consultants report_acil allen_with alt text · web viewprojected demand growth under the scenario...

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

REPORT TODEPARTMENT OF THE ENVIRONMENT

APRIL 2015

ELECTRICITY SECTOR EMISSIONSMODELLING OF THE AUSTRALIAN GENERATION SECTOR

For information on this report please contact:Owen KelpPrincipalTelephone (07) 3009 8711Email [email protected]

ACIL ALLEN CONSULTING PTY LTDABN 68 102 652 148

161 WAKEFIELD STREETADELAIDE SA 5000AUSTRALIAT +61 0412 089 043

LEVEL FIFTEEN127 CREEK STREETBRISBANE QLD 4000AUSTRALIAT+61 7 3009 8700F+61 7 3009 8799

LEVEL TWO33 AINSLIE PLACECANBERRA ACT 2600AUSTRALIAT+61 2 6103 8200F+61 2 6103 8233

LEVEL NINE60 COLLINS STREETMELBOURNE VIC 3000AUSTRALIAT+61 3 8650 6000F+61 3 9654 6363

LEVEL ONE50 PITT STREETSYDNEY NSW 2000AUSTRALIAT+61 2 8272 5100F+61 2 9247 2455

LEVEL TWELVE, BGC CENTRE28 THE ESPLANADEPERTH WA 6000AUSTRALIAT+61 8 9449 9600F+61 8 9322 3955

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 THE REPORT MUST NOT BE PUBLISHED, QUOTED OR DISSEMINATED TO ANY OTHER PARTY WITHOUT ACIL ALLEN CONSULTING’S PRIOR WRITTEN CONSENT.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

© ACIL ALLEN CONSULTING 2015

mailto:[email protected]


Executive summary

The Department of the Environment (the Department) commissioned ACIL Allen Consulting (ACIL Allen) to undertake projections of greenhouse gas emissions from Australia’s electricity generation sector over the period 2008-09 to 2034-35. This assignment is intended to inform emissions projections and support the annual publication of greenhouse gas emissions projections undertaken by the Department in tracking Australia’s progress towards its 2020 emissions reduction target.

The electricity sector currently accounts for around one third of Australia’s emissions as reported under the National Greenhouse Gas Inventory and is the largest single source.

The projections are intended to cover all electricity generation facilities – including those previously deemed to be part of vertically integrated processes such as mining/minerals processing and LNG plants. As such, emission values will be higher than those reported from the 2013 emission projections study as they include some emissions which were previously covered under ‘Direct Combustion’ activities.

Greenhouse gas Global Warming Potentials (GWPs) for carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) have been taken from the Intergovernmental Panel on Climate Change Fourth Assessment Report (AR4) in accordance with the current reporting convention within the National Greenhouse Gas Inventory.

Modelling has been undertaken for two scenarios:

Baseline scenario: Current electricity demand and supply side measures continue for the length of the projection (or until the planned end date of a particular program). The scenario includes modifications to the current Large-scale Renewable Energy Target (LRET) scheme so that mandated GWh targets in 2020 represent an estimated 20 per cent of Australian electricity consumption. This was estimated to be 27,000 GWh. All defined Energy Intensive Trade Exposed (EITE) activities are 100 per cent shielded from RET impacts. The scenario does not include any changes to the Small-scale Renewable Energy Scheme (SRES). New conventional coal-fired developments (plants which do not employ carbon capture and storage) are constrained from entering in this scenario.

No supply-side measures: Current electricity demand side measures remain in place for the length of the projection (or until the planned end date of a particular program). Electricity supply side measures cease from 1 July 2014. Under this scenario, the only supply-side measures assumed to cease was the RET scheme and the assumption of restricting conventional coal-fired new entry is relaxed.

