Documente Academic
Documente Profesional
Documente Cultură
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
ACILALLEN.COM.AU
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.
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.
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Real 30% case End of calendar year 2030 No change to Reference case No change to Reference case
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Closed to New Entrants
50% Growth
Reference
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
Reference
50% Growth
Core Carbon High capital costs High demand Low demand Permanent
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
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 $164/tonne.
vii
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
viii
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
20,260
20,451
20,074
20,037
21,000
19,887
19,771
19,754
19,634
19,464
19,453
20,500
19,358
19,349
19,305
19,213
19,193
19,181
19,182
19,119
19,119
19,092
19,003
20,000
NPV of retail bills ($)
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
Core Carbon High capital costs High demand Low demand Permanent
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Executive summary ii
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
3.1 Scenario description 25
3.2 Results 25
3.2.1 RET outcomes 28
3.2.2 Resource costs, profitability and emissions 29
3.2.3 Retail price outcomes 32
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
8.2.1 Reference case results 87
8.2.2 Other policy cases 90
8.3 Shadow carbon price 93
8.3.1 Reference case results 95
8.3.2 Other policy cases 97
8.4 High capital costs 99
8.4.1 Reference case results 101
8.4.2 Other policy cases 103
8.5 Permanent retirements 106
8.5.1 Reference case results 107
8.5.2 Other policy cases 109
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.
AC I L AL L E N C O N S U L T IN G
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
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
1
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
A more detailed description of the modelling approach and key assumptions used is
presented in Appendix A.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
4
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
Direct Indirect
economic economic
contribution contribution
Factors of
Supply chain production
impacts impacts
Taxes less
Value added subsidies
Crowding out Productivity
impacts impacts
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.
5
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
6
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
2 Reference case
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
Grid / demand segment Annual energy (TWh) Average annual growth rate
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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%
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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).
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
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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
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
500 500
400
400 400
MW
MW
MW
300
300 300
200
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
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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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%
Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other Baseload Gas
10.3%
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
14
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
2,500 30.0
Real 2014 $m
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
0.82
100
0.80
80
0.78
60
0.76
40
0.74
20
0.72
0
0.70
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
22
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 17 Weighted average Australian retail prices and direct RET costs: Reference case
Average retail cost Non-exempt consumers
20
30 LRET SRES
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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%
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
3 Repeal case
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.
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
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 = SWIS, NWIS, DKIS and Mt Isa).
Source: ACIL Allen
25
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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 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.
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
27
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
28
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
-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.
2,500 -10.0
Real 2014 $m
-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
Note: Sum of projected LGC price multiplied by mandated LRET targets and projected STC creation by assumed STC price
Source: ACIL Allen
29
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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 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 $12.8
billion using a discount rate of 10% pre-tax real. The 7% discount rate is more reflective of a social discount rate whereas
10% is more suited for commercial project evaluation.
30
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Wind -14.2
1,000
Hydro 1.0
0
Baseload Gas 1.3
-1,000
Peaking Gas 0.1
-2,000
Brown coal 4.2
-3,000
Black coal 12.1
Black coal Brown coal Peaking Gas Baseload Gas -20 -15 -10 -5 0 5 10 15
Hydro Wind Solar Geothermal NPV ($ billion)
Note: EBITDA measure calculated as modelled pool revenues (energy and LGCs) less fixed operating and maintenance costs and variable
generating costs
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.
0.88
25.0
0.86
Tonnes CO2-e/MWh sent-out
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
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 and off-grid generation ‘Other’ category includes
cogeneration, liquid fuels, CCS-equipped technologies, biomass and geothermal.
Source: ACIL Allen
31
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
$400
0.50
NPV ($)
-0.50 $0
($1)
-1.00 ($200)
-1.50 ($247)
($400)
($367)
-2.00
($600)
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Calendar Incumbent Allowance for Total LGC Voluntary LGC Adjusted LRET
year LGC Creation banked LGCs supply demand demand
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
35
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
-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
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
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.
3,000 -7.9
-5.0
-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
Note: Sum of projected LGC price multiplied by mandated LRET targets and projected STC creation by assumed STC price
Source: ACIL Allen
36
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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 $0
-1.50 ($269)
($400)
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
37
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
39
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
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 = SWIS, NWIS, DKIS and Mt Isa).
