and other policy settings
 New entrant technology candidates, includes generator characteristics, availability for
each region, annual and aggregate build limits, capital and operating costs
 A series of generic constraints which can be applied to range of model variables.
A.1.5 Maintenance of reliability standards

ACIL Allen conditions its PowerMark inputs such that outputs are designed to represent
median (or P50) wholesale market outcomes. Naturally, there are a range of stochastic
factors such as weather driven demand, unplanned outages and intermittent resource
availability (particularly wind and solar) that can results in materially different annual price
outcomes.38 The distribution of these price outcomes is skewed toward higher prices, such
that the mean (average) price outcome is above the median. It is unlikely we would see any
unserved energy under any median (P50) scenario.
Market operators such as AEMO and the IMO use reserve plant levels to determine when
the level of unserved energy may potentially breach the reliability criteria of 0.002% of
energy. These are developed through a comprehensive Monte Carlo modelling exercise
looking at forced outage rates, extreme peak demands and interconnector capacities to
38 ACIL Allen have developed a sophisticated method of profiling wind farms consistent with the standard weather corrected
demand profile - which has the advantage of providing realistic output on peak demand days.
A3
estimate a level of reserve plant that the operator can be confident that unserved energy will
remain below the threshold level.
Rather than replicate this time consuming process or undertake additional modelling using
more extreme P10 demand/outage cases, ACIL Allen compared resulting aggregate plant
capacities against these already developed reserve plant margins for each region. In this
context, intermittent plant capacity was adjusted for anticipated peak contribution as per the
recent AEMO Statement of Opportunities report.
The structure of the WEM in Western Australia is somewhat different in that the Reserve
Capacity Mechanism ensures that sufficient reserve plant margin is always available
through explicit capacity payments. However, the WEM structure is not explicitly
incorporated within PowerMark LT. Rather, the WEM is modelled to operate as an energy
only market.
A.1.6 LRET implementation in PowerMark LT

The LRET is explicitly incorporated into PowerMark LT, inputs comprise of:
 Certificate demand: mandated targets, voluntary demand through GreenPower and
desalination and other voluntary surrenders
 Supply from existing generators (both scheduled generators and non-scheduled
generators)
 Cumulative banked certificates at the start of the model horizon and certificate financial
holding costs
 Constraints on banking and borrowing, incorporating the three year “make-good”
allowance for liable entity shortfalls
 New entrants generators creating LGCs are those flagged as eligible technologies
within PowerMark LT’s new entrant list. A certificate multiplier (typically one) can be
applied for each generation technology per MWh of electrical output.
The key features of the LRET are implemented within PowerMark LT as discussed in the
following sections.
Banking/borrowing
As per the schemes design, unlimited banking of permits is allowed. That is, permits created
can be created and withheld for surrender in later years. PowerMark LT allows an unlimited
number of LGCs to be banked throughout the scheme. Note that all banked LGCs up until
the end of calendar year 2010 will be eligible to be used against the LRET, regardless of
how they were created.
Borrowing under the scheme is effectively limited to 10% of each liable entity’s liability.39
This provision is provided because it is often difficult for a retailer to accurately predict what
its liability will be. The 10% provides liable parties some leeway in estimating liabilities. With
perfect foresight, this provision could be gamed, with liable parties only surrendering 90% of
required LGCs and carrying forward the shortfall. As the target increases, this could become
significant. However in reality this would be a difficult strategy to employ and therefore the
model does not allow for any explicit borrowing of certificates.
39 Renewable Energy (Electricity) Act 2000, Section 36(2)
A4
Shortfall penalty
The shortfall charge as specified within the regulation is $65 per MWh not-indexed (constant
in nominal terms over the life of the scheme). As penalties paid are not deductible business
expenses (they are treated as fines), the effective pre-tax penalty is therefore $92.86/REC
($65/(1-30%), assuming a 30% marginal tax rate). The penalty is not indexed so it declines
in real terms over the period to 2030.
Certificate demand
There are three sources of demand for LGCs: demand for LGCs to offset mandatory
obligations under the scheme, LGCs to acquit GreenPower sales40 and certificates
associated with desalination plants/other voluntary schemes. While the requirement to
surrender LGCs applies to each individual entity, PowerMark LT treats the demand-side as
a single entity. As such, it does not distinguish between parties and their respective LGC
positions.41
It also assumes there is zero mandated demand for LGCs at prices above the tax-adjusted
shortfall penalty price. While some have suggested liable entities may be willing to buy
certificates at prices above these levels to avoid reputational damage, the model does not
explicitly account for this. Note that the demand figures include the 850 GWh allowance for
Waste Coal Mine Gas (WCMG) to 2020. This is offset by a 750 GWh supply-side
assumption for pre-existing WCMG operators42, such that the inclusion has very little impact
upon LRET outcomes.
Figure A3 Reference case total LGC demand to 2030
50,000
45,000
40,000
35,000
30,000
LGCs ('000)
25,000
20,000
15,000
10,000
5,000
0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Mandated demand WCMG GreenPower Desalination Other
Note: Mandated demand includes adjustment under Section 40 (1A) of the Act
40 Voluntary GreenPower demand is assumed to remain at 1,500 GWh per annum over the period to 2030.
41 Another way of thinking of this is that all parties freely trade with one another without any transaction costs.
42 This is based on observed LGC creation to-date by these accredited generators under the scheme.
A5
Certificate supply
The modelling considers two types of certificate supply: existing/committed accredited

generators and potential new entrants.
Contribution from existing accredited generators and those under construction are done at
the individual power station level. For most, this involves projecting LGC creation rates at
levels similar to recent history. Those that are currently under construction have
assumptions about commissioning timing and production ramp up.
A range of specific projects and various generic new entrant technologies are presented to
the model for deployment. With a number of the smaller, niche renewables technologies, it
is difficult to project deployment when modelling the LRET at the macro level. These
include:
 Landfill gas where projects are very site specific and local transmission connection
costs can be a significant component of capital costs. Ultimately the resource base is
limited by suitable landfill sites
 Bagasse where projects are mill specific and the timing of which, is determined by the
need for mill refurbishment more so than the economics of the cogeneration units. The
resource base is also limited by the amount of sugar cane crop processed.
 Wood and wood waste plants which are typically small-scale developments where
feedstock availability and network connection are key variables. Larger projects (such
as the project proposed by Gunns’ Bell Bay) are reliant upon the underlying paper mill
development rather than the economics of generation. Fuel transport and handling
costs typically are constraining factors.
 Embedded solar PV systems above the current 100 kW LRET cut-off (but not
considered utility scale). These mostly occur in remote off-grid locations, particularly in
regions where diesel generation displacement can occur.
 Other technologies such as those using agricultural/food wastes and municipal wastes
which are small and it is often difficult to obtain generic representative capital cost
estimates.
To account for uptake of these technologies, ACIL Allen makes projections of LRET
contribution based on historical growth and ultimate resource potential rather than explicitly
‘modelling’ deployment through PowerMark LT.
A.1.7 Supply-demand balance

Figure A4 shows historical and projected LGC creation from existing renewable generators,
generators that are under construction, from waste coal mine gas (WCMG) generators
entitled to create LGCs, and from niche small-scale generators such as landfill gas, bagasse
and small-scale solar above 100 kW but below utility scale. Figure A4 also shows aggregate
demand for LGCs over the period to 2030 as defined by the annual legislated LRET target.
PowerMark LT seeks to fill the gap between committed and assumed future supply and
demand by deploying further LGC-eligible generation at least cost over the period to 2030
and explicitly considers the economics of those installations for the period beyond 2030.
A6
Figure A4 LGC supply demand balance 2001 to 2030
50,000
45,000
40,000
35,000
LGCs ('000)
30,000
25,000
20,000
15,000
10,000
5,000
Existing generators Under construction Waste coal mine gas

Solar hot water SGUs Assumed niche LGCs
Total Demand
Note: Includes assumed contribution from existing, under construction, waste coal mine gas and niche technologies (Landfill gas, Bagasse,
Wood, Sewage Gas, and embedded solar PV above 100 kW in size). Historical REC Registry data current to 20 March 2013
A.1.8 Output from baseline generators

According to the CER the aggregate baseline for accredited large-scale generators is
currently 16,598 GWh.43 Of this, pre-existing hydro accounts for around 94% of this
generation, with other sources including bagasse, landfill gas and wood amongst others as
shown in Figure A5. The aggregate baseline has changed only very slightly since 2002 with
the CER modifying values based on new accreditations or revisions to baselines of plants
which were in operation prior to the scheme start and de-accreditation of baselined stations
due to retirements.
Figure A5 Baseline generation by primary fuel type for 2014
18,000
15,624.7
16,000
14,000
12,000
10,000
GWh
8,000
6,000
4,000
2,000 492.9 263.7 18.8 5.5 192.2 0.1
0
Bagasse Hydro Landfill Gas Sewage Wind Wood Other
Gas
Source: ACIL Allen based on CER REC Registry data
A consideration in determining the level of renewable generation by 2020 is the level of

generation from pre-existing stations which have baselines. As much of this is hydro whose
43 Clean Energy Regulator, REC Registry (https://www.rec-registry.gov.au/home.shtml), accessed 10 July 2014.
A7
output varies year-to-year, we rely on long-term average generation volumes, with average
generation assumed to occur every year.44
Figure A6 provides a breakdown of the current baseline into four components: Snowy hydro;
Hydro Tasmania (registered as Hydro-Electric Corporation); ‘Other’ hydro and the remaining
fuel types which are included into the ‘Other’ category.
Figure A6 Baseline breakdown
10,000 8,884.5
9,000
8,000
7,000
6,000
GWh
5,000 4,491.9
4,000
3,000 2,248.3
2,000 973.1
1,000
0
Snowy Hydro Limited Hydro-Electric Other hydro Other
Corporation
Note: ‘Other’ category includes all other non-hydro technologies

The modelling assumes long-term output from Snowy hydro facilities of 4,700 GWh and
from hydro Tasmania of 9,100 GWh. This aligns with ACIL Allen’s experiences working with
these entities and historical performance.45 So in this regard, modelled output exceeds
baseline levels in every year of the projection period for these facilities. Accepting this
already accounts for over 13,300 GWh of the total baseline energy.
Not all facilities within the ‘Other hydro’ and ‘Other’ categories have publicly available
dispatch figures so for some it is difficult to determine historic generation relative to
baselines. One way is to examine whether these categories create LGCs – if a baselined
facility does, it is, by definition, generating above its baseline levels. Figure A7 shows LGC
creation from these categories historically. This demonstrates that at least in an aggregate
sense, output from these categories has exceeded historical baseline levels in all years.
44 An allowance is made within the LRET analysis for historical LGC creation from baseline generators in the supply mix in
accordance with average historical LGC creation. This allowance takes into account the historical experience and potential
for ‘gaming’ baselines by energy constrained generation such as hydro.
45 Some commentators have suggested output from these facilities will be lower based on analysis of output through 2005 to
2010, however this period included severe drought conditions which is not consistent with long-term expectations.
A8
Figure A7 Historical LGC creation from other hydro and other sources
1,200
Other hydro Other
1,000
800
LGCs ('000)
600
400
200
0
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
However, as for the large Tasmanian and Snowy Hydro schemes, it is possible that output
from individual stations has been below baseline levels, whilst others are generating above
baseline levels, resulting in LGC creation. For the larger hydro stations, historical dispatch
data is available. Figure A8 shows a comparison of average historical generation (2001-
2013) against individual baselines.46 Where data is available, average output has been
similar to baseline values. Kareeya and McKay Creek stations have both exceeded
baselines on average over the period, whilst Barron Gorge dispatch has been approximately
the same. In other cases where data is available, average historical generation has been
50-75% of baseline levels.
Figure A8 Historical generation from ‘Other hydro’ stations

500,000
450,000
Baseline (MWh) Average Generation (MWh)
400,000
350,000
300,000
MWh
250,000
200,000
150,000
100,000
50,000
0
Note: Average generation volumes presented for those where historical generation data is available. Historical data from 2001-2013
Source: ACIL Allen based on CER and AEMO data
Similar analysis was also carried out for other fuel types at an individual station level where
possible. In the ‘Other’ category, stations accounting for over 750 GWh of the total 973 GWh
of baselined energy created LCGs in calendar year 2013. Generation data from stations
which did not create LGCs was generally not available, but could have ranged from zero to
just under the individual station baseline.
46 This period covers both the drought period 2005-2009 and also artificially higher dispatch periods more recently due to
carbon pricing, with the two events largely cancelling each other out.
A9
Based on examination of individual station LGC creation, ACIL Allen has made an
allowance of around 2,770 GWh of the combined 3,200 GWh of baseline energy from the
‘Other hydro’ and ‘Other’ categories in its assessment of baseline energy.
In total therefore, ACIL Allen makes an allowance for a total of 16,148 GWh of the total
baseline energy (a reduction of 450 GWh from the total of 16,598 GWh).
ACIL Allen is aware of historical analysis by some stakeholders that indicate total ‘below-
baseline’ generation has been up to 2,500 GWh less than the baseline levels for the Snowy
and Tasmanian Hydro systems. To some extent historical behaviour will have been
influenced by drought conditions affecting these systems (five years over 2005-2009 were
particularly affected), but potentially also through inter-temporal ‘gaming’ of the baselines,
although commercial and regulatory factors would tend to work against this.
Certificate creation from baseline generators

For accredited renewable generators that are not explicitly modelled through PowerMark LT,
ACIL Allen makes assumptions regarding future LGC creation from these facilities. This is
done on a station-by-station basis.
Figure A9 presents the assumed LGC creation from baseline generators, broken down into
the four components from the previous section. Aggregate LGC creation from baselined
hydro generators is around 1.1 million per year. The ‘Other’ category which includes
bagasse, landfill gas amongst others, is assumed to create around 1 million LGCs per
annum, with this declining slightly over time based on projections of individual stations,
some of which exhibit declining output as resources are depleted (for example landfill gas
sites).
A10
Figure A9 Assumed certificate creation from baseline generators

Snowy Hydro Limited Hydro-Electric Corporation
1,800 3,000
1,600
2,500
1,400
1,200 2,000
LGCs ('000)
LGCs ('000)
1,000
1,500
800
600 1,000
400 324 535
500
200
0 0
Other hydro Other

700 1,200
1,041
600 1,000
500
800
LGCs ('000)
LGCs ('000)
400
600
300 256
400
200
100 200
0 0
Source: ACIL Allen based on REC Registry data
A.1.9 PowerMark simulation model

ACIL Allen has also employed its more detailed simulation model – PowerMark – in
modelling the NEM. This is due to the volatile nature of the price formation process of the
NEM’s energy-only spot market which is difficult to replicate in long-term planning models.
PowerMark simulates the NEM on a time sequential basis using hourly resolution input data.
The simulation model was employed alongside PowerMark LT for select spot years. This
was to ensure that results from the planning model are realistic and any nuanced results
relating to the intermittency of renewables is properly accounted for within PowerMark LT.
Use of the simulation model also aided the construction of retail hedging costs in building up
the retail price paths and assisted with emulating strategic actions by generator portfolios
such as retirement/mothballing decisions.
A.1.10 Outputs
Outputs from PowerMark LT consist of aggregate output variables, as well as detailed
results down to individual generator level. Wholesale price series are extracted for each
point on the sampled load duration curve, and these are weighted appropriately to produce
load-weighted and time-weighted wholesale price series.
ACIL Allen’s methodology provides the following reporting outcomes:
 Renewable energy generation (in GWh) and capacity (in MW) by technology (covering
large and small scale and including solar/heat pump water heaters), and overall
A11
 Generation of electricity from non-renewable resources (GWh) and capacity (MW), by

technology and overall
 The share of renewables as a percentage of total Australia-wide electricity supply
 The quantity of Large-scale Generation Certificates (LGCs) and Small-scale
Technology Certificates (STCs) created and surrendered, net stock and price
(spot/forward and bilateral contract-based) and whether liable entities are projected to
pay the shortfall charge rather than purchase LGCs
 Investment in generation capacity by technology and in aggregate, including
regional/urban split of the renewable generation capacity (both large and small scale).
Analysis of investment in scalable technologies such as solar, hydro, geothermal and
wind should also distinguish between installations by households, commercial
businesses and large-scale power generators
 Greenhouse gas emissions from the electricity generation sector, noting that it
excludes non-scheduled generation, self-generation and off-grid generators
 The resource cost of the scheme
 The resource cost of abatement for the RET, based on two different methodologies for
treating the emissions component of the calculation
 Wholesale electricity prices
 Retail electricity prices for households, SMEs and larger electricity users
 Net revenues from generation technology types for the purpose of calculating
profitability measures.
A.2 Macro inputs

The modelling of the various policy scenarios utilises consensus central assumptions for all
input variables. The following sections outline the key assumptions used.
A.2.1 Electricity demand