Figure ES 1 shows historical and projected emissions for the Australian generation sector over the period 2008-09 to 2034-35 under the two scenarios modelled. Emissions have fallen over the last 6 years from 211 Mt CO2-e in 2008-09 to a low of around 180 Mt CO2-e in 2013-14 (a 15% decline). This has been due to a combination of factors including the decline in electricity demand, the introduction of renewable generation and carbon pricing. In addition, generation from hydro power stations was well above normal levels during the carbon tax period resulting in a temporary lowering of emissions as hydro stations displaced fossil fuelled stations and drew down water storages.

Under the Baseline scenario, emissions are projected to rise in the near-term, reaching 201 Mt CO2-e by 2019-20. This represents a 7.7% increase over 2012-13 levels (equivalent

ELECTRICITY SECTOR EMISSIONSELECTRICITY SECTOR EMISSIONS MODELLING OF THE AUSTRALIAN GENERATION SECTORMODELLING OF THE AUSTRALIAN GENERATION SECTOR

ii


to average annual growth rate of 1.1% per annum). The projected increase is due to a number of factors:

The repeal of the carbon tax effective from 1 July 2014 results in a shift back to coal-fired generation and low water storages results in lower than normal hydro output over coming years as hydro storage levels are restored

This is further exacerbated by the decrease in gas-fired generation owing to wholesale gas price increases (which makes gas-fired generation less competitive) and oversupplied wholesale markets (which results in limited requirement for high-cost peaking plant operation)

Projected demand growth under the scenario which results in some mothballed coal-fired generators returning to service over time as market conditions improve.

Development of renewable capacity occurs over the period to 2020 in order to meet the ‘Real 20%’ LRET and ongoing installation of rooftop solar PV occurs with support under the SRES. The introduction of this renewable capacity tends to dampen the growth in emissions from the generation sector, particularly from 2017-18 to 2020-21 when much of the large scale capacity development occurs. Through this period, emissions remain relatively flat despite the growth in electricity demand.

Emissions are projected to continue to increase over the long-term from the sector. Over the period 2019-20 to 2035-35, projected annual emissions increase by 17.3% to 236 Mt CO2-e (equivalent to average annual growth of 1.1% per annum). The modelling finds renewable technologies are generally not commercially competitive in the wholesale market without additional subsidies (beyond the current RET) and new centralised generation is primarily gas-fired. Development of decentralised solar PV continues throughout the projection period.

Figure ES 1 Projected emissions: Baseline and No supply-side scenarios

Source: ACIL Allen

In the ‘No supply-side measures’ scenario, total annual emissions increase from 187 Mt CO2-e in 2012-13 to 214 Mt CO2-e by 2019-20 (a 14.4% increase, equivalent to annual growth of 1.9% per annum). In aggregate, emissions over this period are 1,583 Mt CO2-e which is around 27.2 Mt CO2-e (1.7%) higher than the Baseline scenario. This is due to the lower level of renewables developed and increased utilisation of coal-fired capacity over this period.


iii


By 2034-35, projected emissions from the generation sector reach 248 Mt CO2-e, a 32.4% increase over 2012-13 levels. In the ‘No supply-side measures’ scenario, entry of conventional coal is not restricted and the modelling projects a small amount of new conventional coal would be developed in the SWIS from around 2020 and in Queensland from 2025 onwards.

In the absence of the supply-side measures projected installed capacity of renewables is considerably lower (3,300 MW less wind, 1,800 MW less solar PV by 2025). This is offset by some increase in fossil fuel capacity, most of which is associated with the earlier return to service of mothballed plant and the development of some new coal-fired capacity in the SWIS and Queensland toward the latter part of the projection period.

Aggregate emissions over the projection period (2014-15 to 2034-35) are 201.4 Mt CO2-e (4.5%) higher in the absence of the supply side measures.

A range of summary figures for emissions under the two scenarios are presented in Table ES 1 and Table ES 2.