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
material amount of alternative generation capacity, with the market having sufficient capacity
from incumbent generators to meet its requirements.
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
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 figure also provides the projected LGC spot price trajectory over the period to 2030.
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.
Prices trend upward to 2021 (the period in which banked LGCs exist) in accordance with the
assumed holding cost assumption of 5% real. Prices peak at around $59/LGC in 2021, then
trend down, reaching $30/LGC in 2030.
Figure 40 LRET supply-demand balance and price outcomes: Real 20% case
LRET supply-demand balance Projected price outcomes
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
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. S pot 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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
-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
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
45
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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
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
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.
-2.0 -4.7
3,000
-4.0
2,500 -0.6
Real 2014 $m
-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
Note: Sum of projected LGC price multiplied by mandated LRET targets and projected STC creation by assumed STC price
Source: ACIL Allen
47
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Total Small-scale -1.3
0
SRES: SWHs -0.6
Real 2014 $m
-1,000
SRES: SGUs -0.7
-2,000
Total Large-scale -7.9
-3,000 Unserved Energy 0.0
-4,000 Variable O&M 1.7
-5,000 Fixed O&M -2.1
-6,000 Refurbishment 0.5
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
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
Change (Real 20% case - Reference case) NPV of change (7%)
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
Note: EBITDA measure calculated as modelled pool revenues (energy and LGCs) less fixed operating and maintenance costs and variable
generating costs
Source: ACIL Allen
48
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
16.0 0.88
14.0 0.86
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
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 and off-grid generation ‘Other’ category includes
cogeneration, liquid fuels, CCS-equipped technologies, biomass and geothermal.
Source: ACIL Allen
49
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Other green schemes Losses Direct RET costs Other costs Net cost
Retailing costs Total cost
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
50
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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.
51
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
60,000
50,000
40,000
GWh
30,000
20,000
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
Note: Values exclude 850 GWh allowance for WCMG
Source: ACIL Allen
6.2 Results
Figure 50 shows the projected change in wholesale prices under this scenario and relative
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
80
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
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 = 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.
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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 53 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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
MW
1,000
30,000
0
20,000 -1,000
-2,000
10,000
-3,000
0 -4,000
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
23 The same issue applied when the original MRET scheme was extended to 2030, however no adjustment was made.
55
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
LRET supply-demand balance Projected price outcomes
60,000 100
Spot price
90 Tax-adjusted Penalty price
50,000 Spot price (Reference case)
80
40,000 70
60
30,000
50
20,000 40
30
10,000
20
0 10
0
Incumbent LGC Creation New Entrant LGC Creation
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
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.
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
57
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
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.
3,000 6.0
2,500 4.0
Real 2014 $m
6.2
2,000 2.0
Note: Sum of projected LGC price multiplied by mandated LRET targets and projected STC creation by assumed STC price
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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
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. Table only shows results to 2030, even though scheme extends to 2040.
Source: ACIL Allen
Figure 60 Change in electricity generation sector resource costs: Real 30% case
Change (Real 30% case - Reference case) NPV of change (7%)
3,000
Total -0.1
2,000
Total Small-scale -0.2
1,000
SRES: SWHs 0.0
0
Real 2014 $m
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 61 Electricity generation sector profitability by fuel type: Real 30% case
Change (Real 30% case - Reference case) NPV of change (7%)
2,000
Geothermal 0.7
1,500
500
Real 2014 $m
Wind -2.4
0
Hydro -0.8
-500
Note: EBITDA measure calculated as modelled pool revenues (energy and LGCs) less fixed operating and maintenance costs and variable
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.
10.0
0.85
Tonnes CO2-e/MWh sent-out
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
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 and off-grid generation ‘Other’ category includes
cogeneration, liquid fuels, CCS-equipped technologies, biomass and geothermal.
Source: ACIL Allen
62
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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:
$17 (0.2%) lower in the period to 2020
$233 (1.5%) lower in the period to 2030
$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
-1.00 ($200)
-1.50
($300)
($297)
Network Wholesale energy ($350)
LRET SRES 2015-20 2015-30 2015-40
Other green schemes Losses Direct RET costs Other costs Net cost
Retailing costs Total cost
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
63
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
500
450
400
350
Capacity (MW)
20kW - 100kW
300
10kW - 20kW
250 8kW - 10kW
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
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.