Electricity demand for each scenario is determined exogenously and presented to the
models. Demand is not varied across the modelled scenarios in light of changes to the RET.
This is due to a number of reasons:
 Based on recent publicly released modelling commissioned by the AEMC and the
Climate Change Authority, expected retail price impacts from the RET are relatively
small as the RET only comprises a small component of retail bills.
 Price changes for Emissions Intensive Trade Exposed loads, which are probably the
most price sensitive sector, are largely insulated from RET effects through the partial
exemption framework
 Any modifications to demand in light of the RET are likely to be very small when
considering the potential range of forecast error over this time period.
NEM
AEMO has provided ACIL Allen with an updated energy forecast for use within this project
which have formed the basis of the 2014 National Electricity Forecasting Report (NEFR).47
The data provided includes Native and Operational energy forecasts by NEM region and
also assumptions in relation to the level of energy generated from behind the meter solar PV
47 Note that at the time the modelling was undertaken, the 2014 NEFR had not been released.
A12
systems. This data represents AEMO’s view of expected energy growth under Medium
economic conditions (i.e. the mid-case) and High/Low energy growth cases.
AEMO has confirmed that forecast PV output is presented at the meter, so adjustments are
required for network losses to gross this energy up to an Operational level.
Figure A10 shows a comparison of the latest energy forecast against the previous published
figure by AEMO. Estimates for financial year 2013-14 are around 2% lower than those
estimated in November 2013. The energy is around 10% lower by 2020 (-21,500 GWh) and
14% lower by 2032 (-31,200 GWh) compared with the previous figures.
Figure A10 Comparison of latest AEMO energy forecast against 2013 NEFR
250,000 0%
-2%
200,000
-4%
Percentage change
Sent-out GWh
-6%
150,000
-8%
100,000
-10%
-12%
50,000
-14%
0 -16%
2013 NEFR New figures % Change (RHS)
Note: NEFR = National Electricity Forecasting Report. Energy presented on a Native, sent-out basis. 2013 NEFR figure for 2013-14 represents
the updated lower value published in November 2013.
Source: ACIL Allen based on AEMO
Beyond the period of AEMO forecasts, (from 2034-35 to 2040-41) electricity demand in each
NEM region is assumed to continue to grow at the same year-on-year rates as per the last
five years of the AEMO forecast (0.3% per annum).
ACIL Allen has scaled down the peak summer and winter demand forecasts for each region
from the 2013 NEFR by the same level as the energy reduction (i.e. the load factor will
remain the same) as AEMO was unable to provide any update to the peak demand
forecasts at the time the assumptions were finalised.
SWIS
The Independent Market Operator (IMO) has also provided an update to its forecast energy
and peak demand for the purposes of this study. The revised forecasts are actually a slight
increase over the 2013 Statement of Opportunities (SOO) report however the energy value
for the final year (2023-24) has been left unchanged as shown in Figure A11.
A13
Figure A11 Comparison of IMO energy forecast 2013 SOO and 2014 values
23,000
22,000
21,000
GWh sent-out
20,000
19,000
18,000
17,000
16,000
15,000
2014 2013
Note: Market demand net of solar PV assumed by IMO

Source: ACIL Allen based on IMO data
Beyond the period of IMO forecasts, electricity demand in the SWIS would be assumed to
continue to grow at the same year-on-year rate as per the last 5 years of the IMO forecast
(1.97% per annum).
There is also some generation and consumption on the SWIS grid that falls outside the
IMO’s forecast. This largely relates to on-site self-generation by steam turbines at Alcoa’s
alumina plants. This energy is not measured and published publicly. ACIL Allen has made
estimates of this energy separately and hold constant into the future.
A.2.2 Other grids and off-grid generation

Other key grids modelled explicitly by ACIL Allen are the North-West Interconnected System
(NWIS) or Pilbara grid, the Darwin-Katherine Interconnected System (DKIS) and the grid
servicing Mt Isa. ACIL Allen has recently undertaken an extensive analysis of off-grid
electricity generation for the Commonwealth Government, published as a Bureau of
Resources and Energy Economics (BREE) report in October 2013. This study directly
estimated the amount of electricity sent out to these grids in 2011-12 using a range of
sources, in particular data from the National Greenhouse and Energy Reporting scheme
(NGERs). The use of NGERs data means that this study offers an accurate and
comprehensive estimate of demand in 2011-12, from which future demand can be
estimated. ACIL Allen has grown demand for each of these grids using growth rates within
our 2013 emissions projections work.
Similarly, the 2013 off-grid study for BREE estimated 2011-12 electricity consumption (on a
sent out basis) for off-grid areas of Western Australia (excluding the NWIS), Northern
Territory (excluding the DKIS), Queensland (excluding Mt Isa), New South Wales, Victoria,
South Australia and Tasmania.
Table A1 illustrates the 2011-12 demand assumption (on a sent out basis) and source of
growth assumptions for regional grids and off-grid regions.
A14
Table A1 Regional grid and off-grid demand assumptions

First year using Average emissions
2011-12 electricity
Region Growth assumptions emissions projection projection growth rate to
(GWh)
growth rates 2040-41
Emissions projections
NWIS 2,467 growth rates for north- 2012-13 2.4%
west WA
WA excluding SWIS and NWIS 6,414 growth rates for north- 2012-13 2.4%
west WA
DKIS 1,530 2012-13 2.3%
growth rates for NT
NT excluding DKIS 1,744 2012-13 2.3%
growth rates for NT
Mt Isa 2,239 2012-13 2.1%
growth rates for Mt Isa
Queensland excluding Mt Isa Emissions projections
1,160 2012-13 2.1%
and NEM connected areas growth rates for Mt Isa
AEMO for SA, then
South Australia excluding NEM
236 emissions projections 2034-35 1.3%
connected areas
growth rates
AEMO for TAS, then
Tasmania excluding NEM
22 emissions projections 2034-35 1.3%
connected areas
growth rates
AEMO for relevant
New South Wales and Victoria state, then emissions 1.7% (NSW);
35 2034-35
excluding NEM connected areas projections growth 1.9% (VIC)
rates
Source: BREE 2013, Department of Environment
In addition, self-generation by other major loads were included in this forecast. These
include:
 Embedded generation (generators which are not included within the AEMO/IMO
figures)
 Generation at LNG liquefaction sites (Gladstone, Darwin and regional WA)
 industrial gas- or waste-fired generation or cogeneration in steel/chemical-
petrochemical/pulp-paper industries.
A.2.3 Aggregate Australian demand

Figure A12 provides the composition of aggregate Australian electricity demand on a
calendar year basis for 2014 to 2040 and a comparison against BREE’s aggregate
consumption figure for Australia. The estimated electricity consumption for Australia in 2020
is 255,253 GWh.48
48 This excludes estimated electricity displaced by solar water heaters
A15
Figure A12 Aggregate projected Australian electricity demand
350,000
300,000
250,000 Off-grid
Embedded/self-gen
Sent-out GWh
200,000 Solar PV
Mt Isa
150,000 DKIS
NWIS
100,000
SWIS
NEM
50,000
0
2018
2023
2035
2040
2013
2014
2015
2016
2017
2019
2020
2021
2022
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2036
2037
2038
2039
Source: ACIL Allen, AEMO, IMO, BREE
A.2.4 Foreign exchange rates

The assumed real exchange rate series used is shown in Figure A13. This aligns with the
near-term nominal exchange rates provided within the 2014-15 Federal budget. All foreign
equipment is assumed to be priced in US$, despite the fact that some components for wind
turbines are sourced and purchased in Euros.
Figure A13 Real exchange rate US$/AUD
0.98
0.97
0.96
Real 2009-10 US$/$A
0.95
0.94
0.93
0.92
0.91
0.90
0.89
0.88
2014-15
2024-25
2027-28
2013-14
2015-16
2016-17
2017-18
2018-19
2019-20
2020-21
2021-22
2022-23
2023-24
2025-26
2026-27
2028-29
2029-30
2030-31
2031-32
2032-33
2033-34
2034-35
2035-36
2036-37
2037-38
2038-39
2039-40
Source: ACIL Allen based on 2014-15 Budget and RET Review input
A.2.5 Commodity prices and wage indices

Commodity prices and indices were used as part of the input assumptions in the 2013
emissions projections work and are also used within this work as modifiers to capital costs.
These include ‘Metals’ and domestic wage price indices. The gas prices used in this work
were based on the International Energy Agency’s (IEA) 2013 World Energy Outlook. Within
this publication the IEA produces two gas price series: the New Policies Scenario (which
included global policies to achieve greenhouse gas abatement) and a Gas Price
A16
Convergence Scenario in which global prices converged on the basis of increased exports
from the United States. These projections are shown in Figure A14.
Figure A14 IEA projected global gas prices
Note: Presented in US dollars. 1 Mbtu = 1.0546 GJ

Source: International Energy Agency, 2013 World Energy Outlook
ACIL Allen has utilised the prices for Japan from the Gas Price Convergence Scenario in
calculating the equivalent netback series for domestic gas prices. Figure A15 shows the
calculated netback from the gas price series. Net backs are calculated by deducting from
the Japanese LNG import price:
 An allowance for LNG shipping of US$0.90/mmbtu
 Liquefaction capex component of US$3.00/mmbtu
 Assuming 9% of gas used in liquefaction.
This results in the Australian ‘net back’ price being around A$5.80/GJ lower than the
Japanese import price. These prices are then translated throughout the gas network by
adjusting for transport differentials. Existing gas-fired stations transition from prices under
existing contracts to export netback levels over the first 2-3 years of the modelling.
Figure A15 Calculated netback gas price for Australian gas producers
18.00
16.00
14.00
Real 2014 $/GJ
12.00
10.00
8.00
6.00
4.00
2.00
0.00
2018-19
2024-25
2030-31
2014-15
2015-16
2016-17
2017-18
2019-20
2020-21
2021-22
2022-23
2023-24
2025-26
2026-27
2027-28
2028-29
2029-30
2031-32
2032-33
2033-34
2034-35
2035-36
2036-37
2037-38
2038-39
2039-40
Netback price at LNG plant A$/GJ LNG import price (Japan) US$/mmbtu
Source: ACIL Allen
A17
Within our coal model, export thermal coal prices are brought back to the mine by deducting
port, rail and washery costs and adjusting for yield (thermal coal for domestic power
generation is generally unwashed coal and hence does not incur washery losses). Coal
costs for new entrant coal-fired generators are shown in Figure A16.
Figure A16 Assumed new entrant coal costs
3.50
3.00
Real 2014 $/GJ 2.50
2.00
1.50
1.00
0.50
0.00
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
NSW QLD VIC WA
Source: ACIL Allen
Future domestic inflation is assumed to be constant at 2.5% per annum.
A.2.6 Government policy settings

The Clean Energy Future Act 2011 (Commonwealth) implemented a multi-sectoral, nation-
wide carbon price, with the core effect of imposing a cost on emitting greenhouse gases
from a range of activities including fossil fuel combustion (e.g. in electricity generation) and
fugitive emissions from gas production and coal mining. The Reference case modelling
reflects repeal of this and supporting legislation from 1 July 2014.
The Expert Panel has instructed ACIL Allen to assume that no explicit carbon pricing
scheme is assumed to be developed in the period to 2040 under the Reference or Policy
cases. A potential future (post-2020) carbon price is considered as a separate sensitivity.
While the Emissions Reduction Fund (ERF) could have some impact in electricity markets,
depending on its final design, there was a lack of information at the time of modelling about
the nature and magnitude of any impacts. Consequently, the ERF is not incorporated into
the modelling.
A.2.7 Conventional coal entry

ACIL Allen’s standard assumption when undertaking market outlook studies is to restrict
conventional coal (i.e. coal-based technologies which do not employ carbon capture and
storage) from entry. This is due to:
 Community views and corporate sustainability policies
 Potential difficulty obtaining generation licences from State and Territory governments
 Long-term risk of explicit carbon pricing being reintroduced
 Difficulty in securing financing on a commercial basis due to these risks.
A18
As requested by the Expert Panel, conventional coal entry will be permitted in the policy
scenarios based purely on the economics of the technology within the market. As ACIL
Allen’s modelling assumes perfect foresight though, it ignores the risk of future carbon
policies which may have a negative effect upon the commercial performance of such plant.
A.2.8 Price basis

All price results are presented in real mid-2014 dollars unless otherwise stated.
A.3 Existing generators

The modelling incorporates a total of around 190 existing generators across the nine regions
modelled as shown in Table A2. For the NEM, these generators represent those that are
scheduled and semi-scheduled (i.e. those that report and participate in AEMO’s central
dispatch functions). These are generally those generators with a nameplate capacity above
30 MW in the NEM and above 1 MW in the SWIS.
Non-scheduled, embedded ‘behind the meter’ and off-grid generation are handled outside of
PowerMark LT.
For the SWIS, the generators and their capacity corresponds with capacity offered to the
Independent Market Operator (IMO) as part of the wholesale markets net pool functions.
This means that capacity and energy related to own-use consumption (most notably from
cogeneration projects) is not included explicitly and is handled outside the modelling.
For NWIS, DKIS and Mt Isa regions no formal market structure exists and generators
include all major grid connected plants.
Table A2 Existing and committed generators: type, capacity and life

Technical
Technical Capacity
Region Generator Plant type Fuel type Commissioned Retirement
Life (Years) (gross MW)
Year
AGL SF PV Broken Hill Solar PV Solar 2014 30 2044 53

AGL SF PV Nyngan Solar PV Solar 2014 30 2044 106
Bayswater Subcritical pf Black coal 1983 53 2036 2,720
Bendeela Pumps Pump n/a 1977 150 2127 240
Blowering Hydro Hydro 1969 150 2119 80
Boco Rock WF Wind turbine Wind 2014 25 2036 113
Colongra OCGT Natural gas 2009 30 2039 664
Eraring Subcritical pf Black coal 1983 50 2033 2,880
Gullen Range WF Wind turbine Wind 2014 25 2036 165.5
Gunning WF Wind turbine Wind 2011 25 2036 47
Guthega Hydro Hydro 1955 150 2105 60
Hume NSW Hydro Hydro 1957 150 2107 29
Hunter Valley GT OCGT Liquid fuel 1988 30 2018 51
NSW
Liddell Subcritical pf Black coal 1972 60 2032 2,100
Mt Piper Subcritical pf Black coal 1993 50 2043 1,340
Redbank Subcritical pf Black coal 2001 50 2051 150
Shoalhaven Bendeela Hydro Hydro 1977 150 2127 240
Smithfield CCGT Natural gas 1997 30 2027 176
Tallawarra CCGT Natural gas 2009 30 2039 430
Taralga WF Wind turbine Wind 2014 25 2036 106.8
Tumut 1 Hydro Hydro 1959 150 2109 616
Tumut 3 Hydro Hydro 1973 150 2123 1,500
Tumut 3 Pumps Pump n/a 1973 150 2123 400
Uranquinty OCGT Natural gas 2009 30 2039 664
Vales Point B Subcritical pf Black coal 1978 50 2028 1,320
Wallerawang C Subcritical pf Black coal 1978 45 2023 960
A19
Technical
Technical Capacity
Year
Woodlawn WF Wind turbine Wind 2011 25 2036 48