Table ES 1Summary of emission outcomes by fuel type (Mt CO2-e): Baseline scenario

Annual emissions for selected years Change in emissions Cumulative emissions

2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 117.2 140.9 -16.3% 18.7% 20.3% 855.6 1,892.5

Brown coal 69.0 59.6 61.1 65.4 -13.7% 2.5% 7.1% 506.4 995.2

Peaking Gas 5.5 5.0 1.0 1.6 -9.6% -78.9% 53.0% 17.0 15.3

Baseload Gas 13.6 17.4 17.9 24.4 28.3% 2.5% 36.4% 134.9 320.0

Cogen 3.3 3.9 1.9 1.4 18.9% -51.9% -27.8% 23.1 23.4

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -4.1% 9.1% 17.4 35.3

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 201.4 236.2 -11.4% 7.7% 17.3% 1,555.5 3,283.1

Source: ACIL Allen

Table ES 2Summary of emission outcomes by fuel type (Mt CO2-e): No supply-side measures


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 119.5 154.6 -16.3% 21.1% 29.3% 854.9 2,021.0

Brown coal 69.0 59.6 69.7 65.4 -13.7% 16.9% -6.1% 531.7 1,035.5

Peaking Gas 5.5 5.0 1.5 1.4 -9.6% -69.0% -5.7% 18.2 18.0

Baseload Gas 13.6 17.4 19.0 22.3 28.3% 8.8% 17.6% 136.3 322.1

Cogen 3.3 3.9 2.0 1.3 18.9% -50.1% -33.4% 23.3 23.7

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -3.9% 9.1% 17.4 35.4

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 214.0 247.6 -11.4% 14.4% 15.7% 1,582.7 3,457.3

Source: ACIL Allen


iv


Figure ES 2 presents projected emission intensity trends (on a ‘sent out’ basis) for the two scenarios. The Baseline scenario see emissions intensity increase sharply from an estimated 0.77 tonnes CO2-e/MWh sent-out in 2013-14 up to 0.81 tonnes CO2-e/MWh sent-out by 2015-16, before falling over the remainder of the period.

In the absence of supply-side measures, the Australian average intensity increases to 0.82 tonnes CO2-e/MWh sent-out by 2019-20 and 0.78 tonnes CO2-e/MWh sent-out by 2034-35. The impact of the modelled supply-side measures is to reduce the emissions intensity of the generation sector by around 0.04 tonnes CO2-e/MWh.

Figure ES 2 Projected emission intensity Baseline and ‘No supply-side measures’ scenarios

Source: ACIL Allen

A number of sensitivities were run against the Baseline scenario to test key drivers and assess the potential range for emission outcomes. Sensitivities included:

High/Low electricity demand

High/Low fuel costs

Uptake of electric vehicles (increased demand)

Aggregate High/Low sensitivities which included combinations of the above.

Figure ES 3 summarises the emission outcomes for each of these sensitivities. The largest determinant of emission was found to be the demand assumptions used, whereas fuel price changes had little effect. The projected bounds for annual emission outcomes in 2019-20 were 3.3% higher under the Aggregate High sensitivity and 3.3% lower under the Aggregate Low sensitivity. The variation grows over time reaching +8.9% and -9.7% relative to the Baseline scenario by 2034-35 under the respective sensitivities as detailed in Table ES 3.


v


Figure ES 3 Emission outcomes from Baseline sensitivities

Source: ACIL Allen

Table ES 3 Summary of emission outcomes from Baseline sensitivities

Annual emissions in 2019-20Cumulative emissions(2012-13 to 2019-20)

Annual emissions in 2034-35Cumulative emissions(2012-13 to 2034-35)

Mt CO2-eChange

(Mt)

Change

(%)Mt CO2-e

Change

(Mt)

Change

(%)Mt CO2-e

Change

(Mt)

Change

(%)Mt CO2-e

Change

(Mt)

Change

(%)