Figure 66 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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
80
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
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 = SWIS, NWIS, DKIS and Mt Isa).
Source: ACIL Allen
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
66
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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 69 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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
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
Black coal Brown coal Peaking Gas Baseload Gas Black coal Brown coal Peaking Gas Baseload Gas
Hydro Wind Solar Other 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
68
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
The figure also provides the projected LGC spot price trajectory over the period to 2030.
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.
Prices trend upward to 2021 (the period in which banked LGCs exist) in accordance with the
assumed holding cost assumption of 5% real. Prices peak at around $59/LGC in 2021, then
trend down, reaching $30/LGC in 2030.
Figure 71 LRET supply-demand balance and price outcomes: 50% Growth case
LRET supply-demand balance Projected price outcomes
30,000 100
90
25,000
80
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
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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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.
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
70
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
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.
-2.0 -3.3
3,000
-4.0 -1.4
2,500
Real 2014 $m
-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
Note: Sum of projected LGC price multiplied by mandated LRET targets and projected STC creation by assumed STC price
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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
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
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
Total Small-scale -2.7
0
SRES: SWHs -0.5
Real 2014 $m
-1,000
SRES: SGUs -2.2
-2,000
Total Large-scale -7.8
-3,000 Unserved Energy 0.0
-4,000 Variable O&M 1.7
-5,000 Fixed O&M -2.1
-6,000 Refurbishment 0.5
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Note: EBITDA measure calculated as modelled pool revenues (energy and LGCs) less fixed operating and maintenance costs and variable
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.
16.0 0.88
14.0 0.86
Tonnes CO2-e/MWh sent-out
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
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 and off-grid generation ‘Other’ category includes
cogeneration, liquid fuels, CCS-equipped technologies, biomass and geothermal.
Source: ACIL Allen
74
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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)
Network Wholesale energy ($400)
LRET SRES 2015-20 2015-30 2015-40
Other green schemes Losses Direct RET costs Other costs Net cost
Retailing costs Total cost
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
75
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
8 Sensitivities
This chapter provides assumptions and results for the various sensitivities modelled for the
Reference and other policy cases.
-1.7% -24.9%
Sent-out TWh
150 150
100 100
Reference Low case
50 50
0 0
2014 2020 2030 2040
-50
Calendar year Calendar year
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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 $/MWh
-10,000
60
GWh sent-out
-20,000
50
40 -30,000
30 -40,000
20
-50,000
10
0 -60,000
-70,000
Reference case NEM
Reference case: Low demand NEM
Reference case Other grids Black coal Brown coal Peaking Gas Baseload Gas
Reference case: Low demand Other grids Hydro Wind Solar Other
79
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
NPV of change in generator EBITDA (7%) NPV of change in residential retail bills
NPV ($)
Hydro -1.4
$400
Note: All NPV values calculated using a 7% real discount rate over the period 2015-40
Source: ACIL Allen
80
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
Reference case: Low demand NEM
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Mt CO2-e
15.0
Total Large-scale -11.8
10.0
Unserved Energy 0.0
-120%-100% -80% -60% -40% -20% 0% -120% -100% -80% -60% -40% -20% 0%
Percentage change Percentage change
$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)
($1,000)
-20 -15 -10 -5 0 5 10 2015-20 2015-30 2015-40
NPV ($ 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
82
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 83 Sensitivity results: Real 20% case: Low demand (change from Reference case: Low demand)
70
Real 2014 $/MWh
10,000
60
GWh sent-out
5,000
50
40 0
30 -5,000
20
-10,000
10
0 -15,000