Barcaldine CCGT Natural gas 1996 30 2026 55
Barron Gorge Hydro Hydro 1963 150 2113 60
Braemar 1 OCGT Natural gas 2006 30 2036 504
Braemar 2 OCGT Natural gas 2009 30 2039 459
Callide B Subcritical pf Black coal 1989 50 2039 700
Callide C Supercritical pf Black coal 2001 50 2051 810
Collinsville Subcritical pf Black coal 1998 30 2028 190
Condamine CCGT Natural gas 2009 30 2039 140
Darling Downs CCGT Natural gas 2010 30 2040 630
Gladstone Subcritical pf Black coal 1980 50 2030 1,680
Kareeya Hydro Hydro 1958 150 2108 81
Kogan Creek Supercritical pf Black coal 2007 50 2057 750
QLD Mackay GT OCGT Liquid fuel 1975 45 2020 34
Millmerran Supercritical pf Black coal 2002 50 2052 851
Mt Stuart OCGT Liquid fuel 1998 40 2038 418
Oakey OCGT Natural gas 2000 30 2030 282
Roma OCGT Natural gas 1999 30 2029 80
Stanwell Subcritical pf Black coal 1995 50 2045 1,440
Swanbank E CCGT Natural gas 2002 30 2032 385
Tarong Subcritical pf Black coal 1985 50 2035 1,400
Tarong North Supercritical pf Black coal 2002 50 2052 443
Townsville CCGT Natural gas 2005 30 2035 240
Wivenhoe Hydro Hydro 1984 150 2134 500
Wivenhoe Pump Pump n/a 1984 150 2134 480
Yarwun Cogeneration Natural gas 2010 30 2040 168
Bluff WF Wind turbine Wind 2011 25 2036 53
Clements Gap WF Wind turbine Wind 2008 25 2033 57
Dry Creek OCGT Natural gas 1973 45 2018 156
Hallett OCGT Natural gas 2002 30 2032 200
Hallett 2 WF Wind turbine Wind 2008 25 2033 71
Hallett WF Wind turbine Wind 2007 25 2032 95
Ladbroke Grove OCGT Natural gas 2000 30 2030 80
Lake Bonney 2 WF Wind turbine Wind 2008 25 2033 159
Lake Bonney 3 WF Wind turbine Wind 2010 25 2035 39
Mintaro OCGT Natural gas 1984 30 2014 90
North Brown Hill WF Wind turbine Wind 2011 25 2036 132
Northern Subcritical pf Brown coal 1985 50 2035 530
SA
Osborne CCGT Natural gas 1998 30 2028 180
Pelican Point CCGT Natural gas 2000 35 2035 485
Playford B Subcritical pf Brown coal 1960 60 2020 231
Port Lincoln OCGT Liquid fuel 1999 30 2029 74
Quarantine OCGT Natural gas 2002 30 2032 216
Snowtown 2 North WF Wind turbine Wind 2014 25 2039 144
Snowtown 2 South WF Wind turbine Wind 2014 25 2039 126
Snowtown WF Wind turbine Wind 2008 25 2033 99
Snuggery OCGT Liquid fuel 1997 30 2027 63
Torrens Island A Steam turbbine Natural gas 1967 52 2019 480
Torrens Island B Steam turbine Natural gas 1977 50 2027 800
Waterloo WF Wind turbine Wind 2011 25 2036 111
Bastyan Hydro Hydro 1983 150 2133 80
Bell Bay Subcritical pf Natural gas 1971 38 2009 240
Bell Bay Three OCGT Natural gas 2006 30 2036 120
Cethana Hydro Hydro 1971 150 2121 85
TAS Devils Gate Hydro Hydro 1969 150 2119 60
Fisher Hydro Hydro 1973 150 2123 43
Gordon Hydro Hydro 1978 150 2128 432
John Butters Hydro Hydro 1992 150 2142 144
Lake Echo Hydro Hydro 1956 150 2106 32
A20
Technical
Technical Capacity
Year
Lemonthyme_Wilmot Hydro Hydro 1970 150 2120 82

Liapootah_Wayatinah_Catagunya Hydro Hydro 1960 150 2110 170
Mackintosh Hydro Hydro 1982 150 2132 80
Meadowbank Hydro Hydro 1967 150 2117 40
Musselroe WF Wind turbine Wind 2013 25 2038 168
Poatina Hydro Hydro 1964 150 2114 300
Reece Hydro Hydro 1986 150 2136 231
Tamar Valley CCGT Natural gas 2010 30 2040 200
Tamar Valley GT OCGT Natural gas 2009 30 2039 58
Tarraleah Hydro Hydro 1938 150 2088 90
Trevallyn Hydro Hydro 1955 150 2105 80
Tribute Hydro Hydro 1994 150 2144 83
Tungatinah Hydro Hydro 1953 150 2103 125
Anglesea Subcritical pf Brown coal 1969 52 2021 160
Bairnsdale OCGT Natural gas 2001 30 2031 92
Dartmouth Hydro Hydro 1960 150 2110 158
Eildon Hydro Hydro 1957 150 2107 120
Energy Brix Subcritical pf Brown coal 1960 58 2018 195
Hazelwood Subcritical pf Brown coal 1968 63 2031 1,640
Hume VIC Hydro Hydro 1957 150 2107 29
Jeeralang A OCGT Natural gas 1979 50 2029 228
Jeeralang B OCGT Natural gas 1980 50 2030 255
Laverton North OCGT Natural gas 2006 30 2036 312
Loy Yang A Subcritical pf Brown coal 1986 50 2036 2,180
VIC Loy Yang B Subcritical pf Brown coal 1995 50 2045 1,050
Macarthur WF Wind turbine Wind 2013 25 2038 420
McKay Hydro Hydro 1980 150 2130 300
Mortlake OCGT Natural gas 2011 40 2051 566
Mt Mercer WF Wind turbine Wind 2014 25 2039 131
Murray Hydro Hydro 1968 150 2118 1,500
Newport Steam turbine Natural gas 1980 50 2030 500
Oaklands Hill WF Wind turbine Wind 2011 25 2036 63
Somerton OCGT Natural gas 2002 30 2032 160
Valley Power OCGT Natural gas 2002 30 2032 300
West Kiewa Hydro Hydro 1956 150 2106 62
Yallourn Subcritical pf Brown coal 1980 55 2035 1,538
Albany Wind turbine Wind 2001 25 2026 22
Alcoa Kwinana Cogen Cogeneration Natural gas 1998 30 2028 5
Alcoa Pinjarra Cogen Cogeneration Natural gas 1985 35 2020 10
Alcoa Wagerup Cogen Cogeneration Natural gas 1990 30 2020 25
Bluewaters Subcritical pf Black coal 2009 40 2049 441
BP Cogen Cogeneration Natural gas 1996 30 2026 81
Canning/Melville LFG Reciprocating engine Landfill gas 2007 15 2022 9
Cockburn CCGT Natural gas 2003 30 2033 246
Collgar WF Wind turbine Wind 2012 25 2037 206
Collie Subcritical pf Black coal 1999 40 2039 333
Emu downs Wind turbine Wind 2006 25 2031 80
SWIS Geraldton OCGT Distillate 1973 40 2013 21
Grasmere Wind turbine Wind 2012 25 2037 14
Greenough River Solar PV Solar 2012 30 2042 10
Kalgoorlie OCGT Distillate 1990 30 2020 63
Kalgoorlie Nickel OCGT Natural gas 1996 30 2026 10
Kemerton OCGT Natural gas 2005 30 2035 310
Kwinana A Steam turbine Natural gas 1970 41 2011 245
Kwinana B Steam turbine Natural gas 1974 34 2008 0
Kwinana C Steam turbine Natural gas 1976 39 2015 385
Kwinana GT OCGT Distillate 1975 40 2015 21
Kwinana HEGT OCGT Natural gas 2011 30 2041 201
Muja A&B Subcritical pf Black coal 1968 40 2008 240
A21
Technical
Technical Capacity
Year
Muja C Subcritical pf Black coal 1981 40 2021 398

Muja D Subcritical pf Black coal 1986 40 2026 454
Mumbida Wind turbine Wind 2012 25 2037 55
Mungarra OCGT Natural gas 1991 30 2021 113
Namarkkon OCGT Distillate 2012 30 2042 70
Neerabup Peaker OCGT Natural gas 2009 30 2039 330
Newgen Power CCGT Natural gas 2007 30 2037 314
Parkeston SCE OCGT Natural gas 1996 30 2026 68
Pinjar A B OCGT Natural gas 1990 30 2020 228
Pinjar C OCGT Natural gas 1992 30 2022 233
Pinjar D OCGT Natural gas 1996 30 2026 124
Pinjarra Alinta Cogen Cogeneration Natural gas 2007 30 2037 280
Tesla (various sites) OCGT Distillate 2012 30 2042 40
Tiwest Cogen Cogeneration Natural gas 1999 30 2029 37
Wagerup Alinta Peaker OCGT Distillate 2007 30 2037 323
Walkaway Wind turbine Wind 2005 25 2030 89
Western Energy Peaker OCGT Natural gas 2011 30 2041 106
Worsley Cogeneration Black coal 1990 40 2030 0
Worsley SWCJV Cogeneration Natural gas 2000 25 2025 116
Burrup Peninsula OCGT Natural gas 2006 30 2036 74
Cape Lambert CCGT Natural gas 1996 30 2026 105
Cape Preston CCGT Natural gas 2009 30 2039 450
Dampier Steam Turbine Natural gas 2000 50 2050 120
Dampier C Steam Turbine Natural gas 1970 50 2020 120
NWIS
Karratha Steam turbine Natural gas 2005 50 2055 44
Karratha ATCO OCGT Natural gas 2010 30 2040 86
Paraburdoo Reciprocating Engine Liquid fuel 1985 30 2015 20
Port Hedland OCGT Natural gas 1997 30 2027 180
South Hedland CCGT Natural gas 2017 30 2047 150
Berrimah OCGT Liquid fuel 1979 30 2009 30
Channel Island u1-3 OCGT Natural gas 1986 30 2016 95
Channel Island u4-6 CCGT Natural gas 1998 30 2028 95
Channel Island u7 OCGT Natural gas 2006 30 2036 42
DKIS Channel Island u8-9 OCGT Natural gas 2012 30 2042 90
Katherine OCGT Natural gas 1987 30 2017 34
LMS Shoal Bay Reciprocating engine Landfill gas 2005 15 2020 1
Pine Creek CCGT CCGT Natural gas 1989 30 2019 27
Weddell OCGT Natural gas 2008 30 2038 128
APA Xstrata OCGT OCGT Natural gas 2008 30 2038 30
Diamantina CCGT CCGT Natural gas 2014 30 2044 242
Diamantina OCGT OCGT Natural gas 2014 30 2044 60
Ernest Henry Reciprocating Engine Liquid fuel 1997 30 2027 32
Mica Creek A CCGT CCGT Natural gas 2000 30 2030 103
Mt Isa
Mica Creek A GT OCGT Natural gas 2000 30 2030 132
Mica Creek B OCGT Natural gas 2000 30 2030 35
Mica Creek C CCGT Natural gas 2000 30 2030 55
Mt Isa Mines Station Steam turbine Natural gas 1974 50 2024 38
Phosphate Hill OCGT Natural gas 1999 30 2029 42
Source: ACIL Allen
Table A3 provides the assumed thermal efficiencies, auxiliary use, emissions factors, O&M
costs, outage rates and marginal loss factor (MLF) values for each existing and committed
generator. These values are taken from ACIL Allen’s generator database.
A22
Table A3 Existing and committed generators: efficiency, emissions and O&M costs
Scope 1
Scope 1
Thermal emission
emission FOM VOM Forced Planned
efficiency Auxiliaries intensity
Region Generator factor (kg ($/MW ($/MWh outage rate outage rate MLF
HHV (%) (%) (tonnes
CO2-e/GJ of gross/year) sent-out) (%) (%)
sent-out CO2-e/MWh
fuel)
sent-out)
AGL SF PV Broken Hill 0.00% 0 0 34,833 0 0.00% 0.00% 1.1026

AGL SF PV Nyngan 0.00% 0 0 34,833 0 0.00% 0.00% 1.1026
Bayswater 35.90% 6.00% 90.2 0.905 46,039 1.11 3.00% 4.00% 0.9653
Bendeela Pumps 0.00% 0 0 48,858 8.67 0.00% 0.00% 0.9877
Blowering 0.00% 0 0 48,858 4.82 0.00% 4.00% 0.9368
Boco Rock WF 0.00% 0 0 32,083 0 0.00% 0.00% 1.0156
Colongra 32.00% 3.00% 51.3 0.577 12,214 9.38 1.50% 0.00% 0.9855
Eraring 35.40% 6.00% 89.5 0.91 46,039 1.11 3.00% 4.00% 0.9850
Gullen Range WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9667
Gunning WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9609
Guthega 0.00% 0 0 48,858 6.74 0.00% 4.00% 0.8987
Hume NSW 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9232
Hunter Valley GT 28.00% 3.00% 69.7 0.896 12,214 8.93 2.50% 0.00% 0.9755
NSW Liddell 33.80% 5.00% 92.8 0.988 48,858 1.11 3.00% 8.00% 0.9663
Mt Piper 37.00% 5.00% 87.4 0.85 46,039 1.23 3.00% 4.00% 0.9698
Redbank 29.30% 8.00% 90 1.106 46,509 1.11 4.00% 4.00% 0.9744
Shoalhaven Bendeela 0.00% 0 0 48,858 8.67 0.00% 4.00% 0.9737
Smithfield 41.00% 5.00% 51.3 0.45 23,489 2.23 2.50% 2.00% 1.0026
Tallawarra 50.00% 3.00% 51.3 0.369 30,249 1.1 3.00% 2.00% 0.9839
Taralga WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9860
Tumut 1 0.00% 0 0 48,858 6.74 0.00% 4.00% 0.9363
Tumut 3 0.00% 0 0 48,858 10.6 0.00% 4.00% 0.9309
Tumut 3 Pumps 0.00% 0 0 48,858 0 0.00% 0.00% 0.9490
Uranquinty 32.00% 1.00% 51.3 0.577 12,214 9.38 1.50% 0.00% 0.8675
Vales Point B 35.40% 5.00% 89.8 0.913 46,039 1.11 3.00% 8.00% 0.9865
Wallerawang C 33.10% 7.00% 87.4 0.951 48,858 1.23 3.00% 8.00% 0.9729
Woodlawn WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9618
Barcaldine 40.00% 1.00% 51.3 0.462 23,489 2.23 2.50% 4.00% 0.9871
Barron Gorge 0.00% 0 0 48,858 10.6 0.00% 4.00% 1.0883
Braemar 1 30.00% 2.50% 51.3 0.616 12,214 7.33 1.50% 0.00% 0.9480
Braemar 2 30.00% 1.00% 51.3 0.616 12,214 7.33 1.50% 0.00% 0.9480
Callide B 36.10% 9.00% 93 0.927 46,509 1.12 4.00% 4.00% 0.9525
Callide C 36.50% 4.80% 95 0.937 46,509 2.54 6.00% 5.00% 0.9525
Collinsville 27.70% 10.00% 89.4 1.162 61,072 1.23 4.00% 2.00% 1.0389
Condamine 48.00% 3.00% 51.3 0.385 30,249 1.1 1.50% 4.00% 0.9519
Darling Downs 46.00% 6.00% 51.3 0.401 30,249 1.1 3.00% 4.00% 0.9480
Gladstone 35.20% 8.00% 92.1 0.942 48,858 1.11 4.00% 4.00% 0.9797
Kareeya 0.00% 0 0 48,858 5.78 0.00% 4.00% 1.0870
Kogan Creek 37.50% 9.00% 94 0.902 45,099 1.17 4.00% 4.00% 0.9493
QLD Mackay GT 28.00% 3.00% 69.7 0.896 12,214 8.4 1.50% 0.00% 1.0305
Millmerran 36.90% 6.00% 92 0.898 45,099 2.64 5.00% 8.00% 0.9532
Mt Stuart 30.00% 1.00% 69.7 0.836 12,214 8.4 2.50% 2.00% 1.0041
Oakey 32.60% 1.00% 51.3 0.567 12,214 8.93 2.00% 0.00% 0.9519
Roma 30.00% 1.00% 51.3 0.616 12,214 8.93 3.00% 0.00% 0.9404
Stanwell 36.40% 9.00% 90.4 0.894 46,039 2.99 2.50% 4.00% 0.9693
Swanbank E 47.00% 3.00% 51.3 0.393 30,249 1.1 3.00% 2.00% 0.9934
Tarong 36.20% 8.00% 92.1 0.916 46,509 6.98 3.00% 4.00% 0.9677
Tarong North 39.20% 6.00% 92.1 0.846 45,099 1.33 3.00% 4.00% 0.9678
Townsville 46.00% 3.00% 51.3 0.401 30,249 1.1 3.00% 2.00% 1.0292
Wivenhoe 0.00% 0 0 48,858 0 0.00% 4.00% 0.9891
Wivenhoe Pump 0.00% 0 0 28,187 0 0.00% 0.00% 0.9930
Yarwun 34.00% 2.00% 51.3 0.543 23,489 0 3.00% 0.00% 0.9837
Bluff WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9795
SA Clements Gap WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9658
Dry Creek 26.00% 1.00% 51.3 0.71 12,214 8.93 3.00% 0.00% 1.0022
A23
Scope 1
Scope 1
Thermal emission
HHV (%) (%) (tonnes
sent-out CO2-e/MWh
fuel)
sent-out)
Hallett 24.00% 1.00% 51.3 0.77 12,214 8.93 1.50% 0.00% 0.9869
Hallett 2 WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9843
Hallett WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9869
Ladbroke Grove 30.00% 1.00% 51.3 0.616 12,214 3.34 3.00% 4.00% 0.9884
Lake Bonney 2 WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9665
Lake Bonney 3 WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9665
Mintaro 28.00% 1.00% 51.3 0.66 12,214 8.93 1.50% 0.00% 0.9879
North Brown Hill WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9795
Northern 34.90% 10.00% 91 0.939 51,676 1.11 5.00% 8.00% 0.9744
Osborne 42.00% 5.00% 51.3 0.44 23,489 4.72 3.00% 2.00% 0.9990
Pelican Point 48.00% 2.00% 51.3 0.385 30,249 1.1 3.00% 4.00% 0.9994
Playford B 21.90% 8.00% 91 1.496 65,770 2.79 10.00% 8.00% 0.9882
Port Lincoln 26.00% 8.00% 67.9 0.94 12,214 8.93 1.50% 0.00% 0.9108
Quarantine 32.00% 5.00% 51.3 0.577 12,214 8.93 2.50% 0.00% 0.9939
Snowtown 2 North WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9861
Snowtown 2 South WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9861
Snowtown WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9136
Snuggery 26.00% 3.00% 67.9 0.94 12,214 8.93 2.00% 0.00% 0.9944
Torrens Island A 27.60% 5.00% 51.3 0.669 36,666 2.05 4.50% 4.00% 1.0004
Torrens Island B 30.00% 5.00% 51.3 0.616 36,666 2.05 4.50% 4.00% 1.0004
Waterloo WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9819
Bastyan 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9396
Bell Bay 29.00% 2.50% 51.3 0.637 36,666 2.05 12.00% 8.00% 0.9999
Bell Bay Three 29.00% 1.00% 51.3 0.637 12,214 7.33 3.00% 0.00% 0.9999
Cethana 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9593
Devils Gate 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9639
Fisher 0.00% 0 0 48,858 4.82 0.00% 4.00% 0.9645
Gordon 0.00% 0 0 48,858 4.82 0.00% 4.00% 0.9046
John Butters 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9405
Lake Echo 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9028
Lemonthyme_Wilmot 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9683
Liapootah_Wayatinah_Catagunya 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9702
TAS
Mackintosh 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9296
Meadowbank 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9644
Musselroe WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.8957
Poatina 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9681
Reece 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9317
Tamar Valley 48.00% 3.00% 51.3 0.385 30,249 1.1 3.00% 2.00% 0.9990
Tamar Valley GT 28.00% 2.00% 51.3 0.66 12,214 8.93 3.00% 2.00% 0.9999
Tarraleah 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9195
Trevallyn 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9745
Tribute 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9339
Tungatinah 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9122
Anglesea 27.20% 10.00% 91 1.204 124,962 1.11 3.00% 2.00% 0.9849
Bairnsdale 34.00% 1.00% 51.3 0.543 12,214 2.09 2.50% 0.00% 0.9745
Dartmouth 0.00% 0 0 48,858 5.78 0.00% 4.00% 1.0097
Eildon 0.00% 0 0 48,858 8.67 0.00% 4.00% 1.0078
Energy Brix 24.00% 15.00% 99 1.485 93,957 2.05 2.50% 4.00% 0.9682
Hazelwood 22.00% 10.00% 93 1.522 131,539 1.11 3.50% 8.00% 0.9691
VIC Hume VIC 0.00% 0 0 48,858 5.78 0.00% 4.00% 0.9912
Jeeralang A 22.90% 3.00% 51.3 0.806 12,214 8.4 2.50% 0.00% 0.9647
Jeeralang B 22.90% 3.00% 51.3 0.806 12,214 8.4 2.50% 0.00% 0.9647
Laverton North 30.40% 1.00% 51.3 0.608 12,214 7.33 1.50% 2.00% 0.9972
Loy Yang A 27.20% 9.00% 91.5 1.211 122,144 1.11 3.00% 2.00% 0.9704
Loy Yang B 26.60% 8.00% 91.5 1.238 87,738 1.11 4.00% 2.00% 0.9704
Macarthur WF 0.00% 0 0 32,083 0 0.00% 0.00% 0.9946
A24
Scope 1
Scope 1
Thermal emission
HHV (%) (%) (tonnes
sent-out CO2-e/MWh
fuel)
sent-out)
McKay 0.00% 0 0 48,858 6.74 0.00% 4.00% 1.0083