Baseline 201.4 1,555.5 236.2 4,838.6

High demand 208.1 6.7 3.3% 1,577.6 22.1 1.4% 252.1 15.9 6.7% 5,063.4 224.8 4.6%

Low demand 194.6 -6.7 -3.3% 1,534.8 -20.7 -1.3% 213.9 -22.3 -9.4% 4,593.6 -245.1 -5.1%

High fuel 201.9 0.5 0.3% 1,557.2 1.7 0.1% 235.8 -0.5 -0.2% 4,846.4 7.8 0.2%

Low fuel 200.8 -0.5 -0.3% 1,556.2 0.7 0.0% 237.0 0.8 0.3% 4,835.4 -3.3 -0.1%

EV uptake 201.8 0.4 0.2% 1,556.7 1.2 0.1% 242.5 6.3 2.7% 4,883.0 44.4 0.9%

Aggregate High 208.0 6.7 3.3% 1,579.4 23.9 1.5% 257.3 21.1 8.9% 5,102.3 263.6 5.4%

Aggregate Low 194.7 -6.7 -3.3% 1,534.5 -20.9 -1.3% 213.4 -22.8 -9.7% 4,595.0 -243.6 -5.0%

Source: ACIL Allen


vi


C o n t e n t sExecutive summary ii

1 Introduction 3

2 Project overview 5

2.1 Modelling process 5

2.1.1 PowerMark LT 5

2.1.2 LRET implementation in PowerMark LT 7

2.1.3 Retail price model 8

2.1.4 SRES projection model 8

2.1.5 Off-grid and embedded generation 8

2.2 Scenarios 9

2.2.1 Baseline scenario 9

2.2.2 No supply-side measures scenario 10

2.3 Sensitivities 11

2.4 Historical calibration with National Greenhouse Gas Inventory 11

3 Scenario results 14

3.1 Baseline scenario 14

3.2 No supply-side measures 20

4 Sensitivity results 26

4.1 High/Low demand 26

4.2 High/Low fuel prices 30

4.3 High uptake of electric vehicles 33

4.4 Aggregate High/Low scenarios 35

Appendix A Input assumptions 1

List of figuresFigure ES 1 Projected emissions: Baseline and No supply-side scenarios iii

Figure ES 2 Projected emission intensity Baseline and ‘No supply-side measures’ scenarios v

Figure ES 3 Emission outcomes from Baseline sensitivities vi

vii


Figure 1 Modelling framework used 5

Figure 2 Modelled electricity grids 7

Figure 3 Model calibration against actual National Greenhouse Gas Inventory values 12

Figure 4 Comparison with National Greenhouse Gas Inventory by year by State (Mt CO2-e) 13

Figure 5 Australian generation by fuel type: Baseline scenario 14

Figure 6 Australian emissions by fuel type: Baseline scenario 15

Figure 7 Generation by fuel type by grid: Baseline scenario 17

Figure 8 Generation by fuel type by jurisdiction: Baseline scenario 18

Figure 9 Projected emission intensity: Baseline scenario 19

Figure 10 Projected generation capacity: Baseline scenario 19

Figure 11 Australian generation by fuel type: No supply-side measures 20

Figure 12 Australian generation by fuel type: No supply-side measures change from Baseline scenario 20

Figure 13 Australian emissions by fuel type: No supply-side measures 21

Figure 14 Impact of supply-side measures on emissions 22

Figure 15 Australian emissions by fuel type: No supply-side measures change from Baseline scenario 22

Figure 16 Generation by fuel type by grid: No supply-side measures 23

Figure 17 Generation by fuel type by jurisdiction: No supply-side measures 24

Figure 18 Projected emission intensity: No supply-side measures 25

Figure 19 Projected generation capacity: No supply-side measures change from Baseline scenario 25

Figure 20 Demand sensitivity inputs: Australia 26

Figure 21 Demand sensitivity aggregate emission outcomes: Australia 28

Figure 22 Australian emissions by fuel type: High demand sensitivity change from Baseline scenario 29