-20,000
Reference case: Low demand NEM
Real 20% case: Low demand NEM
Reference case: Low demand Other grids Black coal Brown coal Peaking Gas Baseload Gas
Real 20% case: Low demand Other grids Hydro Wind Solar Other
Refurbishment 0.1
-5.0
New build -8.0
-15 -10 -5 0 5
Black coal Brown coal Peaking Gas Baseload Gas Other
NPV ($ billion)
83
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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)
Note: All NPV values calculated using a 7% real discount rate over the period 2015-40
Source: ACIL Allen
Figure 84 Sensitivity results: Real 30% case: Low demand (change from Reference case: Low demand)
100
Wholesale price outcomes Change in dispatch
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
Reference case: Low demand NEM
Real 30% case: Low demand NEM
Reference case: Low demand Other grids Black coal Brown coal Peaking Gas Baseload Gas
Real 30% case: Low demand Other grids Hydro Wind Solar Other
Refurbishment -0.2
-15.0
New build 0.7
84
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
-10% 0% 10% 20% 30% 40% 50% -2.5% -2.0% -1.5% -1.0% -0.5% 0.0%
Percentage change Percentage change
Geothermal 0.0 $0
Solar 0.6
($50) ($37)
Wind -2.0
($76) ($75)
($100)
NPV ($)
($211)
Brown coal -0.9
($250)
Black coal -1.7
($267)
($300) ($285)
-3 -2 -2 -1 -1 0 1 1 2015-20 2015-30 2015-40
NPV ($ 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
Figure 85 Sensitivity results: 50% Growth case: Low demand (change from Reference case: Low demand)
100
Wholesale price outcomes Change in dispatch
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
Reference case: Low demand NEM
50% Growth case: Low demand NEM
Reference case: Low demand Other grids Black coal Brown coal Peaking Gas Baseload Gas
50% Growth case: Low demand Other grids Hydro Wind Solar Other
85
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Mt CO2-e
15.0
Total Large-scale -12.0
10.0
Unserved Energy 0.0
Hydro 0.2
$0
($800)
-20 -15 -10 -5 0 5 10 2015-20 2015-30 2015-40
NPV ($ 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
86
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
350 379.9
300 15.6%
Sent-out TWh
Sent-out TWh
87
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Wholesale price outcomes Change in dispatch
80,000
90
70,000
80
70 60,000
Real 2014 $/MWh
60 50,000
GWh sent-out
50 40,000
40
30,000
30
20,000
20
10 10,000
0 0
-10,000
Reference case NEM
Reference case: High demand NEM
Reference case Other grids Black coal Brown coal Peaking Gas Baseload Gas
Reference case: High demand Other grids Hydro Wind Solar Other
88
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Mt CO2-e
Total Large-scale 15.2 30.0
Refurbishment 1.0
0.0
New build 3.5
0 5 10 15 20
Black coal Brown coal Peaking Gas Baseload Gas Other
NPV ($ billion)
-16%-14%-12%-10%-8% -6% -4% -2% 0% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0%
Percentage change Percentage change
Hydro 2.7
$400
Baseload Gas 1.5
Note: All NPV values calculated using a 7% real discount rate over the period 2015-40
Source: ACIL Allen
89
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 88 Sensitivity results: Repeal case: High demand (change from Reference case: High demand)
20,000
60
GWh sent-out
50 10,000
40
0
30
20 -10,000
10
0 -20,000
-30,000
Reference case: High demand NEM
Repeal case: High demand NEM
Reference case: High demand Other grids Black coal Brown coal Peaking Gas Baseload Gas
Repeal case: High demand Other grids Hydro Wind Solar Other
90
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Mt CO2-e
Total Large-scale -8.8 10.0
Refurbishment 0.0
-5.0
New build -9.8
-15 -10 -5 0 5 10
Black coal Brown coal Peaking Gas Baseload Gas Other
NPV ($ billion)
-105%-100% -95% -90% -85% -80% -75% -120% -100% -80% -60% -40% -20% 0%
Percentage change Percentage change
($579) ($579)
Black coal 19.5
($1,000)
-20 -15 -10 -5 0 5 10 15 20 25 2015-20 2015-30 2015-40
NPV ($ 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
91
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 89 Sensitivity results: 50% Growth case: High demand (change from Reference case: High demand)
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
Reference case: High demand NEM
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
4.0
Total Large-scale -2.5
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
Black coal Brown coal Peaking Gas Baseload Gas Other
NPV ($ billion)
92
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
NPV ($)
Hydro 0.6 $192
$200
Note: All NPV values calculated using a 7% real discount rate over the period 2015-40
Source: ACIL Allen
93
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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).
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
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 ($9.50/tonne
in Real 2014 dollars). Prices in subsequent years increase by approximately 5.5% nominal
(3% real).