Mortlake 32.00% 1.00% 51.3 0.577 12,214 7.73 2.50% 0.00% 0.9922
Mt Mercer WF 0.00% 0 0 32,083 0 0.00% 0.00% 1.0064
Murray 0.00% 0 0 48,858 5.78 0.00% 4.00% 1.0178
Newport 33.30% 5.00% 51.3 0.555 37,583 2.09 2.00% 4.00% 0.9953
Oaklands Hill WF 0.00% 0 0 32,083 0 0.00% 0.00% 1.0262
Somerton 24.00% 1.00% 51.3 0.77 12,214 8.93 1.50% 0.00% 0.9927
Valley Power 24.00% 1.00% 51.3 0.77 12,214 8.93 1.50% 0.00% 0.9704
West Kiewa 0.00% 0 0 48,858 6.74 0.00% 4.00% 1.0416
Yallourn 23.50% 10.00% 92.5 1.417 126,842 1.11 4.00% 4.00% 0.9538
Albany 0.00% 0 0 42,000 1.05 0.00% 0.00% 1.0699
Alcoa Kwinana Cogen 30.00% 1.00% 51.3 0.616 25,000 0 3.80% 5.20% 1.0214
Alcoa Pinjarra Cogen 30.00% 1.00% 51.3 0.616 25,000 0 3.80% 5.20% 0.9951
Alcoa Wagerup Cogen 30.00% 1.00% 51.3 0.616 25,000 0 3.80% 5.20% 0.9868
Bluewaters 36.10% 7.50% 93.1 0.928 52,000 1.58 3.00% 4.90% 0.9992
BP Cogen 33.00% 2.00% 51.3 0.56 23,489 0 5.00% 4.10% 1.0246
Canning/Melville LFG 30.00% 0.00% 0 0 50,000 3.68 5.00% 0.00% 1.0296
Cockburn 48.00% 2.40% 51.3 0.385 30,249 4.73 4.20% 10.10% 1.0277
Collgar WF 0.00% 0 0 42,000 1.05 0.00% 0.00% 1.0146
Collie 36.00% 7.90% 93.1 0.931 52,000 1.58 3.20% 8.50% 0.9956
Emu downs 0.00% 0 0 42,000 1.05 0.00% 0.00% 0.9937
Geraldton 29.00% 0.50% 67.9 0.843 12,214 9.46 5.90% 9.00% 1.0507
Grasmere 0.00% 0 0 42,000 1.05 0.00% 0.00% 1.0699
Greenough River 0.10% 0 0 50,000 0 0.00% 0.00% 1.0227
Kalgoorlie 33.00% 0.50% 67.9 0.741 12,214 9.46 5.90% 4.10% 1.0782
Kalgoorlie Nickel 33.00% 0.50% 51.3 0.56 12,214 9.46 5.20% 4.70% 1.2253
Kemerton 34.00% 0.50% 51.3 0.543 12,214 9.46 6.00% 7.90% 1.0079
Kwinana A 32.00% 9.00% 51.3 0.577 40,000 8.41 5.40% 14.80% 1.0201
Kwinana B 32.00% 9.00% 51.3 0.577 40,000 8.41 5.40% 14.80% 1.0201
Kwinana C 33.00% 4.00% 51.3 0.56 40,000 7.35 5.20% 9.90% 1.0201
Kwinana GT 32.00% 0.50% 67.9 0.764 12,214 9.46 5.20% 9.90% 1.0201
SWIS
Kwinana HEGT 40.00% 0.50% 51.3 0.462 12,214 1.31 5.20% 4.10% 1.0201
Muja A&B 26.40% 8.50% 93.1 1.27 60,000 1.58 4.20% 10.00% 1.0000
Muja C 34.60% 8.00% 93.1 0.97 52,000 1.58 4.20% 9.90% 1.0000
Muja D 35.60% 8.00% 93.1 0.942 52,000 1.58 4.90% 9.90% 1.0000
Mumbida 0.00% 0 0 42,000 1.05 0.00% 0.00% 1.0353
Mungarra 29.00% 0.50% 51.3 0.637 12,214 9.46 5.20% 9.90% 1.0353
Namarkkon 30.00% 1.00% 67.9 0.815 12,214 9.46 4.00% 4.00% 1.0405
Neerabup Peaker 32.00% 2.00% 51.3 0.577 12,214 9.46 3.90% 2.20% 1.0379
Newgen Power 48.00% 2.00% 51.3 0.385 30,249 1.1 4.00% 3.30% 1.0224
Parkeston SCE 33.00% 0.50% 51.3 0.56 12,214 9.46 5.20% 4.90% 1.2012
Pinjar A B 29.00% 0.50% 51.3 0.637 12,214 9.46 5.20% 9.90% 1.0312
Pinjar C 29.00% 0.50% 51.3 0.637 12,214 9.46 5.20% 9.90% 1.0312
Pinjar D 29.00% 0.50% 51.3 0.637 12,214 9.46 5.20% 9.90% 1.0312
Pinjarra Alinta Cogen 34.10% 2.40% 51.3 0.542 25,000 0 3.90% 4.10% 1.0184
Tesla (various sites) 28.00% 1.00% 67.9 0.873 12,214 9.46 4.00% 4.00% 1.1229
Tiwest Cogen 32.00% 1.50% 51.3 0.577 25,000 0 5.90% 4.10% 1.0177
Wagerup Alinta Peaker 34.10% 0.50% 67.9 0.717 12,214 9.46 3.90% 4.10% 0.9868
Walkaway 0.00% 0 0 42,000 1.05 0.00% 0.00% 0.9560
Western Energy Peaker 32.00% 0.50% 51.3 0.577 12,214 9.46 5.20% 4.10% 1.0204
Worsley 28.00% 0.00% 93.1 1.197 25,000 0 4.80% 4.10% 0.9886
Worsley SWCJV 33.00% 2.00% 51.3 0.56 25,000 0 5.00% 4.10% 0.9846
Burrup Peninsula 29.00% 2.00% 51.3 0.637 12,214 9.61 3.00% 8.00% 1.0000
Cape Lambert 45.00% 2.00% 51.3 0.41 30,249 1.1 3.00% 8.00% 1.0000
NWIS
Cape Preston 50.00% 3.00% 51.3 0.369 30,249 1.1 3.00% 8.00% 1.0000
Dampier 30.00% 5.00% 51.3 0.616 40,000 2.25 3.00% 4.00% 1.0000
A25
Scope 1
Scope 1
Thermal emission
HHV (%) (%) (tonnes
sent-out CO2-e/MWh
fuel)
sent-out)
Dampier C 30.00% 5.00% 51.3 0.616 40,000 2.25 3.00% 4.00% 1.0000
Karratha 30.00% 5.00% 51.3 0.616 40,000 2.25 3.00% 4.00% 1.0000
Karratha ATCO 40.00% 2.00% 51.3 0.462 12,214 9.61 3.00% 8.00% 1.0000
Paraburdoo 29.00% 2.00% 67.9 0.843 13,000 9.61 3.00% 4.00% 1.0000
Port Hedland 29.00% 2.00% 51.3 0.637 12,214 9.61 3.00% 8.00% 1.0000
South Hedland 46.00% 2.00% 51.3 0.393 40,000 2.25 3.00% 4.00% 1.0000
Berrimah 24.00% 1.00% 67.9 1.019 12,214 9.61 3.00% 8.00% 1.0000
Channel Island u1-3 27.00% 1.00% 51.3 0.684 12,214 9.61 3.00% 8.00% 1.0000
Channel Island u4-6 48.00% 2.00% 51.3 0.385 30,249 1.1 3.00% 8.00% 1.0000
Channel Island u7 37.00% 1.00% 51.3 0.499 12,214 9.61 3.00% 8.00% 1.0000
DKIS Channel Island u8-9 37.00% 1.00% 51.3 0.499 12,214 9.61 3.00% 8.00% 1.0000
Katherine 25.00% 1.00% 51.3 0.739 12,214 9.61 3.00% 8.00% 1.0000
LMS Shoal Bay 40.00% 2.00% 0 0 80,000 4 3.00% 5.00% 1.0000
Pine Creek CCGT 47.00% 2.00% 51.3 0.393 30,249 1.1 3.00% 8.00% 1.0000
Weddell 35.00% 1.00% 51.3 0.528 12,214 9.61 3.00% 4.00% 1.0000
APA Xstrata OCGT 36.00% 1.00% 51.3 0.513 12,214 9.61 3.00% 8.00% 1.0000
Diamantina CCGT 48.00% 2.00% 51.3 0.385 30,249 1.05 3.00% 4.00% 1.0000
Diamantina OCGT 32.00% 2.00% 51.3 0.577 12,214 9.61 3.00% 5.00% 1.0000
Ernest Henry 29.00% 2.00% 67.9 0.843 13,000 9.61 3.00% 4.00% 1.0000
Mica Creek A CCGT 43.00% 2.00% 51.3 0.429 30,249 1.05 3.00% 8.00% 1.0000
Mt Isa
Mica Creek A GT 27.00% 3.00% 51.3 0.684 12,214 9.61 3.00% 8.00% 1.0000
Mica Creek B 27.00% 3.00% 51.3 0.684 12,214 9.61 3.00% 8.00% 1.0000
Mica Creek C 43.00% 2.00% 51.3 0.429 30,249 9.61 3.00% 8.00% 1.0000
Mt Isa Mines Station 25.00% 1.00% 51.3 0.739 40,000 9.61 3.00% 8.00% 1.0000
Phosphate Hill 27.00% 3.00% 51.3 0.684 12,214 1.05 3.00% 8.00% 1.0000
Source: ACIL Allen
A.3.1 Energy constrained and intermittent generation
Hydro
Within PowerMark LT the annual output of hydro stations can be constrained explicitly to
desired levels.49 Aside from run of river output which occurs independently of wholesale
prices, the model will naturally schedule hydro output during high priced periods in order to
minimise system production costs.
It should be recognised that hydro output does fluctuate considerably year to year and is
also susceptible to drought and flood events as witnessed over the last decade. Whilst the
modelling can account for changes to long-term averages, it is not typically used to predict
fluctuations due to cyclical changes in weather conditions.
Output from the Snowy Mountains Hydro-electric Scheme (Snowy Hydro) has averaged
around 4,000 GWh over the last 10 years. ACIL Allen assumes that over the long-term
output averages 4,700 GWh with a 60/40 split between NSW and Victorian regions, which is
slightly higher than the recent average reflecting prevailing drought conditions for much of
the past decade. Similarly, Tasmanian hydro output has averaged approximately
8,000 GWh over the same period. The modelling assumes 9,100 GWh of output which
corresponds to Hydro Tasmania’s long-term assumption.
49 Simulation models typically use the notion of an opportunity cost for the water which attempts to maximise the net revenue of
the plant but not break the energy constraint.
A26
Wind
For wind farms, annual output is limited to capacity factors which approximate recent actual
outcomes (if available) or assumed levels based on corresponding nearby operating
facilities. Wind output is profiled according to 30 minute resolution wind traces for a range of
wind regimes across Australia. These wind traces are then mapped back to the sampled
demand profiles in order to ensure wind output correlates properly with demand.
Solar
Solar plants are also limited by annual capacity factor constraints according to the
technologies capability. The only committed large-scale scheduled solar systems within the
modelling are AGL Energy’s 159 MW solar flagship developments in NSW and the 10 MW
Greenough River project in the SWIS.50
ACIL Allen incorporates representative solar PV output profiles for these projects which vary
by time of day and month.
A.3.2 Mothballing, end of life and refurbishment
Mothballing
Where power station profitability is negatively affected by market events (such as the
introduction of large amount of additional renewable capacity or a temporary decline in
demand) it is natural for supply to respond. For generation portfolios, a typical response is to
withdraw some capacity from the market through mothballing. Incumbent generators can
withdraw from the market in a variety of ways including initially through partial mothballing
and eventually full closure and retirement. There are a number of drivers for this supply-side
response by existing generators.
In the old government-owned vertically integrated industry, mothballing and retirement
decisions were taken based on engineering concepts and older capacity was retired to
make way for more efficient new technologies in a cost minimisation planning framework. In
the context of market-based structures, decision making regarding capacity is undertaken by
individual participants in a decentralised manner. The focus of participants is solely on
commercial profitability of their own operations which is likely at times to result in materially
different outcomes to that of a centralised generation planner.
As a general rule, capacity will be withdrawn from the market if the cash returns from
operation (revenues less variable operation and maintenance costs) do not cover the fixed
operating and maintenance costs of the station.
As wholesale electricity prices are somewhat volatile and prices can at times be
suppressed, this negative profitability trigger will only occur if it is apparent that market
conditions are unlikely to improve in the near-to-medium term. It is at this point that a
generator may mothball capacity. In doing so, the aggregate supply is reduced and spot
prices will rise, making any remaining capacity more profitable (all else being equal). The
rise in prices also has a positive effect upon the balance of a generator’s portfolio and
competing generators.
For a power station with multiple units, mothballing a portion of capacity may improve the
profitability of the remaining station. However the profitability of the remaining units may
50 Other smaller existing solar developments are treated as non-scheduled or embedded generation and are therefore handled
outside of the PowerMark LT modelling.
A27
need to cover virtually the same fixed O&M burden as some fixed elements would not vary
with the number of units running (for example, mine costs or wages and salaries).
A decision to mothball capacity is taken where there is a reasonable prospect of market
conditions improving in the medium term which would allow the unit to return to service.
A retirement decision is taken where there is little prospect of market conditions improving
such that the generator would earn net revenues above fixed O&M costs. When this occurs,
its economic value is zero and even a sale process would likely yield a zero or negative
result. Permanent retirement of a power station is likely to trigger the obligations of site
rehabilitation depending on the conditions specified with the generating licence with the
State Government. These decisions are therefore not taken lightly as site rehabilitation can
run into many millions of dollars and this cost is difficult to estimate prior to closure. 51
In recent times there has been a large amount of coal-fired capacity mothballed – primarily
due to the declining demand conditions, but also due to the impost of carbon pricing on the
less profitable generators. Table A4 summarises the plant that have had capacity withdrawn
from the market in recent years.
Table A4 Recently mothballed or retired units by NEM region