Figure 23 Australian emissions by fuel type: Low demand sensitivity change from Baseline scenario 30

Figure 24 Fuel cost variations: International fuel price markers 31

Figure 25 Fuel cost sensitivity aggregate emission outcomes: Australia 31

Figure 26 Australian emissions by fuel type: High fuel cost sensitivity change from Baseline scenario 32

Figure 27 Australian emissions by fuel type: Low fuel cost sensitivity change from Baseline scenario 33

Figure 28 Incremental demand for EV sensitivity 34

Figure 29 Australian emissions by fuel type: EV sensitivity change from Baseline scenario 35

Figure 30 Aggregate High/Low sensitivity emission outcomes: Australia 35

Figure 31 Australian emissions by fuel type: Aggregate High sensitivity change from Baseline scenario 37

Figure 32 Australian emissions by fuel type: Aggregate Low sensitivity change from Baseline scenario 38

Figure A1 Comparison of Australian energy demand to 2020 3

Figure A3 Real wage index used 3

viii


Figure A4 Calculated netback gas price for Australian gas producers 4

Figure A5 Assumed new entrant coal costs 4

Figure A6 Base capital cost comparison with AETA 18

Figure A7 Final capital costs for new entrant technologies for selected years A-3

Figure A8 National average historic PV installation cost (2011$/kW) A-10

Figure A9 National average projected PV installation cost A-11

Figure A10 REC/STC prices (nominal $/certificate) A-13

Figure A11 Share of installed capacity by system size class A-18

Figure A12 Energy impact of electric vehicles by State (medium scenario) A-25

Figure A13 Aggregate energy impact of electric vehicles across scenarios A-26

List of tablesTable ES 1 Summary of emission outcomes by fuel type (Mt CO2-e): Baseline

scenario iv

Table ES 2 Summary of emission outcomes by fuel type (Mt CO2-e): No supply-side measures iv

Table ES 3 Summary of emission outcomes from Baseline sensitivities vi

Table 1 Real 20% by 2020 annual LRET targets 10

Table 2 Australian generation by fuel type (sent-out TWh): Baseline scenario 15

Table 3 Summary of emission outcomes by fuel type (Mt CO2-e): Baseline scenario 16

Table 4 Australian generation by fuel type (sent-out TWh): No supply-side measures 21

Table 5 Summary of emission outcomes by fuel type (Mt CO2-e): No supply-side measures 23

Table 6 Demand sensitivity inputs for selected years and change from baseline: Australia 26

Table 7 Electricity demand projection for selected years by grid (GWh): High demand sensitivity 27

Table 8 Electricity demand projection for selected years by grid (GWh): Low demand sensitivity 27

Table 7 Summary of emission outcomes by fuel type (Mt CO2-e): High demand sensitivity 28

Table 8 Summary of emission outcomes by fuel type (Mt CO2-e): Low demand sensitivity 30

Table 9 Fuel cost variations relative to baseline assumptions 31

Table 10 Summary of emission outcomes by fuel type (Mt CO2-e): High fuel cost sensitivity 32

Table 11 Summary of emission outcomes by fuel type (Mt CO2-e): Low fuel cost sensitivity 33

Table 12 Summary of emission outcomes by fuel type (Mt CO2-e): EV sensitivity 34

Table 13 Summary of emission outcomes by fuel type (Mt CO2-e): Aggregate High sensitivity 36

Table 14 Summary of emission outcomes by fuel type (Mt CO2-e): Aggregate Low sensitivity 38

Table A1 Breakdown of electricity demand comparison for 2012-13 1

ix


Table A2 Baseline electricity demand projection for selected years by jurisdiction (GWh) 2

Table A3 Baseline electricity demand projection for selected years by grid (GWh) 2

Table A4 Existing and committed generators: type, capacity and life 6

Table A5 Existing and committed generators: efficiency, emissions and O&M costs 9