94
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
35
30
25
A$/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 $
95
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 93 Sensitivity results: Reference case: Carbon (change from Reference case)
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 Other grids Black coal Brown coal Peaking Gas Baseload Gas
Reference case: Carbon Other grids Hydro Wind Solar Other
-1.0
Total Large-scale 0.4
-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
96
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Solar 0.0
$800
Wind 1.2
$600 $515
NPV ($)
Hydro 1.0 $468
$400
Baseload Gas 0.2
Brown coal-1.8 $7
$0
Black coal -0.6 ($14) ($7)
($47) ($47)
($200)
-2 -2 -1 -1 0 1 1 2 2015-20 2015-30 2015-40
NPV ($ 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
25 A small change to the up-take of small scale technologies is observed but considered immaterial to wholesale market
outcomes.
97
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 94 Sensitivity results: Repeal case: Carbon (change from Reference case: Carbon)
100 30,000
Real 2014 $/MWh
80 20,000
GWh sent-out
60 10,000
40 0
20 -10,000
0 -20,000
-30,000
Reference case: Carbon NEM
Repeal case: Carbon NEM
Reference case: Carbon Other grids Black coal Brown coal Peaking Gas Baseload Gas
Repeal case: Carbon Other grids Hydro Wind Solar Other
15.0
Total Large-scale -11.3
10.0
Unserved Energy 0.0
-105% -100% -95% -90% -85% -80% -120% -100% -80% -60% -40% -20% 0%
Percentage change Percentage change
98
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
NPV ($)
Hydro 1.2 $152 $186
$200 $73
Baseload Gas 1.3
$0
Note: All NPV values calculated using a 7% real discount rate over the period 2015-40
Source: ACIL Allen
3,000
Reference High
2,396
2,500 2,284 2,211 2,237
2,164
Real 2014 A$/kW installed
1,500
1,000
500
0
2014 2020 2030 2040
Calendar year
99
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Solar
As of March 2014 the national average cost of a residential solar installation is estimated to
be $2,560/kW excluding GST and excluding the STC rebate. The cost of solar installations
consists of a local technology component, a foreign technology component and a local
labour component. For future cost projections each of these components is indexed by the
projected cost reduction factors sourced from the 2013 Australian Energy Technology
Assessment (AETA). While the cost indices of foreign and local equipment are assumed to
decline, the local labour cost component is assumed to increase in these projections as
wages are projected to increase in real terms. The Reference case capital costs are
provided in Figure 96 for residential and commercial rooftop systems as well as for fixed flat
plate utility-scale solar PV.
It should be noted that utility scale PV attracts a higher capital cost initially due to increased
costs associated with transformer and network connection, which offset the economies of
scale.26 This capital cost differential is projected to reduce over time due to the higher
proportion of imported materials in the overall build relative to small-scale installations which
conversely have a high domestic labour component.
Other solar technologies such as solar thermal (parabolic trough, central receiver and
compact linear Fresnel) follow a similar decline path but are omitted from the chart for
clarity.
3,000
2,524 2,602 Residential roof top solar
Real 2014 A$/kW installed
500
0
2014 2020 2030 2040
Calendar year
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
1,000
500
0
2014 2020 2030 2040
Calendar year
101
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 98 Sensitivity results: Reference case: High capital cost (change from Reference case)
100 4,000
3,000
Real 2014 $/MWh
80
2,000
GWh sent-out
60 1,000
0
40
-1,000
-2,000
20
-3,000
0 -4,000
-5,000
Reference case NEM
Reference case: High capital costs NEM
Reference case Other grids Black coal Brown coal Peaking Gas Baseload Gas
Reference case: High capital costs Other grids Hydro Wind Solar Other
Refurbishment -0.1
-1.0
New build 2.1
-1 0 1 2 3
Black coal Brown coal Peaking Gas Baseload Gas Other
NPV ($ billion)
-40% -30% -20% -10% 0% 10% 20% -20% -15% -10% -5% 0%
Percentage change Percentage change
102
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
$0
-0.5 0.0 0.5 1.0 1.5 2.0 2015-20 2015-30 2015-40
NPV ($ 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
27 A small change to the up-take of small scale technologies is observed but considered immaterial to wholesale market
outcomes.