Capacity
Portfolio Power Station Fuel type Public comments/assumptions on period affected
(MW)
Queensland
Mothballed Oct 2012, back online late 2014, early
Stanwell Tarong (2 units) Coal 700 2015. I unit to return in Oct 2014 to replace
Swanbank E
Currently operating but due to be mothballed Oct
Stanwell Swanbank E Gas 385
2014 due to fuel source being on-sold
Permanently retired, units decommissioned in April
Stanwell Swanbank B Coal 480 2010, June 2010, 2011, and May 2012, due to the
plant reaching the end of its operational life
Mothballed December 2012, not expected to return to
RATCH Collinsville Coal 190
service
New South Wales
Delta Electricity Munmorah Coal 600 Retired July 2012
Mothballed whole station March 2014 until viable for
EnergyAustralia Wallerawang Coal 1000
return to service. One unit permanently retired
Victoria
EnergyBrix Morwell (2 units) Coal 95 From July 2012 until viable
EnergyAustralia Yallourn (1 unit) Coal 350 From mid-2012 until viable
South Australia
Alinta Energy Northern Coal 540 Mothballed Apr-Sept until 2015
From March 2012 until viable; not expected to return
Alinta Energy Playford Coal 200
to service
Refurbishment
All generating plant have a technical design life for which an allowance of ‘stay-in-business’
capital expenditure is provided through annual fixed operating and maintenance costs. The
fixed operating and maintenance cost assumptions however do not provide for abnormal
capital expenditure required for life extension.
51 It has been suggested that some generators who have mothballed capacity are effectively deferring costs associated with
rehabilitation.
A28
Design lives range from 20-30 years for wind and solar, 30 years for gas and 40+ years for
coal. However, as has often been the experience in Australia, most generating plant have
had operational lives extended through refurbishment programmes. Refurbishment requires
a large lump of capital expenditure to refresh/upgrade various components of the power
station. The decision on whether to proceed with a refurbishment is an economic one and is
dependent upon the commercial outlook (present value of expected net revenues against
upfront capital expenditure).
The capital costs for refurbishment will vary greatly across technologies and, often, be site
specific. Therefore some simplifying generic assumptions are required.
Table A5 provides the proposed refurbishment capital costs for plant which reach the end of
their stated technical life. Capital expenditure for the refurbishment is expressed as a
percentage of new entry costs for the same technology and results in the plant being
operational beyond its technical retirement date for a set number of years. The modelling
allows for more than one refurbishment so for example, a subcritical coal plant would incur a
refurbishment cost of 25% of a new coal plant every 15 years after the end of its technical
retirement date.
PowerMark LT evaluates the economics of refurbishment against expected net revenues
able to be earned from continued operation and projects whether the refurbishment would
go ahead or the station be retired. If refurbishment occurs, output from the power station is
reduced in the refurbishment year, reflecting the time units are out of service.
Table A5 Refurbishment costs for incumbent plant

Economic life of Additional life Additional life
Refurbishment
Technology new plant (% of original from refurb
cost (% of new)
(years) life) (years)
CCGT 30 70% 100% 30
Cogeneration 30 70% 100% 30
OCGT 30 85% 100% 30
Solar PV 35 75% 100% 35
Steam turbine 50 25% 30% 15
Subcritical pf 50 25% 30% 15
Supercritical pf 50 25% 30% 15
Wind turbine 25 30% 100% 25
Source: ACIL Allen
Retirement criteria
Existing plant may cease operating if net operating revenues from the market (revenue less
variable O&M) fail to cover their avoidable fixed overheads.52 This profitability metric is
assessed on a standardised per kW basis for each generator. Once this metric turns
negative on a sustained basis (i.e. over several years and there is no prospect of recovery),
the station is retired regardless of its remaining technical asset life. Retirement may be
sculpted over a number of years to avoid large single year shocks to the market and reflects
gradual unit retirement.
52 For integrated mine mouth brown coal power stations, fixed overheads also include mine overheads as in most cases the
closure of the power station would also result in closure of the mine.
A29
A.4 New entrant generators

A range of new entrant generating technologies are made available within the modelling
over the period to 2040. PowerMark LT determines a least cost plant mix for each modelled
region on a dynamic inter-temporal basis.
New capacity is introduced to each region through the use of continuous capacity variables,
that is, generation increments are not set to predetermined sizes and the model allows entry
of any optimal increment.53
A range of cost and generation characteristics are required for each new entrant technology
to solve the model in a way that minimises overall resource costs on a net present value
basis. The key proposed inputs for each of these elements is discussed in the following
sections.
Costs are presented in this section in real 2011-12 dollars to enable comparisons against
the AETA 2012 work. Capital costs will be escalated to today’s dollars prior to modelling
using published CPI.
A.4.1 Capital costs

Capital costs comprise one of the key inputs for long-term electricity sector modelling as
capital is the largest cost component for most generation technologies.
The methodology employed for this study is to commence with a starting capital cost value
(termed the ‘base’ capital cost) and break this down into its component parts: local labour;
local equipment and commodities; and foreign equipment and commodities.
These component parts are then projected forward individually before being recombined into
a final capital cost estimate. This process allows for the influences of learning rates (both
foreign and local), labour costs, and exchange rates to be properly incorporated into the
final cost estimates.
For the most part, the base capital cost estimates for most technologies were taken from the
2012 Australian Energy Technology Assessment (AETA) published by the Bureau of
Resource and Energy Economics (BREE), with adjustments made in the 2013 update to the
study.54
ACIL Allen has selected a sub-set of 29 of the 40 technologies examined within the AETA
study. Technologies excluded include exotic coal-based technologies that do not employ
carbon capture and storage (IGCC, oxy-fuel and direct injection), solar hybrids, offshore
wind, landfill gas, bagasse and nuclear options.
Table A6 presents the proposed capital costs for each of the technologies considered within
the modelling. The table also includes the headline splits for the cost components taken
from the AETA study.
These capital costs are presented on an ‘overnight’ basis – interest during construction
(IDC) and financing costs are excluded.55 For plants that employ carbon capture, the capital
costs include capture and compression of CO 2, but exclude transport and storage costs.
53 The PowerMark LT model is formulated as a linear program. A mixed integer linear program (MILP) formulation is required to
introduce standard increments of new entrant capacity however this increases solution time enormously.
54 Note that the 2013 update only updated operating and maintenance costs for a select few technologies and no update to
capital costs.
55 Interest during construction is added within the modelling.
A30
ACIL Allen made some minor modifications to base capital costs for a number of selected
technologies where it has direct recent experience with actual proposed projects in
Australia. Figure A17 shows a comparison of the proposed capital cost figures against those
within the AETA study.
Modifications to the base capital costs were made for the following technologies:
 Natural gas-fired CCGT (7% higher)
 Natural gas-fired OCGT (12% higher)
 Solar PV (20% lower) including corresponding changes to tracking options
 Onshore wind (9% lower).
Figure A17 Base capital cost comparison with AETA
12,000
Adopted for study AETA 2012
Real 2011-12 A$/kW Installed
10,000
8,000
6,000
4,000
2,000
Source: ACIL Allen, BREE (AETA 2012)
Hydro-electric generation is not included as a model as a new entrant technology. This

reflects the fact that few commercially viable large-scale hydro-electric sites remain in
Australia for exploitation.
A31
Table A6 Base capital costs and cost component splits
2011-12 Base capital 2011-12 Base capital Foreign equipment and Local equipment and
Category Technology Labour
cost (A$/kW installed) cost (A$/kW net) commodities commodities
Coal PC Supercritical – Brown Coal 3,451 3,788 29% 38% 33%

PC Supercritical Black Coal 2,974 3,124 30% 39% 31%
PC Supercritical Black Coal (SWIS Scale) 3,192 3,381 31% 40% 29%
Natural gas CCGT 1,100 1,127 26% 56% 18%
CCGT SWIS Scale 1,078 1,111 26% 56% 18%
OCGT 800 808 11% 79% 10%
Solar CLFR 4,802 5,220 20% 55% 25%
CLFR with storage 8,550 9,500 25% 55% 20%
Parabolic trough 4,526 4,920 20% 55% 25%
Parabolic trough with storage 8,055 8,950 25% 55% 20%
Central Receiver 5,570 5,900 30% 55% 15%
Central Receiver with storage 7,477 8,308 25% 55% 20%
Solar PV Solar PV fixed 2,700 2,700 15% 70% 15%
Solar PV single axis tracking 3,180 3,180 15% 70% 15%
Solar PV dual axis tracking 4,730 4,730 15% 70% 15%
Wind On-shore Wind Farm 2,300 2,312 15% 72% 13%
Ocean/Wave 5,900 5,900 30% 40% 30%
Biomass Other Biomass Waste 4,400 5,000 18% 27% 55%
Geothermal Geothermal Hot Sedimentary Aquifer (HSA) 6,300 7,000 34% 23% 43%
Geothermal Engineered Geothermal System (EGS) 9,646 10,600 37% 17% 46%
CCS PC Supercritical with CCS – Brown Coal 5,902 7,766 29% 35% 36%
PC Supercritical with CCS – Bituminous Coal 4,559 5,434 29% 35% 36%
PC Oxy Combustion Supercritical with CCS 4,274 5,776 33% 35% 32%
CCGT with CCS 2,495 2,772 19% 67% 14%
IGCC with CCS – Bituminous Coal 4,984 7,330 27% 52% 21%
IGCC with CCS – Brown Coal 5,083 8,616 27% 52% 21%
CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 2,493 3,945 30% 30% 40%
PC Subcritical Black Coal - Retrofit CCS 1,611 2,244 30% 30% 40%
Existing CCGT with retrofit CCS 1,392 1,547 12% 78% 10%
Note: CCS capital costs are inclusive of capture, but exclude transport and storage costs as these are treated separately.
Source: ACIL Allen, BREE (AETA 2012).

A.4.2 Learning rates
Learning rates are applied to the Base capital costs to reflect cost changes over time
through technology and manufacturing improvements and learning by doing.
Learning rates for each major technology have been taken from CSIRO’s Global and Local
Learning Model (GALLM) as part of the AETA 2012 study. For some technologies
differential learning rates were provided for foreign and local content components and these
have been applied to the respective foreign equipment and local equipment/local labour
components respectively. Learning rates in the GALLM model are endogenous and respond
to the rate of deployment of each technology both locally and internationally.
A complication in this process is the adjustments made by ACIL Allen to the base capital
costs for solar PV and wind technologies from the AETA figures. As these represent a
reduction in the starting base capital cost, it was decided that the learning rates should be
reduced in the early years such that the capital cost for 2020 remained unchanged from the
AETA work. The reported learning rates for these technologies in the period to 2020 will
therefore differ due to the lower starting value.
A.4.3 Other adjustments

Adjustments are made to the some of the capital cost components based on macro
assumptions as reported in section A.2.5. These include:
 local labour costs are modified through the application of the real labour cost index
 the composite metals price index is used to adjust 25% and 40% of the local and
foreign equipment cost component respectively.
 exchange rates are used to convert the foreign equipment and commodities cost
component (which are projected in US dollars) back into Australian dollars.
A.4.4 Final capital costs

Table A7 presents the final capital costs for each of the technologies after all adjustments
for learning, labour, metals and exchange rates are made. These are also shown graphically
in Figure A18. Table A8 shows the average year-on-year percentage change in capital costs
for each decade of the projection.
A-33
Table A7 Final capital costs for new entrant technologies for selected years (Real 2011-12 $/kW installed)
Technology 2011-12 2019-20 2029-30 2039-40 2049-50
Coal PC Supercritical – Brown Coal 3,451 3,595 3,623 3,651 3,731
PC Supercritical Black Coal 2,974 2,735 2,767 2,801 2,876
PC Supercritical Black Coal (SWIS Scale) 3,192 2,942 2,978 3,018 3,104
Natural gas CCGT 1,100 1,178 1,207 1,233 1,286
CCGT SWIS Scale 1,078 1,154 1,183 1,208 1,260
OCGT 800 842 844 834 830
Solar CLFR 4,802 2,612 2,164 2,195 2,262
CLFR with storage 8,550 4,683 3,892 3,970 4,124
Parabolic trough 4,526 2,462 2,040 2,069 2,132
Parabolic trough with storage 8,055 4,412 3,667 3,740 3,885
Central Receiver 5,570 3,073 2,560 2,626 2,749
Central Receiver with storage 7,477 4,096 3,404 3,471 3,606
Solar PV Solar PV fixed 2,700 1,813 1,473 943 903
Solar PV single axis tracking 3,180 2,135 1,735 1,111 1,063
Solar PV dual axis tracking 4,730 3,176 2,580 1,653 1,581
Wind On-shore Wind Farm 2,300 1,767 1,806 1,829 1,815
Ocean/Wave 5,900 6,312 3,072 3,168 3,142
Biomass Other Biomass Waste 4,400 4,576 4,637 4,498 4,646
Geothermal Geothermal HSA 6,300 6,684 6,901 7,194 7,624
Geothermal EGS 9,646 10,228 10,605 11,096 11,813
CCS PC Supercritical with CCS – Brown Coal 5,902 6,185 4,667 4,678 4,805
PC Supercritical with CCS – Bituminous Coal 4,559 4,778 3,605 3,613 3,712
PC Oxy Combustion Supercritical with CCS 4,274 4,505 3,409 3,435 3,551
CCGT with CCS 2,495 2,649 1,931 1,903 1,914
IGCC with CCS – Bituminous Coal 4,984 5,295 3,778 3,778 3,878
IGCC with CCS – Brown Coal 5,083 5,400 3,853 3,854 3,955
CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 2,493 2,604 1,997 2,004 2,061
PC Subcritical Black Coal - Retrofit CCS 1,611 1,683 1,291 1,295 1,332
Existing CCGT with retrofit CCS 1,392 1,478 1,070 1,046 1,040
Small CCGT Small Scale 1,886 2,019 2,070 2,114 2,205
Note: CCS capital costs are inclusive of capture, but exclude transport and storage costs which are treated separately.
Source: ACIL Allen
A-34
Figure A18 Final capital costs for new entrant technologies for selected years
Capital cost (Real 2011-12 $/kW installed)
0 2,000 4,000 6,000 8,000 10,000 12,000 14,000
PC Supercritical – Brown Coal
PC Supercritical Black Coal
PC Supercritical Black Coal (SWIS Scale)
CCGT
CCGT SWIS Scale
OCGT
CLFR
CLFR with storage
Parabolic trough
Parabolic trough with storage
Central Receiver
Central Receiver with storage
Solar PV fixed
Solar PV single axis tracking
Solar PV dual axis tracking
On-shore Wind Farm
Ocean/Wave
Other Biomass Waste
Geothermal HSA
Geothermal EGS
PC Supercritical with CCS – Brown Coal
PC Supercritical with CCS – Bituminous Coal
PC Oxy Combustion Supercritical with CCS 2011-12
CCGT with CCS

2019-20
IGCC with CCS – Bituminous Coal
IGCC with CCS – Brown Coal 2029-30
PC Subcritical Brown Coal - Retrofit CCS

2039-40
PC Subcritical Black Coal - Retrofit CCS
2049-50
Existing CCGT with retrofit CCS
CCGT Small Scale
Source: ACIL Allen
A-35
Table A8 Average real year-on-year capital cost change for each decade
2011-12 to 2019-20 to 2029-30 to 2039-40 to
Technology
2019-20 2029-30 2039-40 2049-50
Coal PC Supercritical – Brown Coal 0.5% 0.1% 0.1% 0.2%
PC Supercritical Black Coal -1.0% 0.1% 0.1% 0.3%
PC Supercritical Black Coal (SWIS Scale) -1.0% 0.1% 0.1% 0.3%
Natural gas CCGT 0.9% 0.2% 0.2% 0.4%
CCGT SWIS Scale 0.9% 0.2% 0.2% 0.4%
OCGT 0.6% 0.0% -0.1% -0.1%
Solar CLFR -7.3% -1.9% 0.1% 0.3%
CLFR with storage -7.2% -1.8% 0.2% 0.4%
Parabolic trough -7.3% -1.9% 0.1% 0.3%
Parabolic trough with storage -7.2% -1.8% 0.2% 0.4%
Central Receiver -7.2% -1.8% 0.3% 0.5%
Central Receiver with storage -7.2% -1.8% 0.2% 0.4%
Solar PV Solar PV fixed -4.9% -2.1% -4.4% -0.4%
Solar PV single axis tracking -4.9% -2.1% -4.4% -0.4%
Solar PV dual axis tracking -4.9% -2.1% -4.4% -0.4%
Wind On-shore Wind Farm -3.2% 0.2% 0.1% -0.1%
Ocean/Wave 0.8% -6.9% 0.3% -0.1%
Biomass Other Biomass Waste 0.5% 0.1% -0.3% 0.3%
Geothermal Geothermal HSA 0.7% 0.3% 0.4% 0.6%
Geothermal EGS 0.7% 0.4% 0.5% 0.6%
CCS PC Supercritical with CCS – Brown Coal 0.6% -2.8% 0.0% 0.3%
PC Supercritical with CCS – Bituminous Coal 0.6% -2.8% 0.0% 0.3%
PC Oxy Combustion Supercritical with CCS 0.7% -2.7% 0.1% 0.3%
CCGT with CCS 0.8% -3.1% -0.1% 0.1%
IGCC with CCS – Bituminous Coal 0.8% -3.3% 0.0% 0.3%
IGCC with CCS – Brown Coal 0.8% -3.3% 0.0% 0.3%
CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 0.5% -2.6% 0.0% 0.3%
PC Subcritical Black Coal - Retrofit CCS 0.5% -2.6% 0.0% 0.3%
Existing CCGT with retrofit CCS 0.7% -3.2% -0.2% -0.1%
Small CCGT Small Scale 0.9% 0.2% 0.2% 0.4%
Source: ACIL Allen
A.4.5 Other new entrant parameters