Table A6 Recently mothballed or retired units by NEM region 15

Table A7 Refurbishment costs for incumbent plant 16

Table A8 Base capital costs and cost component splits 19

Table A9 Final capital costs for new entrant technologies for selected years (Real 2011-12 $/kW installed) A-2

Table A10 Average real year-on-year capital cost change for each decade A-4

Table A11 New entrant parameters A-4

Table A12 Technology availability and construction profiles A-5

Table A13 Technology life and refurbishment costs A-6

Table A14 Assumed CO2 transport and storage costs A-7

Table A15 Lag between installation and registration of PV installations with the CER A-9

Table A16 PV installation premium/discount by system size A-11

Table A17 State/territory variation in system cost A-12

Table A18 Solar Credits multiplier A-13

Table A19 Feed-in tariffs by jurisdiction A-15

Table A20 Solar zone ratings A-15

Table A21 Installed capacity by solar zone A-16

Table A22 Export rates A-16

Table A23 Uplift factors by customer class A-20

Table A24 Hedging factors and capacity credit costs by customer class A-20

Table A25 Basis of network cost calculation A-21

Table A26 PEC creation and EITE load A-22

x


GlossaryAcronym or term Explanation

AEMO Australian Energy Market Operator, the entity that manages dispatch and planning in the National Electricity Market.

AETA Australian Energy Technology Assessment, an analysis of future generation costs from various electricity supply technologies undertaken by BREE in 2012.

ARENA The Australian Renewable Energy Agency, a statutory authority of the Commonwealth Government to support renewable energy

Bagasse A renewable fuel produced from sugar cane waste.

BREE Bureau of Resources and Energy Economics, a Commonwealth Government research agency.

Capacity factor A measure of the intensity with which a generator operates, calculated as the generator’s average output divided by its maximum possible output, and typically expressed as a percentage.

CCGT Combined-cycle gas turbine, a gas turbine generator where waste heat from the turbine exhaust is captured and used to drive an auxiliary steam turbine.

CCS Carbon capture and storage, the capturing of carbon dioxide produced in the process of generating electricity (or some other industrial process) and storing

CGEComputable General Equilibrium modelling, a form of modelling that relates the inputs and outputs of 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.

Cogeneration, or ‘cogen’

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, 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.

CSIRO The Commonwealth Scientific Industrial and Research Organisation, an Australian Government scientific research agency

DKIS Darwin-Katherine Interconnected System, the interconnected electricity grid servicing the main population centres of the northern part of the Northern Territory.

Dual axisIn the context of solar PV generation, this refers to solar PV plates that can change angle to track the sun on two axes, an axis to track daily east-west movement of the sun across the sky and a second axis to adjust to changes in the sun’s angle (north-south) with the seasons. See also ‘fixed axis’ and ‘single axis’.

EGS Engineered geothermal system, a form of geothermal generation technology also sometimes known as ‘hot fractured rocks’.

Fixed axis In the context of solar PV generation, this refers to solar PV plates that are mounted in a fixed position and do not track the sun. See also ‘single axis’ and ‘dual axis’.

FOM Fixed operating and maintenance costs. These are represented in ACIL Allen’s modelling as a fixed annual payment required to keep a power station operational.

GALLM Global and Local Learning Model, CSIRO’s model of generation technology costs.

GGAS Greenhouse Gas Abatement Scheme, the NSW Government’s former emissions reduction scheme

GWh Gigawatt-hour, a unit of electricity output or consumption measured over time, which is equivalent to one gigawatt being produced/consumed continuously for one hour, or one thousand megawatt-hours.

HEGT High efficiency gas turbine.

HSA Hot sedimentary aquifer, a form of geothermal generation technology.

IGCC Integrated gasification combined cycle, a form of generation technology that uses coal as the fuel, and which converts the coal to a synthetic gas to drive a gas turbine through an integrated process.