103
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 99 Sensitivity results: Repeal case: High capital cost (change from Reference case: High capital
cost)
120
Wholesale price outcomes Change in dispatch
40,000
100
30,000
Real 2014 $/MWh
80 20,000
GWh sent-out
60 10,000
40 0
20 -10,000
0 -20,000
-30,000
Reference case: High capital costs NEM
Repeal case: High capital NEM
Reference case: High capital costs Other grids Black coal Brown coal Peaking Gas Baseload Gas
Repeal case: High capital Other grids Hydro Wind Solar Other
15.0
Total Large-scale -12.8
10.0
Unserved Energy 0.0
-120%-100% -80% -60% -40% -20% 0% -120% -100% -80% -60% -40% -20% 0%
Percentage change Percentage change
104
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
NPV ($)
Hydro 0.8 $200 $101
$26
$0
Baseload Gas 1.3
($200) ($81)
Peaking Gas 0.1
($400) ($285)
($386)
Brown coal 3.8 ($600)
Note: All NPV values calculated using a 7% real discount rate over the period 2015-40
Source: ACIL Allen
Figure 100 Sensitivity results: 50% Growth case: High capital cost (change from Reference case: High
capital cost)
120
Wholesale price outcomes Change in dispatch
20,000
100 15,000
Real 2014 $/MWh
80 10,000
GWh sent-out
5,000
60
0
40
-5,000
20 -10,000
0 -15,000
-20,000
Reference case: High capital costs NEM
50% Growth case: High capital NEM
Reference case: High capital costs Other grids Black coal Brown coal Peaking Gas Baseload Gas
50% Growth case: High capital Other grids Hydro Wind Solar Other
8.0
Total Large-scale -8.8
6.0
Unserved Energy 0.0 4.0
Variable O&M 2.3 2.0
-15 -10 -5 0 5
Black coal Brown coal Peaking Gas Baseload Gas Other
NPV ($ billion)
105
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
-70%-60%-50%-40%-30%-20%-10% 0% -80%-70%-60%-50%-40%-30%-20%-10% 0%
Percentage change Percentage change
($200) ($135)
Brown coal 2.1
($300)
Black coal 6.1
($317) ($317)
($400)
-8 -6 -4 -2 0 2 4 6 8 2015-20 2015-30 2015-40
NPV ($ 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
106
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 101 Sensitivity results: Reference case: Permanent retirements (change from Reference case)
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 Other grids Black coal Brown coal Peaking Gas Baseload Gas
Reference case: Permanent retirements Other grids Hydro Wind Solar Other
-2 -1 0 1 2
Black coal Brown coal Peaking Gas Baseload Gas Other
NPV ($ billion)
108
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
NPV ($)
Hydro 1.8 $500
$400
Baseload Gas 0.7
$300
Peaking Gas 0.5
$200 $152 $140
Note: All NPV values calculated using a 7% real discount rate over the period 2015-40
Source: ACIL Allen
109
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 102 Sensitivity results: Repeal case (change from Reference case: Permanent retirements)
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
Reference case: Permanent retirements NEM
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
-120%-100% -80% -60% -40% -20% 0% -120% -100% -80% -60% -40% -20% 0%
Percentage change Percentage change
110
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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)
Black coal 4.6 ($585)
($631) ($631)
($700) ($653)
-20 -15 -10 -5 0 5 10 2015-20 2015-30 2015-40
NPV ($ 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
Figure 103 Sensitivity results: 50% Growth case: Permanent retirements (change from Reference case:
Permanent retirements)
70
Real 2014 $/MWh
10,000
60
GWh sent-out
5,000
50
40 0
30 -5,000
20
-10,000
10
0 -15,000
-20,000
Reference case: Permanent retirements NEM
50% Growth case: Permanent retirements NEM
Reference case: Permanent retirements Other grids Black coal Brown coal Peaking Gas Baseload Gas
50% Growth case: Permanent retirements Other grids Hydro Wind Solar Other
10.0
Total Large-scale -8.2
8.0
Unserved Energy 0.0 6.0
Variable O&M 1.2 4.0
-15 -10 -5 0 5
Black coal Brown coal Peaking Gas Baseload Gas Other
NPV ($ billion)
111
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
-70%-60%-50%-40%-30%-20%-10% 0% -80%-70%-60%-50%-40%-30%-20%-10% 0%
Percentage change Percentage change
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)
Brown coal 0.5 ($243)
($263) ($263)
($300)
Black coal 2.0
($308)
($350)
-8 -6 -4 -2 0 2 4 2015-20 2015-30 2015-40
NPV ($ 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
112
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
9 Abatement costs
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
$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
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
Reference case: Carbon 10,668 294 $36 9,279 280 $33 1,389 15 $95
Reference case: High capital costs 11,902 299 $40 10,514 285 $37 1,388 13 $104
50% Growth case: High capital 4,025 115 $35 3,338 108 $31 687 7 $103
Reference case: High demand 8,708 250 $35 7,544 238 $32 1,164 12 $101
50% Growth case: High demand 5,951 174 $34 5,492 171 $32 460 4 $131