Table A9 provides other technical parameters and cost assumptions for the new entrant
technologies. For the most part these are aligned with the AETA 2012 and 2013 studies,
with a few modifications.
Table A9 New entrant parameters

Thermal FOM (Real VOM (Real
Auxiliary load
Category Technology efficiency (% 2011-12 2011-12
(%)
HHV sent-out) $/MW/year) $/MWh)
Coal PC Supercritical – Brown Coal 32.3% 8.9% 85,000 1.30
PC Supercritical Black Coal 41.9% 4.8% 52,000 1.30
PC Supercritical Black Coal (SWIS Scale) 41.4% 5.6% 55,500 8.00
Natural gas CCGT 49.5% 2.4% 33,000 1.20
CCGT SWIS Scale 49.3% 3.0% 10,000 4.00
OCGT 32.0% 1.0% 14,000 8.00
Solar CLFR 100.0% 8.0% 64,107 15.19
CLFR with storage 100.0% 10.0% 72,381 11.39
A-36
Thermal FOM (Real VOM (Real
Auxiliary load
Category Technology efficiency (% 2011-12 2011-12
(%)
HHV sent-out) $/MW/year) $/MWh)
Parabolic trough 100.0% 8.0% 59,176 15.19
Parabolic trough with storage 100.0% 10.0% 72,381 11.39
Central Receiver 100.0% 5.6% 58,285 7.07
Central Receiver with storage 100.0% 10.0% 71,372 5.65
Solar PV Solar PV fixed 100.0% 0.0% 25,000 0.00
Solar PV single axis tracking 100.0% 0.0% 30,000 0.00
Solar PV dual axis tracking 100.0% 0.0% 39,000 0.00
Wind On-shore Wind Farm 100.0% 0.5% 32,500 10.00
Ocean/Wave 100.0% 0.0% 190,000 0.00
Biomass Other Biomass Waste 27.0% 12.0% 125,000 8.00
Geothermal Geothermal HSA 100.0% 10.0% 200,000 0.00
Geothermal EGS 100.0% 9.0% 170,000 0.00
CCS PC Supercritical with CCS – Brown Coal 20.8% 24.0% 91,500 15.00
PC Supercritical with CCS – Bituminous Coal 31.4% 16.1% 73,200 12.00
PC Oxy Combustion Supercritical with CCS 32.5% 26.0% 62,000 14.00
CCGT with CCS 43.1% 10.0% 17,000 9.00
IGCC with CCS – Bituminous Coal 28.9% 32.0% 98,700 8.00
IGCC with CCS – Brown Coal 25.5% 41.0% 123,400 10.00
CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 17.0% 36.8% 37,200 8.40
PC Subcritical Black Coal - Retrofit CCS 26.6% 28.2% 31,000 7.00
Existing CCGT with retrofit CCS 43.0% 10.0% 17,000 9.00
Small scale CCGT Small Scale 49.3% 3.0% 10,000 4.00
Note: FOM = Fixed operating and maintenance; VOM = Variable operating and maintenance
Source: ACIL Allen AETA (2012, 2013)
Both fixed and variable O&M charges are assumed to escalate at CPI (constant in real
terms).
Table A10 shows the availability and construction profiles for each of the technologies. It is
assumed that CCS based plant would not be available prior to 2030 based on slow
international progress on demonstration plants.
Table A10 Technology availability and construction profiles

First Construction
Category Technology year period Yr1 Yr2 Yr3 Yr4
available (years)
Coal PC Supercritical – Brown Coal 2018 4 35% 35% 20% 10%
PC Supercritical Black Coal 2018 4 35% 35% 20% 10%
PC Supercritical Black Coal (SW IS Scale) 2018 4 35% 35% 20% 10%
Natural gas CCGT 2017 2 60% 40% 0% 0%
CCGT SWIS Scale 2017 2 60% 40% 0% 0%
OCGT 2016 1 100% 0% 0% 0%
Solar CLFR 2018 3 50% 30% 20% 0%
CLFR with storage 2018 3 50% 30% 20% 0%
Parabolic trough 2018 3 50% 30% 20% 0%
Parabolic trough with storage 2018 3 50% 30% 20% 0%
Central Receiver 2018 3 20% 60% 20% 0%
Central Receiver with storage 2018 3 50% 30% 20% 0%
Solar PV Solar PV fixed 2016 2 70% 30% 0% 0%
Solar PV single axis tracking 2016 2 70% 30% 0% 0%
Solar PV dual axis tracking 2016 2 70% 30% 0% 0%
A-37
First Construction
Category Technology year period Yr1 Yr2 Yr3 Yr4
available (years)
Wind On-shore Wind Farm 2016 2 80% 20% 0% 0%
Ocean/Wave 2025 2 60% 40% 0% 0%
Biomass Other Biomass Waste 2017 2 20% 80% 0% 0%
Geothermal Geothermal HSA 2020 3 40% 40% 20% 0%
Geothermal EGS 2020 3 40% 45% 15% 0%
CCS PC Supercritical with CCS – Brown Coal 2030 4 35% 35% 20% 10%
PC Supercritical with CCS – Bituminous Coal 2030 4 35% 35% 20% 10%
PC Oxy Combustion Supercritical with CCS 2030 4 35% 35% 20% 10%
CCGT with CCS 2030 2 60% 40% 0% 0%
IGCC with CCS – Bituminous Coal 2030 3 20% 60% 20% 0%
IGCC with CCS – Brown Coal 2030 3 20% 60% 20% 0%
CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 2030 3 25% 60% 15% 0%
PC Subcritical Black Coal - Retrofit CCS 2030 3 25% 60% 15% 0%
Existing CCGT with retrofit CCS 2030 3 25% 60% 15% 0%
Small scale CCGT Small Scale 2017 2 60% 40% 0% 0%
Source: ACIL Allen, AETA (2012, 2013)
Table A11 shows the assumed economic life for each technology taken from AETA. As with
incumbent generation, refurbishments are also applied to new entrants with the
refurbishment capital cost expressed as a percentage of a new facility and resulting in a life
extension expressed as a percentage of the original life. Installations can undergo multiple
refurbishments within the projection horizon.
Table A11 Technology life and refurbishment costs
Refurbishment Additional life Additional life

Economic life
Category Technology cost (% of (% of original from refurb
(years)
new) life) (years)
Coal PC Supercritical – Brown Coal 50 25% 30% 15

PC Supercritical Black Coal 50 25% 30% 15
PC Supercritical Black Coal (SWIS Scale) 50 25% 30% 15

Natural gas CCGT 30 70% 100% 30
CCGT SWIS Scale 30 70% 100% 30
OCGT 30 85% 100% 30
Solar CLFR 40 75% 100% 40
CLFR with storage 40 75% 100% 40
Parabolic trough 35 75% 100% 35
Parabolic trough with storage 35 75% 100% 35
Central Receiver 35 75% 100% 35
Central Receiver with storage 40 75% 100% 40

Solar PV Solar PV fixed 35 75% 100% 35
Solar PV single axis tracking 35 75% 100% 35
Solar PV dual axis tracking 35 75% 100% 35

Wind On-shore Wind Farm 25 50% 100% 25
Ocean/Wave 25 75% 100% 25
Biomass Other Biomass Waste 30 75% 100% 30
Geothermal Geothermal HSA 40 75% 100% 40
Geothermal EGS 40 75% 100% 40
A-38
Refurbishment Additional life Additional life
Economic life
Category Technology cost (% of (% of original from refurb
(years)
new) life) (years)
CCS PC Supercritical with CCS – Brown Coal 50 25% 30% 15

PC Supercritical with CCS – Bituminous Coal 50 25% 30% 15
PC Oxy Combustion Supercritical with CCS 50 25% 30% 15
CCGT with CCS 45 50% 50% 23
IGCC with CCS – Bituminous Coal 30 50% 50% 15
IGCC with CCS – Brown Coal 30 50% 50% 15

CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 30 25% 30% 9
PC Subcritical Black Coal - Retrofit CCS 30 25% 30% 9
Existing CCGT with retrofit CCS 30 50% 50% 15

Source: ACIL Allen, AETA (2012)
A.4.6 Carbon transport and storage costs

For plant that utilise carbon capture, transport and storage costs are applied separately. As
the majority of costs related to transport and storage of CO2 are large upfront fixed costs
(pipeline construction and drilling costs), it is appropriate for these to be levied to new
entrant technologies as a fixed charge rather than through variable charges. This can be
done either through an addition to the capital cost or through an addition charge to the fixed
O&M cost. ACIL Allen adds the cost to fixed O&M values.
Costs for CO2 transport and storage are uncertain and highly dependent upon the scale of
the development for both transmission pipelines and injection infrastructure. A larger CO2
pipeline grid would result in significant economies of scale over a single coal-fired power
station development.
The assumed transport and storage costs are presented in Table A12. These assumptions
have been informed by the AETA 2012 study. Costs are assumed to remain constant in real
terms over the modelling period.
Table A12 Assumed CO2 transport and storage costs

Region $/tonne CO2-e
NSW 72
QLD 23
SA n/a
TAS n/a
VIC 22
SWIS 14
NWIS 19
DKIS n/a
Mt Isa n/a
Source: AETA (2012)
A.5 SRES model and assumptions
A.5.1 Model overview

ACIL Allen’s forecasts for uptake of small-scale generation units (SGUs) are based on a
regression model relating historic uptake to historic net financial returns to installing solar PV
A-39
systems (the most common form of SGUs). This historic relationship is then applied to the
forecast level of net financial returns to predict future uptake of solar PV.
SGUs comprise renewable generators of less than 100 kW capacity that are eligible to
create STCs under the SRES. These generators can be installed by households and
commercial or industrial premises. Available data on these installations do not distinguish
between installations by different classes of customer and so it is difficult to separately
analyse these different customer types. In practice, residential and commercial/industrial
installations are incorporated within a single regression model (reflecting the undifferentiated
underlying uptake data) and delineated using the simplifying assumption that all installations
of more than 7.5 kW are commercial or industrial installations, and the remainder are
residential. This assumption is consistent with the observation that the vast majority of
historic PV installations have been made by households. Changes in future PV uptake
trends, particularly increasing rates of commercial installations, are discussed further in
section A.5.9.
The model uses a quarterly resolution and separately estimates uptake for each state and
territory. The regression model is based on observations for 21 quarters, from the start of
2009 to the first quarter of 2014.
Model assumptions relate principally to either historic uptake of solar PV (the regression
model’s ‘dependent variable’) or to the real net financial return to solar PV installation (the
regression model’s key ‘explanatory variable’). These are discussed separately below.
Further, as real financial returns are driven by several distinct factors, these are discussed
separately. These factors are:
 PV system installation costs
 Rebates and subsidies
 Electricity prices
 Payments for exported electricity, generally known as ‘feed-in tariffs’ or ‘buyback rates’
 System output and export assumptions.
A.5.2 Historic uptake

Historic uptake is based on Clean Energy Regulator (CER) data on STC creation by solar
installations. This data was up to date as of the end of April 2014. As STC creation can
occur up to 12 months after installation, the CER data was adjusted for lag to estimate the
likely ‘underlying’ level of installations in the period April 2013 to March 2014 based on raw
STC creation data.
Table A14 shows the lag factors used to adjust installed capacity in each month based on
data on installations or the period April 2012 to Mach 2013.
A-40
Table A13 Lag between installation and registration of PV installations with
the CER
Time since installation Percentage of capacity registered Lag Factor
Months % -
0 39% 2.57
1 74% 1.34
2 85% 1.18
3 90% 1.11
4 93% 1.08
5 95% 1.06
6 96% 1.04
7 97% 1.03
8 98% 1.02
9 98% 1.02
10 99% 1.01
11 99% 1.01
12 100% 1
Note: 0 months since installation indicates the installation was registered in the same month as the
installation occurred
Source: ACIL Allen Analysis of CER data
As noted above, this historic data set does not distinguish between residential and non-
residential PV installations, and ACIL Allen separates installations into these two categories
on the basis that all installations under 7.5 kW are residential.
To facilitate this delineation, solar PV uptake data will be broken down into eight size
categories for each state and territory in each quarter of analysis. The eight sizes are:
 Less than 1.5 kW
 1.5 to 2.5 kW
 2.5 to 3.5 kW
 3.5 to 4.5 kW
 4.5 to 5 kW (this category was chosen as some feed-in tariffs were not available for
systems of more than 5 kW of capacity)
 5 to 7.5 kW
 7.5 to 10 kW (this category was chosen as some feed-in tariffs were not available for
systems of more than 10 kW of capacity)
 More than 10 kW.
The first six categories are taken to consist of residential systems and the last two will be
commercial or industrial (‘non-residential’) systems. Shares of installed capacity in each
state and territory for systems in these size categories will be tracked through the historic
data, and the net financial return per kilowatt for each category will be weighted on this basis
to produce a single explanatory variable for each jurisdiction: the ‘weighted real net financial
return per kilowatt’.
A.5.3 PV system costs
Average cost per kW

The cost of installing a PV system has varied over time. ACIL Allen’s estimates of historic
system cost were derived by taking a national average system cost which was scaled to
account for differences in cost due to system size and to account for differences in system
A-41
costs between different states and territories. No allowance was made for the cost of
inverter replacement or for ongoing system maintenance costs.
For the period from October 2012 to May 2014 (inclusive) the national average cost of
installing a PV system was based on SolarChoice’s PV Price Check publication.56
That publication sets out prices for systems of different sizes in each capital city, which were
adjusted for GST and Small-scale Technology Certificate (STC) values to estimate an
underlying system cost. The city level estimates were used to derive a national average
system cost by weighting in proportion to the number of systems installed in each state or
territory.
Before December 2012, PV Price Check was unavailable, so different data sources were
used. The estimated national average cost of installing a PV system between January 2009
and September 2012 (inclusive) was based on:
 from 2009 to mid-2010, AECOM analysis of PV system costs for the NSW Government
(published October 2010)
 from 2010 to November 2011, ACIL Allen (then ACIL Tasman) reviews of internet quotes
for PV systems undertaken as part of analysis for the Clean Energy Regulator (late
2010, mid-2011, late 2011)
 between November 2011 and September 2012 the cost was assumed to move in a
linear fashion between ACIL Allen’s last estimate and Solar Choice estimates.
The national average system cost values are summarised in Figure A19.
Figure A19 National average historic PV installation cost (2011$/kW)

$10,000
$9,000
$8,000
$/kW installed (2011)
$7,000
$6,000
$5,000
$4,000
$3,000
$2,000
$1,000
$0
2009 2010 2011 2012 2013
Note: Cost excludes rebates, subsidies, and GST

Source: AECOM; ACIL Allen; SolarChoice
Small-scale costs were projected by:

 taking the latest data on system cost reported by Solar Choice and adjusted for STC
rebates as a starting point
56 See www.solarchoice.net.au. These are also published from time to time in sources such as Climate Spectator. See for
example, http://www.businessspectator.com.au/article/2013/12/13/solar-energy/solar-pv-price-check-%E2%80%93-
december
A-42
 assuming a split of total installation cost into labour cost, foreign equipment and local
equipment cost
 growing the separate cost components of roof-top PV installations at the same rate as
for large-scale PV from AETA (including exchange rate adjustments as relevant),
noting that the local labour and components element of small-scale PV will be larger
than for large-scale PV
 summing the three cost elements to produce a total cost series.
Figure A20 shows the average cost of installing a residential and commercial roof-top
solar system projected on this basis and in caparison to large-scale utilise PV systems
on a per kW basis.
Figure A20 National average projected PV installation cost (2014$/kW)
3,500
3,000
2,500
Real 2014 $/kW
2,000
1,500
1,000
500
0
2017
2018
2036
2037
2038
2014
2015
2016
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2039
2040
Residential roof top solar Commercial roof top solar Utility scale PV fixed
Note: Cost excludes rebates, subsidies, and GST