IMO Independent Market Operator, the the entity that manages dispatch and planning in the South-West Interconnected System.

kW Kilowatt, a unit of (instantaneous) electricity output or consumption, equal to one one-thousandth of a megawatt.


1


Acronym or term Explanation

LDC Load duration curve, a representation of the variation in electricity demand over a period of time created by ordering the electricity demand (or ‘load’) in descending order.

LGCLarge-scale Generation Certificate, the certificate that can be created and traded by renewable generators under the LRET. Sometimes referred to as a ‘REC’, or Renewable Energy Certificate. LGCs are different from the ‘Small-scale Technology Certificates’ or STCs created under the SRES.

LP Linear programming

LRETLarge-scale Renewable Energy Target, the Commonwealth Government’s scheme to promote large-scale renewable electricity generation. Formerly known as the Mandatory Renewable Energy Target (MRET), and sometimes referred to simply as the RET.

MLF Marginal loss factor, the level of transmission losses between a given generator and the point of 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, equal to one thousand kilowatts.

MWh Megawatt-hour, a unit of electricity output or consumption measured over time, which is equivalent to one megawatt being produced/consumed continuously for one hour.

NEM National Electricity Market, the interconnected electricity grid covering most of Queensland, New South Wales, Victoria, Tasmania and South Australia.

NWIS North-West Interconnected System, the interconnected electricity grid covering the Pilbara region of north-western Western Australia.

O&M Operating and maintenance costs – see also FOM and VOM.

OCGT Open cycle gas turbine, a gas turbine generator where waste heat is vented to the atmosphere rather than captured to generate electricity or steam, as in a combined-cycle gas turbine (CCGT) or cogeneration plant.

Oxy combustionA technique used to improve the efficiency of CCS, by firing coal in a primarily oxygen and non-combustible gases (importantly, in the absence of nitrogen), so as to produce a relatively pure stream of CO2 suitable for capture and storage.

PC Pulverised coal. See also ‘pf’

pf Pulverised fuel, typically coal. See also ‘PC’.

POE Probability of exceedence, representing the probability that a given forecast will be exceeded in the relevant forecast period.

PV Photovoltaic, a form of generation that converts solar radiation to direct current electricity using semi-conductors that exhibit the photovoltaic effect.

SF Solar Flagships, the Commonwealth Government’s program to promote large-scale solar generation projects.

Single axis In the context of solar PV generation, this refers to solar PV plates that can change angle to track the east-west daily movement of the sun across the sky. See also ‘fixed axis’ and ‘double axis’.

SRESSmall-scale Renewable Energy Scheme, the Commonwealth Government’s scheme to promote small-scale renewable energy technologies, principally solar PV and solar water heaters. The incentives for these technologies were formerly combined with those for large-scale renewables through the MRET.

SRMCShort-Run Marginal Cost, an economic interpretation of the extent to which production costs, in this case electricity generation costs, vary at the margin when key inputs, particularly the capital equipment comprising the generator, cannot be varied.

SWIS South-West Interconnected System, the interconnected electricity grid covering south-western Western Australia. Also known as the Wholesale Electricity Market, or WEM.

VOM Variable operating and maintenance costs. These are represented in ACIL Allen’s modelling as costs which vary linearly with the amount of electricity produced by a given power station (i.e. as a cost in $/MWh).

WACC Weighted average cost of capital, a benchmark rate of return on capital investments representing an assumed level of equity and debt financing, and specific rates of return to each of equity and debt.

WCMG Waste coal mine gas


2


1 IntroductionThe Department of the Environment (the Department) commissioned ACIL Allen Consulting (ACIL Allen) to undertake projections of greenhouse gas emissions from Australia’s electricity generation sector over the period 2008-09 to 2034-35. This assignment is intended to inform emissions projections and support the annual publication of greenhouse gas emissions projections undertaken by the Department in tracking Australia’s progress towards its 2020 emissions reduction target.