50% Growth case: Low Demand 640 7 $90 0 0 n/a 640 7 $90
Reference case: Low demand 10,932 302 $36 9,199 284 $32 1,733 18 $96
Real 20% case: Low demand 3,972 106 $38 3,066 96 $32 906 10 $94
Real 30% case: Low demand 9,821 279 $35 8,237 262 $31 1,584 16 $97
Reference case: Perm retirements 11,010 346 $32 9,229 326 $28 1,781 20 $89
50% Growth case: Perm retirements 3,552 142 $25 2,727 133 $20 825 9 $90
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
116
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Figure 105 Calculated abatement costs for the RET to 2030: Method 2 (abatement stream undiscounted)
$140 131
RET LRET Solar PV
$120
Abatement cost $/tonne CO2-e
$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
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Table 11 Calculated abatement costs for the RET to 2040: Method 1 (abatement stream discounted)
RET LRET Solar PV
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
Reference case: Carbon 13,127 205 $64 11,271 195 $58 1,856 10 $187
Reference case: High capital costs 14,711 208 $71 12,831 198 $65 1,880 9 $204
50% Growth case: High capital 4,948 81 $61 4,081 77 $53 868 4 $193
Reference case: High demand 10,264 174 $59 8,824 166 $53 1,440 7 $192
50% Growth case: High demand 6,814 121 $56 6,368 118 $54 447 2 $185
50% Growth case: Low Demand 866 5 $172 0 0 n/a 866 5 $172
Reference case: Low demand 14,217 210 $68 11,798 197 $60 2,419 12 $195
Real 20% case: Low demand 4,901 74 $66 3,714 68 $55 1,187 6 $184
Real 30% case: Low demand 14,716 207 $71 12,545 196 $64 2,171 11 $198
Reference case: Perm retirements 14,191 243 $58 11,572 228 $51 2,618 15 $180
50% Growth case: Perm retirements 4,501 100 $45 3,367 93 $36 1,134 7 $174
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
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
Abatement cost $/tonne CO2-e
$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
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 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Table 12 Calculated abatement costs for the RET to 2040: Method 2 (abatement stream undiscounted)
RET LRET Solar PV
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
Reference case: Carbon 13,127 515 $25 11,271 492 $23 1,856 23 $81
Reference case: High capital costs 14,711 524 $28 12,831 503 $26 1,880 21 $88
50% Growth case: High capital 4,948 209 $24 4,081 199 $20 868 9 $92
Reference case: High demand 10,264 439 $23 8,824 424 $21 1,440 15 $95
50% Growth case: High demand 6,814 303 $22 6,368 299 $21 447 4 $127
50% Growth case: Low Demand 866 12 $73 0 0 n/a 866 12 $73
Reference case: Low demand 14,217 510 $28 11,798 480 $25 2,419 30 $80
Real 20% case: Low demand 4,901 193 $25 3,714 178 $21 1,187 14 $83
Real 30% case: Low demand 14,716 580 $25 12,545 554 $23 2,171 25 $85
Reference case: Perm retirements 14,191 620 $23 11,572 582 $20 2,618 38 $69
50% Growth case: Perm retirements 4,501 255 $18 3,367 239 $14 1,134 16 $73
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
Figure 107 Calculated abatement costs for the RET to 2040: Method 2 (abatement stream undiscounted)
$140
127
RET LRET Solar PV
$120
Abatement cost $/tonne CO2-e
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
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
119
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Closed to New Entrants
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
121
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
Closed to New Entrants
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: 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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
123
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Closed to New Entrants
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: 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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Core Carbon High capital costs High demand Low demand Permanent
retirements
125
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Core Carbon High capital costs High demand Low demand Permanent
retirements
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
Closed to New Entrants
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Core Carbon High capital costs High demand Low demand Permanent
retirements
Note: Sum of projected LGC price multiplied by LRET targets and projected STC creation by assumed STC price
Source: ACIL Allen
127
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
Closed to New Entrants
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: NPV of annual residential bills for average household over the period 2015-40. Uses a 7% real discount rate
Source: ACIL Allen
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
Distribution
Median
Number of observations
Mean
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
130
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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.