Source: ACIL Allen based on Solar Choice, and AETA
A.5.4 Variation by system size and location

Solar Choice’s PV Price Check data were also used to estimate a cost premium or discount
for each state and territory based on averaged variations across the period. Similarly,
smaller and larger systems were given a premium or discount based on observed variation
from the average.
The relative premia/discounts associated with different sized systems are set out in Table
A14.
Table A14 PV installation premium/discount by system size

System size (kW) <1.5 1.5-2.5 2.5-3.5 3.5-4.5 4.5-7.5 >7.5
Premium/discount 12.5% 5.2% -3.0% -5.2% -9.5% -15%
Note: the 4.5 to 7.5 kW category incorporates the 4.5 to 5 kW and 5 to 7.5 kW categories. Similarly, the
>7.5 kW category incorporates the 7.5 to 10 kW category and the >10 kW category.
Source: ACIL Allen analysis of SolarChoice data
The premia/discounts associated with installations in different states and territories are set
out in Table A15.
A-43
Table A15 State/territory variation in system cost
State/territory Installation premium/discount
New South Wales -6.2%
Victoria 3.0%
Queensland 3.8%
South Australia 2.9%
Western Australia -6.4%
Tasmania 3.0%
Northern Territory 3.8%
Australian Capital Territory 3.6%
Note: Northern Territory value is set as the Queensland value
Source: Solar Choice
Although these system size and locational premia/discounts were based only on data from
late 2012 to early 2014, they were applied to system costs throughout the historic and
forecast periods.
A.5.5 Rebates and subsidies

Two sources of upfront rebates and subsidies for PV installations were taken into account:
 the former Solar Homes and Communities Program (SHCP), which provided an
upfront cash rebate
 the indirect subsidy provided by the creation of STCs under the SRES, including the
creation of additional STCs through the ‘Solar Credits multiplier’.
Under SHCP, customers who installed PV systems received an upfront rebate of $8,000.
SHCP was in place at the beginning of 2009, and was closed during June 2009. However,
as systems installed in the second half of that based on prior applications for the rebate, it is
analysed as having an effect on some installations in the second half of 2009.
In addition to the upfront payment through SHCP, PV systems were eligible to create
certificates for the renewable electricity they generate during the historic period. The value
of these certificates (initially RECs created under the Renewable Energy Target and then
STCs created under the SRES) provides an upfront subsidy to installation of PV systems.
The value of this subsidy is depended on system size and certificate price. From June 2009
until 31 December 2012, it also depended on the ‘solar credit multiplier’, which was
established under the Solar Credits scheme and allowed eligible customers who installed
PV systems were deemed to create additional RECs/STCs, thereby increasing the amount
of the subsidy. The multiplier was originally 5, meaning that a PV system would create 5
solar credits for every MWh of electricity it was deemed to generate, for the first 1.5 kW of
capacity installed. The multiplier then declined over time.
The SHCP was phased out in favour of Solar Credits during 2009. Customers could benefit
from either the SHCP or the Solar Credits multiplier, but not both. To address the overlap
between these two policies, 50% of PV installations in quarter 3 2009, and 20% in quarter 4
2009 were assumed to receive the SHCP rebate. The remainder were assumed to use the
Solar Credits multiplier to generate extra STCs (then RECs).
The solar multiplier and certificate values factored into the analysis are shown in Table A16.
In effect, a PV system installed in 2009 was assumed to receive part of the SHCP grant and
part of its entitlement through Solar Credits.
A-44
Table A16 Solar Credits multiplier
Until July Q1 2010 From
Q3 2011 – Q3 & Q4
2009 Q3 2009 Q4 2009 – Q2 January
Q2 2012 2012
2011 2013
Solar
Credits 1 3.0 4.2 5 3 2 1
multiplier
SHCP
$8,000 $4,000 $1,600 $0 $0 $0 $0
value
Note: Q3 2009 and Q4 2009 multipliers are ‘implicit’ multipliers based on relative uptake of Solar Credits
and the SHCP rebate. Years and quarters are shown on a calendar year basis.
Source: ACIL Allen; Renewable Energy (Electricity) Regulations 2001
Unlike the SHCP payment, the value of RECs/STCs, and therefore the total rebate derived
from these certificates, varied over time. The assumed values from 2009 to the present are
shown in Figure A21. Beyond 2014, the certificate price and multiplier were assumed to
remain constant (in nominal terms), at $38 per certificate, which is just below the legislated
maximum.
Figure A21 REC/STC prices (nominal $/certificate)

$60
$50
nominal $/certificate
$40
$30
$20
$10
$0
2009 2010 2011 2012 2013
Note: REC prices prior to Q1 2011, STC prices subsequently.

Source: AFMA; ACIL Allen analysis
All systems are assumed to create 15 years of ‘deemed’ RECs/STCs at the time on
installation, and then cease to be eligible for further certificates after 15 years. The eligibility
timeframe for deeming is limited to 2030 such that as we approach 2030, the number of
years ‘deemed’ declines.
A.5.6 Retail electricity prices

Retail electricity prices are important to the financial return on solar PV as every kWh of
solar output that is consumed by the owner of the system avoids the variable component of
the retail electricity price.
For Tasmania, Western Australia, the Northern Territory and the Australian Capital Territory,
historic retail prices were adopted based on published regulated retail prices in the relevant
periods. This approach was adopted because, for the former three jurisdictions, there was
no retail competition in that period, and so no other prices were available. In the case of the
ACT retail competition was in place, but the regulated prices were sufficiently low and
A-45
competitive that ACIL Allen assumes minimal discounts to that level were available and the
published prices are sufficiently reflective of prices paid by consumers.
In the cases of New South Wales, Victoria, Queensland and South Australia, historic retail
prices are estimated based on a ‘cost-stack’ estimate, combining, wholesale, network,
‘green scheme’ and retail components. The variable component of these tariffs are assumed
to be all costs other than the fixed component of network tariffs, and advanced metering
infrastructure (also known as smart meters’) charges in Victoria. Separate cost series will be
developed for residential and non-residential retail prices, and applied to determine financial
returns to systems in the appropriate size categories.
A.5.7 Feed-in tariffs and buyback rates

When solar PV systems produce more power than is required at the premises at which they
are installed, the electricity is exported to the grid and on-sold to other customers. The value
of this exported electricity is another important component of the financial return to PV
installation.
For clarity, this report distinguishes between ‘feed-in tariffs’ and ‘buyback rates’. Within this
categorisation, ‘feed-in tariff’ refers to a premium rate determined by legislation that must be
paid for exported electricity from eligible PV systems.
In general, exported PV output always displaces electricity that would otherwise be
purchased from the wholesale market, and therefore provides some value to the retailer that
on-sells this electricity. Accordingly, retailers that supply power to owners of PV systems are
generally willing to pay some amount for exported PV output that is separate from, and
additional to, any premium feed-in tariff that might be imposed by legislation. The term
‘buyback rate’ refers to these payments by retailers that reflect the value of exported PV
output to the retailer, and which, whilst sometimes regulated, are not intended to offer a
premium rate or purposefully subsidise PV systems.
Within this categorisation, it is necessary to distinguish between three types of feed-in tariffs:
 A ‘net’ feed-in tariff is the most common form, and pays a premium rate for all exported
PV output
 A ‘gross’ feed-in tariff meters PV output in such a way that all PV output is effectively
exported, earning whatever premium rate is available, and then all of the customer’s
electricity is then imported at the prevailing retail rate
 A ‘one-for-one’ feed-in tariff establishes that the payment for exported PV output must
be equal to the prevailing retail rate.
With this nomenclature established, Table A17 sets out the various feed-in tariffs that have
been or are in operation in Australia.
A-46
Table A17 Feed-in tariffs by jurisdiction
Initial feed-
Subsequent Current
in tariff Comments Comments
feed-in tariff arrangements
regime
Commenced
early 2010, Buyback rate
Breached cap in
New South phased out regulated to be
60c gross 20c gross Q3 2011. Paid to
Wales through early between 6.6 and
end 2016
2011. Paid to 11.2 c
end 2016
October 2009 to
Breached cap in Regulated
September
Victoria 60c net 25c net mid-2012. Paid to minimum buyback
2011. Paid to
end 2016 rate of 8c
end 2024
2008 to mid- Feed-in tariff of 8c
Queensland 44c net 2013. Paid to 8c net until 30 June
June 2028 2014
2008 to
October 2011 to Regulated
South September
44c net 16c net September 2012. minimum buyback
Australia 2011. Paid to
Paid to June 2016 rate of 7.6c
June 2028
Synergy offers a
Mid 2010 to mid- In place July and buyback rate of
Western 2011. Paid for August 2011. 8.8529c; Horizon
40c net 20c net
Australia 10 years from Paid for 10 years offers location-
installation from installation specific rates of
10-50c
One-for-one feed-
Tasmania One-for-one Still in place
in tariff
Northern One-for-one feed-
One-for-one Still in place
Territory in tariff
50.05c
gross, then April 2009 to
Australian ActewAGL offers
45.7c gross mid-2011. Paid One-for- Mid 2011 to mid-
Capital a buyback rate of
from for 20 years from one 2013
Territory 7.5c
October installation
2010
Note: All rates are in nominal cents per kWh.
Buyback rates are generally much lower than feed-in tariffs, typically around 7 to 10 cents
per kilowatt hour, and vary only slightly by jurisdiction.
A.5.8 System output and export rates

System output is estimated based on four solar zones created by the CER for the purpose
of calculating REC and STC creation by solar PV, which have different assumed rates of
solar output per kW of installed capacity. Each postcode is assigned a zone, whereas
multiple solar zones may existing in a given state or territory. Accordingly, the share of
installations in each zone for each state and territory are based on historic installations in
CER data to March 2014.
The solar zone ratings are set out in Table A18, and the historic share of installed capacity
by zone, and implied average output per kW of installed capacity for each state and territory,
are set out in Table A19.
Table A18 Solar zone ratings

Zone 1 Zone 2 Zone 3 Zone 4
Output (MWh per
kW of installed 1.622 1.536 1.382 1.185
capacity)
Source: Clean Energy Regulations 2001, Schedule 5
A-47
Table A19 Installed capacity by solar zone
Implied
average
Zone 1 Zone 2 Zone 3 Zone 4
output
(MWh/kW)
New South
0.0% 5.1% 93.2% 1.7% 1.386
Wales
Victoria 0.0% 0.0% 8.6% 91.4% 1.202
Queensland 0.0% 1.1% 98.9% 0.0% 1.384
South
0.0% 1.3% 95.6% 3.1% 1.378
Australia
Western
1.9% 5.5% 89.5% 3.1% 1.389
Australia
Tasmania 0.0% 0.0% 0.0% 100.0% 1.185
Australian
Capital 0.0% 0.0% 100.0% 0.0% 1.382
Territory
Northern
33.2% 65.6% 1.3% 0.0% 1.563
Territory
Source: ACIL Allen analysis of CER data
Assumed export rates by state and by system size are set out in Table A20. Export rates are
assumed to be 10% for all systems above 7.5 kW on the basis that most larger systems will
be commercial or industrial systems, and will be sized to minimise exports and thereby
maximise avoided network tariffs.
Table A20 Export rates

<1.5 1.5-2.5 2.5-3.5 3.5-4.5 4.5-7.5 >7.5
New South
35% 40% 45% 50% 55% 10%
Wales
Victoria 35% 40% 45% 50% 55% 10%
Queensland 35% 40% 45% 50% 55% 10%
South
40% 45% 50% 55% 60% 10%
Australia
Western
40% 45% 50% 55% 60% 10%
Australia
Tasmania 30% 35% 40% 40% 45% 10%
Australian
Capital 35% 40% 45% 50% 55% 10%
Territory
Northern
35% 40% 45% 50% 55% 10%
Territory
Source: ACIL Allen assumptions
A.5.9 System size trends

Data on PV uptake illustrates clearly a substantial increase in average system size over
time, in particular as the incentives created by the Solar Credits formula to install smaller
systems has dissipated and been overwhelmed by the attraction of lower system costs.
Further, although CER data does not distinguish between residential and non-residential
systems, ACIL Allen analysis for various distribution businesses indicates recent growth in
the share of larger systems being proposed by non-residential customers. This trend is
widely anticipated within the PV industry, and reflects the financial opportunity from low
A-48
installation costs and the opportunity to avoid large variable components of retail prices
across a broad range of retail tariff classes.
Addressing these trends in analysing historic uptake and paybacks is relatively trivial. As
discussed above, net financial return per kilowatt across different system sizes will be
weighted on the basis of their historic share of installed capacity in each jurisdiction to
produce a single ‘weighted real net financial return per kilowatt’.
However, this issue is slightly more complicated for future installations. Of particular
importance to the forecasting approach is the fact that large/non-residential systems
represent only a small share of historic installations, but are widely expected to grow
strongly from this low base in the future. Further, the low levels of uptake of larger systems
cannot itself be explained by net financial returns per kilowatt, as returns for these systems
appear to have been clearly positive in many jurisdictions for at least a year. In simple
terms, it appears to have taken some time for the PV industry to target its marketing efforts
to this niche, and for non-residential customers to understand the financial opportunity of
installing PV. Given the lagged and relatively recent emergence of this market niche,
econometric analysis of historic uptake and paybacks to larger systems may tend to
underestimate the potential response of commercial installations to improving paybacks. To
address this issue, ACIL Allen has undertaken desktop research to reflect the potential
growth of large, non-residential PV installation rates. The approach used was broadly:
 Estimate the potential for growth in larger (non-residential) systems based on various
industry analyses and compare this growth to forecasts of residential uptake rates
 Increase the share of installed capacity attributed to larger systems over time to reflect
this potential growth, and hold this share constant across different modelling scenarios
 This higher weighting should bring down weighted net financial returns in the future, as
large non-residential systems typically have strong paybacks due to low export rates
(implying a high level of avoided network charges) and lower installation costs
 In turn, the overall level of installations and the level of non-residential shares should
grow over time in a way that reflects changes in their payback
 Importantly, the change in overall uptake between policy scenarios (e.g. in response to
phasing out or closing the SRES) will be determined on an internally consistent basis
across various scenarios.
Work done by Beyond Zero Emissions (BZE)57 indicates that commercial roof-top space
suitable for PV installations is much more limited than that on residential premises. BZE’s
estimates are based on an estimate of commercial and residential floor space and average
roof to floor space ratios. Table A21 shows BZE’s estimates of commercial and residential
building footprints.
57 Beyond Zero Emissions, Melbourne Energy Institute, The University of Melbourne, August 2013
A-49
Table A21 Building footprints as of 2011
State/territory Residential CBD Non residential Non CBD Non Total
Residential
Unit km2 km2 km2 km2
NSW 393 0.56 25.1 418
VIC 319 0.45 24.4 344
QLD 279 0.31 16.5 296
SA 97 0.1 5.2 102
WA 146 0.14 11.3 157
TAS 30 0.15 3 33
NT 10 0 0.9 11
ACT 20 0 1.6 21
Note: CBD stands for Central Business District

Source: Beyond Zero Emissions, Melbourne Energy Institute, The University of Melbourne, August 2013
Two studies by Entura on the ratio between floor and roof space were presented by BZE
and reproduced in Table A22.
Table A22 Commercial roof-space apportioning factor

Type Floor-space Roof-space Roof to floor space ratio Study
Unit km2 km2 - -
Commercial 3.11 1.08 0.35 City of Boroondara
Commercial 2.6 1.0 0.38 City of Port Philipp

Data Source: Beyond Zero Emissions, Melbourne Energy Institute, The University of Melbourne, August
2013 citing (Entura, 2011)
The roof-space identified in Table A22 includes space that may not be suitable for the
installation of solar panels such as shaded areas or areas covered by air-conditioning units.
To account for this non-usable space usable space factors (USF) of 62% for residential
roofs, 15% for non-residential roofs in CBD locations and 50% for non-residential roofs
outside CBD locations.
For its estimate of the total PV capacity installable on commercial roofs, ACIL Allen has
adopted the USF ratios presented by BZE and the roof to floor space ratio of the City of Port
Philipp study. Final estimates of installable capacity on commercial roofs are shown below.
Table A23 Estimated installable roof-top capacity

State/Territory Residential CBD Non residential Non CBD Non Residential
Unit MW MW MW
NSW 21,771 5 731
VIC 17,671 4 711
QLD 15,456 3 481
SA 5,373 1 152
WA 8,088 1 329
TAS 1,662 1 87
NT 554 0 26
ACT 1,108 0 47
Source: ACIL Allen based on data from Beyond Zero Emissions, Melbourne Energy Institute, The
University of Melbourne, August 2013 and (Entura, 2011)
A-50
In the first quarter of 2014 82% of installed PV capacity across Australia was installed in
smaller residential systems (i.e. systems smaller than 7.5kW). Figure A22 shows the share
of installed capacity across Australia by system size class.
Figure A22 Share of installed capacity by system size class
100%
90%
80%
% of installed capacity
70%
60%
50%
40%
30%
20%
10%
0%
Quarter Ending
1.5 - kW 2 - kW 3 - kW 4 - kW 5 - kW 6 - kW 8 - kW 10 - kW
Note: 10-kW includes systems of 10-100kW

Source: ACIL Allen Analysis of CER data
ACIL Allen has estimated the growth in the market share of larger commercial systems
based on a logistic growth function. The logistic growth function was fitted to quarterly data
on the share of installed PV capacity that came from systems larger than 7.5 kW by
determining the value for the parameter a in the equation below for each state:
(−𝑎∗𝑡)
𝑆 (𝑡) = 𝑒 ln(𝑆(0))∗𝑒
Where:
 S(t) is the share of installed capacity from large non-residential systems
 S(0) is the share of installed capacity from large non-residential systems for Q1 2009
 t is the number of quarters since Q1 2009
 a is a constant to fit the curve to data for each of the investigated network regions.
The formula presented above grows the share of installed PV capacity consisting of
non-residential systems to eventually approach 100% without taking into account that only
limited roof-space is available for the installation of commercial systems. In order to account
for the predicted saturation of commercial roof-space with PV installation, the predicted
share of installed capacity was reduced to reflect predicted saturation of commercial roof-
space in an iterative process.
A.6 Retail model

Retail prices were modelled using a ‘cost-stack’ approach accounting for the major
components of retail prices, namely:
 Wholesale energy costs
 Network costs
A-51
 RET costs
 Costs associated with other ‘green schemes’
 Costs attributable to losses
 Metering costs, retail operating costs and retail margins.
Retail prices were developed for the six jurisdictions in the NEM (NSW, Victoria,
Queensland, South Australia, Tasmania and the ACT), south-west Western Australia and
the Darwin-Katherine grid in Northern Territory. Retail prices were calculated on a calendar
year basis.
The retail series were developed on the basis of a mix of notional customers within four
broad customer classes: residential, commercial, industrial and emissions-intensive trade-
exposed (EITE) industry. The industrial and EITE categories do not overlap, such that the
industrial category is effectively ‘light industry’. Modelled prices for the EITE category are
adjusted to account for partial exemptions from RET costs, as described in section A.6.3.
Three separate retail series were developed:
 An ‘all-inclusive’ retail series that combines fixed, demand, capacity and variable
(energy basis) charges into a single average cost, presented in terms of cents per
kilowatt hour. This series is the core output reported in this report.
 A series representative of the retail charges avoided by own-consumed PV output,
which excludes all fixed, demand and capacity charges, as well as ‘off-peak’ energy
charges. Further, the wholesale component of this series was adjusted to transition
over time away from the wholesale costs associated with supplying the relevant retail
load, to the wholesale market price of solar generation. This adjustment was made on
the assumption that metering advances and increasing solar penetration would
promote a policy and market transition towards accounting for solar output in a way
that more closely reflects its market value. This is discussed in more detail in section
A.6.5. This series was used as an input into the SRES forecasting model.
 A series representative of the ‘buyback rate’ paid to exported PV output, comprising
the wholesale value series described in relation to own-consumed PV output, adjusted
for losses. This series was used as an input into the SRES forecasting model.
A.6.1 Wholesale energy costs

Slightly different estimates for wholesale energy costs were made based on available
information. For the NEM, the following process was followed:
 Historic NEM demand for the years 2008 to 2012 was decomposed into residential,
commercial, industrial and EITE components by:
 Allocating the net system load profile (NSLP) load to the residential category
 Allocating the demand component identified as ‘industrial’ by AEMO in its 2013
National Electricity Forecasting report to the EITE category, on the basis that this
load is constant over the course of each year
 Separating the remainder into commercial and industrial components on the basis of
an assumed industrial load shape involving demand that is 25% lower in off-peak
times than peak times.
 Calculating ‘uplift’ factors based on this historic decomposition for the non-residential
categories, which are then held constant throughout the projection
 Calculating uplift factors for the residential category that changes over time in
response to modelling outcomes by:
A-52
 Regressing historic NSLP load against historic NEM load and time-of-day and
seasonal characteristics for each NEM region and the ACT
 Using this observed relationship to forecast residential load based on modelled future
NEM region loads
 Adjusting this relationship for the growth in rooftop solar
 Calculating the difference between the average modelled price and the modelled
‘uplifted’ residential price in the Reference case to determine uplift factors for each
NEM region and the ACT in each model year
 Holding these factors constant across all model scenarios and sensitivities.
Uplift factors for each NEM region and the ACT for the non-residential customer classes are
shown in Table A24. For the DKIS, simpler factors were assumed due to a lack of detailed
historic data, also shown in Table A24, and were estimated based on analysis by the AEMC
in 2012 and 2013 of trends in residential retail prices.
Table A24 Uplift factors by customer class

NSW Time varying 8% 5% 0%
Queensland Time varying 4% 4% 0%
South Australia Time varying 15% 7% 0%
Tasmania Time varying 5% 5% 0%
Victoria Time varying 10% 5% 0%
ACT Time varying 8% 5% 0%
DKIS 60% 30% 15% 0%
Source: ACIL Allen analysis of AEMO data and AEMC analysis
Additional hedging factors were assumed for each NEM region to reflect the cost of hedging
against higher price events (e.g. in the event of high demand events or combinations of
outages that cause price spikes). These hedging factors were based on 2011 ACIL Allen
(then ACIL Tasman) analysis for the AEMC on retail price trends, which found that hedging
costs were a relatively stable uplift from load-weighted residential prices. In the SWIS,
constant real capacity credit costs were assumed for each customer class. Assumed
hedging cost factors and capacity credit costs are shown in Table A25.
Table A25 Hedging factors and capacity credit costs by customer class
Hedging factors expressed as a percentage

NSW 13% 13% 10% 10%
Queensland 16% 16% 10% 10%
South Australia 17% 17% 10% 10%
Tasmania 14% 14% 10% 10%
Victoria 15% 15% 10% 10%
ACT 10% 10% 8% 8%
DKIS 10% 10% 8% 8%
Capacity credit costs expressed as real $2014 per MWh
SWIS $61 $44 $33 $22
Carbon costs for the first half of 2014 (before the assumed repeal effective from 1 July
2014) are included within the wholesale energy component.
A-53
A.6.2 Network costs
Network costs can be analysed heavily based on published regulated network charges, and
expected changes in these charges based on forward looking regulatory determinations.
Accordingly, near term trends in network costs can be analysed with a reasonable degree of
confidence.
Published historic prices to 2013-14 were analysed for all regions other than Victoria, with
Victoria’s network prices changing on a calendar year basis and so published prices for
2014 were included in the analysis. Representative tariffs were applied to a range of
notional customers (with assumed average and peak consumption), and then weighted
within customer categories.
Future prices were projected on regulatory determinations where available. Beyond the
regulatory determination period for each region (which ‘roll off’ at varying times), specific
capital and operating expenditure trends for each network were extrapolated and then
gradually harmonised over time to project the core cost components of the return on capital
and operating expenditure using the established ‘building block’ methodology for network
cost calculations. The regulated return on capital (also known as the weighted average cost
of capital, or WACC) was assumed to normalise to 8% for all networks from the end of each
regulatory determination, with this rate reflecting recent Australian Energy Regulator
decisions (e.g. its April 2014 placeholder decision for NSW and ACT distribution networks,
which set a WACC of 8.05%, and its April 2012 decision for the Tasmanian distribution
network, setting a WACC of 8.28%).
Table A26 summarises the historic, regulatory determination based projection and modelled
projection of network costs for each network.
Table A26 Basis of network cost calculation

Projection based on Projection based on
Element Networks Historic regulatory modelled
determinations assumptions
Ausgrid, Endeavour
NSW distribution and
Energy, Essential To 2013-14 2014-15 to 2018-19* From 2019-20
sub-transmission
Energy
NSW transmission TransGrid To 2013-14 2014-15 to 2018-19* From 2019-20
CitiPower, Jemena,
Powercor, SP
Victoria distribution To 2014 2015 From 2016
AusNet, United
Energy
Victoria transmission SP AusNet To 2014 2015 to March 2017 From March 2017
Queensland Energex, Ergon
To 2013-14 2014-15 From 2015-16
distribution Energy
Queensland
Powerlink To 2013-14 2014-15 to 2016-17 From 2017-18
transmission
SA distribution SAPN To 2013-14 2014-15 From 2015-16
SA transmission ElectraNet To 2013-14 2014-15 to 2017-18 From 2018-19
Tasmania
Aurora To 2013-14 2014-15 to 2016-17 From 2017-28
distribution
Tasmania
Transend To 2013-14 2014-15 to 2018-19* From 2019-20
transmission
ACT distribution ActewAGL To 2013-14 2014-15 to 2018-19* From 2019-20
Western Australia Western Power To 2013-14 2014-15 to 2016-17 From 2017-18
Northern Territory NT Power and Water To 2013-14 2014-15 to 2018-19 From 2019-20
* ‘Placeholder decision’ by AER
Source: AER, ERAWA, NT Utilities Commission.
Overall cost trends were divided by energy growth trends to give a cost per unit energy (i.e.
price) trend.
A-54
The share of network costs recovered from fixed, demand and capacity charges was
assumed to increase over time. This did not affect the all-inclusive retail price series, but
affects the portion of network costs avoided by own consumption of PV output. Fixed,
demand and capacity charges were assumed to comprise 50% of residential and
commercial network charges by 2030, commencing a linear transition from present levels in
the first projection year.
A.6.3 RET costs

RET costs for non-EITE customers (i.e. customers without access to partial exemptions
from the RET) are calculated as the simple multiplication of the prevailing LGC or STC price
by the ‘renewable power percentage’ (RPP) or ‘small-scale technology percentage’ (STP).
The RPP is calculated directly for each scenario based on exogenous demand targets and
assumed levels of ‘reduced acquisitions’. Reduced acquisitions are equal to ‘relevant
acquisitions’ under the scheme, less the volume of partial exemption certificates (PECs)
issued.
Relevant acquisitions were calibrated to the CER’s estimate of 2014 relevant acquisitions
(approximately 205 TWh), and then grown in line with modelled electricity demand. PECs
were calculated based on analysis of PEC creation from 2010 to 2013, and the implied level
of EITE load underlying this level of PEC creation. EITE load was assumed to reduce by
2,500 GWh in 2015 to reflect the closure of the Point Henry aluminium smelter, and then
grow in line with total demand from that point. The calibration of historic EITE load based on
historic PEC creation is shown in Table A27.
Table A27 PEC creation and EITE load

2011 2012 2013 2014
Actual Actual Actual CER estimate
Highly (actual) 26,515 32,455 30,437

PEC creation Moderately
647 996 955
(000s) (actual)
Total 27,162 33,451 31,392 27,200
Highly 68.4% 77.5% 74.7% 67.6%
Exemption rate
Moderately 45.6% 51.6% 49.8% 45.0%
Highly 38,767 41,889 40,729 39,029
Implied EITE
Moderately 1,420 1,928 1,917 1,837
load (GWh)
Total 40,187 43,827 42,646 40,867
Note: Partial exemptions differentiate between activities that are ‘highly’ or ‘moderately’ emissions
intensive. The calculation above addresses these two categories separately.
Source: ACIL Allen analysis of CER data
The STP was taken as the pre-determined STP for 2014. The 2015 STP is adjusted for
over/under achievement of the implied STC creation target in 2014 relative to ACIL Allen’s
modelled rate of STC creation. From 2016 onwards the STP is calculated as the modelled
rate of STC creation divided by reduced acquisitions.
LGC costs for the LRET cost component are taken as the certificate price modelled by ACIL
Allen. The market (as opposed to legislated) STC price paid by customers was assumed to
be $38 (constant nominal) in all scenarios other than where the SRES was repealed.
A.6.4 Green scheme costs, losses and retail costs

These relatively small cost components include:
 Various green schemes
A-55
 The Energy Savings Scheme in NSW
 The Victorian Energy Efficiency Target, assumed to close from 2016 in line with
recent Victorian Government announcements
 South Australia’s Residential Energy Efficiency Schemes
 Various feed-in tariff costs
 The cost of losses, comprising the wholesale and green scheme cost components
grossed up for energy losses (which differ by customer class and state/territory)
 Other network costs, including metering costs for examplein Victoria
 Retail operating costs (which differ by customer class and state/territory)
 Market fees and ancillary services costs for NEM regions, with these being based on
historic real costs in these categories
 Retail margins, calculated as a constant share of non-network cost components (with
slight variations by region).
A.6.5 Treatment of solar costs

As discussed above, the wholesale component of the retail cost series that is avoided by
own consumption of PV output, and for ‘buyback rates’ for solar exports, were adjusted to
transition over time towards the wholesale market value of solar generation.
This transition was assumed to occur such that:
 Residential customers would avoid the wholesale cost of serving their typical energy
consumption until 2025, after which time this rate would transition to the point where
50% of customers would avoid only the solar dispatch-weighted price by 2030.
 Commercial customers would avoid the wholesale cost of serving their typical energy
consumption until 2025, after which time this rate would transition to the point where
100% of customers would avoid only the solar dispatch-weighted price by 2035.
 Industrial and EITE customers would transition to a rate 100% reflective of the
wholesale value of solar between 2025 and 2030.
Though approximate, this transition implies that as solar penetration increases, and the
correlated solar output depresses wholesale market prices during sunny periods, typical
wholesale costs of supply and the wholesale market value of solar will diverge over time. In
response, we assume that policy-makers and retailers will implement a regulatory and/or
market response that sees rooftop solar systems metered in a way that allows the value of
their output to be more accurately credited, rather than simply credited as avoided a ‘typical’
unit of energy consumption. This transition could be implemented through metering
requirements for new solar installations, or widespread adoption of time-of-use metering for
retail load allowing differentiation between the cost of serving customers with and without
rooftop solar.
The commercial imperative for such a change is substantial, as our modelling suggests that
the value of solar output declines to a value of around 80% of the average market price by
2030 in most NEM regions, and to around 60% by 2040, as solar penetration increases.
A-56

ACIL RET Review Modelling - Full Report

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

ACIL RET Review Modelling - Full Report

Încărcat de

Drepturi de autor:

Formate disponibile

A C I L A L L E N C O N S U L T I N G

SUITE C2 CENTA BUILDING

RELIANCE AND DISCLAIMER

© ACIL ALLEN CONSULTING 2014

Cases and sensitivities

Figure ES 1 LRET annual targets under the various policy scenarios

Reference case Real 20% case Real 30% case

Table ES 1SRES settings under the various policy scenarios

15 years upfront, with deeming 10 years deeming upfront, with

10 years deeming from 1

10 years from 1 January 2015, 10 years from 1 January 2015,

Analysis and findings

Figure ES 2 Proportion of renewables in Australia’s energy mix: Reference case

Share of Australian energy

Note: Proportion of estimated total Australian electricity demand

Generation sector resource costs

Closed to New Entrants

Emissions and cost of abatement

Figure ES 4 Aggregate emissions from electricity generation: 2015-2040: All scenarios/sensitivities

Closed to New Entrants

As with any Government expenditure or program, an important consideration is whether the

Impacts on retail prices

Figure ES 5 NPV of average household total expenditure on electricity (2015-2040): All

Closed to New Entrants

Closed to New Entrants

1 Introduction and background 1

4 Closed to new entrants 33

5 Real 20% by 2020 38

6 Real 30% by 2030 51

6.1 Scenario description 51

7 50% of growth case 64

9 Abatement costs 113

10 Summary and conclusions 120

10.1.4 RET compliance costs 125

Appendix A Methodology and assumptions A1

Acronym or term Explanation

Solar Flagships, a Commonwealth Government program to promote large-scale solar generation

1 Introduction and background

1.1 Approach to the modelling

Figure 1 Modelling framework

Demand inputs LRET settings

Source: ACIL Allen

1.2 Overview of scenarios examined

Figure 2 Scenario / sensitivity matrix

1.3 Scope of modelling

Figure 3 Estimating the macroeconomic impact of a policy/program

Total economic impact

Economic output impact Terms of Foreign income

Source: ACIL Allen

1.4 Report structure

 Finally, Appendix A provides a more detailed description of the methodology employed

2.1 Scenario description

Figure 4 Proportion of renewables in Australia’s energy mix: Reference case

Share of Australian energy

Note: Proportion of estimated total Australian electricity demand

We note that previous assessments of the proportion of renewables in Australia’s electricity

Table 1 Projected energy by grid: Reference case

2014 2020 2040 2014-20 2014-40

Figure 5 Wholesale price outcomes: Reference case

Baseload Black coal

Figure 7 New entry capacity and capital expenditure: Reference case

Figure 8 New entrant capacity by grid: Reference case

TAS VIC SWIS

NWIS DKIS Mt Isa