A4
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
Note: Mandated demand includes adjustment under Section 40 (1A) of the Act
Source: ACIL Allen analysis
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Certificate supply
A6
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
50,000
45,000
40,000
35,000
LGCs ('000)
30,000
25,000
20,000
15,000
10,000
5,000
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
Source: ACIL Allen analysis
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
A7
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
A10
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
LGCs ('000)
1,000
1,500
800
600 1,000
400 324 535
500
200
0 0
500
800
LGCs ('000)
LGCs ('000)
400
600
300 256
400
200
100 200
0 0
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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%
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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.
A14
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
A15
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
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
A16
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Convergence Scenario in which global prices converged on the basis of increased exports
from the United States. These projections are shown in Figure A14.
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
A17
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
3.50
3.00
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
A18
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
A19
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Technical
Technical Capacity
Region Generator Plant type Fuel type Commissioned Retirement
Life (Years) (gross MW)
Year
A20
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Technical
Technical Capacity
Region Generator Plant type Fuel type Commissioned Retirement
Life (Years) (gross MW)
Year
A21
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Technical
Technical Capacity
Region Generator Plant type Fuel type Commissioned Retirement
Life (Years) (gross MW)
Year
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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 ($/MWh outage rate outage rate MLF
HHV (%) (%) (tonnes
CO2-e/GJ of gross/year) sent-out) (%) (%)
sent-out CO2-e/MWh
fuel)
sent-out)
A23
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Scope 1
Scope 1
Thermal emission
emission FOM VOM Forced Planned
efficiency Auxiliaries intensity
Region Generator factor (kg ($/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)
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Scope 1
Scope 1
Thermal emission
emission FOM VOM Forced Planned
efficiency Auxiliaries intensity
Region Generator factor (kg ($/MW ($/MWh outage rate outage rate MLF
HHV (%) (%) (tonnes
CO2-e/GJ of gross/year) sent-out) (%) (%)
sent-out CO2-e/MWh
fuel)
sent-out)
A25
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
Scope 1
Scope 1
Thermal emission
emission FOM VOM Forced Planned
efficiency Auxiliaries intensity
Region Generator factor (kg ($/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)
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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
AC I L AL L E N C O N S U L T IN G
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).
12,000
Adopted for study AETA 2012
Real 2011-12 A$/kW Installed
10,000
8,000
6,000
4,000
2,000
A31
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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
A-33
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
Figure A18 Final capital costs for new entrant technologies for selected years
CCGT
OCGT
CLFR
Parabolic trough
Central Receiver
Solar PV fixed
Ocean/Wave
Geothermal HSA
Geothermal EGS
A-35
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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-36
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
A-37
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
A-38
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
Refurbishment Additional life Additional life
Economic life
Category Technology cost (% of (% of original from refurb
(years)
new) life) (years)
A-39
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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-40
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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-41
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
$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
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
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
The premia/discounts associated with installations in different states and territories are set
out in Table A15.
A-43
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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-44
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
$50
nominal $/certificate
$40
$30
$20
$10
$0
2009 2010 2011 2012 2013
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-45
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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-46
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
Source: ACIL Allen analysis
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-47
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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
South
0.0% 1.3% 95.6% 3.1% 1.378
Australia
Western
1.9% 5.5% 89.5% 3.1% 1.389
Australia
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.
A-48
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
Table A21 Building footprints as of 2011
State/territory Residential CBD Non residential Non CBD Non Total
Residential
TAS 30 0.15 3 33
NT 10 0 0.9 11
ACT 20 0 1.6 21
Two studies by Entura on the ratio between floor and roof space were presented by BZE
and reproduced in Table A22.
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
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
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-51
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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-52
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
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
Residential Commercial Industrial EITE
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
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
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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.
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-55
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS
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-56
RET REVIEW MODELLING MARKET MODELLING OF VARIOUS RET POLICY OPTIONS