The electricity sector currently accounts for around one third of Australia’s emissions as reported under the National Greenhouse Gas Inventory and is the largest single source. Modelling has been undertaken for two scenarios as follows:

Baseline scenario: Current electricity demand and supply side measures continue for the length of the projection (or until the planned end date of a particular program). The scenario includes modifications to the current Large-scale Renewable Energy Target (LRET) scheme so that mandated GWh targets in 2020 represent an estimated 20 per cent of Australian electricity consumption. This was estimated to be 27,000 GWh. All defined Energy Intensive Trade Exposed (EITE) activities are 100 per cent shielded from RET impacts (both LRET and SRES). The scenario does not include any changes to the SRES.

No supply-side measures: Current electricity demand side measures remain in place for the length of the projection (or until the planned end date of a particular program). Electricity supply side measures do not continue from 1 July 2014. Under this scenario, the only supply-side measures assumed to cease was the RET scheme.

In addition to the above policy scenarios, a range of sensitivities were also undertaken to test the robustness of results and key drivers of particular outcomes. More details on the scenarios and sensitivities can be found in section 2.2 and 2.3 respectively.

ACIL Allen has utilised its long-term dynamic planning model, PowerMark LT for this assignment as well as a range of other sub-models. PowerMark LT was used to estimate development and policy effects in Australia’s major electricity markets namely, the National Electricity Market (NEM), the Western Australian Wholesale Electricity Market which covers the South-West Interconnected System (SWIS), the North-West Interconnected System (NWIS) in the Pilbara region, the Darwin-Katherine Interconnected System (DKIS) and the grid serving Mount Isa. Other generation facilities outside of these major grids and on-grid not covered under these market arrangements were projected in separate sub-models.

In terms of generation coverage, ACIL Allen has drawn upon confidential historical data for years 2011-12 and 2012-13 made available through the National Greenhouse Gas Inventory and the National Greenhouse and Energy Reporting (NGER) Scheme. These were used to calibrate input assumptions and settings for individual facilities to align with historical outcomes as reported by the Department.

The projections are intended to cover all electricity generation facilities – including those previously deemed to be part of vertically integrated processes such as mining/minerals processing and LNG plants. As such, emission values will be higher than those reported from the 2013 emission projections study as they include some emissions which were previously covered under ‘Direct Combustion’ activities.


3


Greenhouse gas Global Warming Potentials (GWPs) for carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) have been taken from the Intergovernmental Panel on Climate Change Fourth Assessment Report (AR4) in accordance with the current reporting convention within the National Greenhouse Gas Inventory.

As part of a separate study, pitt&sherry undertook demand projections to be used under the engagement. This study focused primarily on projected electricity demand on main grids. ACIL Allen was also tasked with providing the balance of the demand assumptions for those demand components not covered by the pitt&sherry work, primarily additional large industrial, mini-grid, off-grid and grid-exempt demand’.

This report represents ACIL Allen’s primary deliverable under this assignment for the Department. It is accompanied by spreadsheet results from the modelling.

The remainder of this report is structured as follows:

Section 2 gives an overview of the project, including methodology, the models used, and a description of the scenarios and sensitivities modelled

Section 3 highlights the key modelling results for scenarios

Section 4 outlines the results from the modelled sensitivities

Appendix A sets out the key modelling assumptions, including those derived from the demand projections and those adopted within the electricity sector modelling.


4


2 Project overview

2.1 Modelling processFigure 1 provides an overview of the modelling process used to undertake the projections for the Department under this project.

Figure 1 Modelling framework used

PowerMark LT

Market charateristics- existing generator inputs- new entrant options and costsCoverage:NEMSWISDKISNWIS and Mt Isa

LRET settings- targets and banking rules- existing generators- new entrant generators- certificate multipliers per MWh

SRES model- projected SWH- projected SGU (solar PV)

Demand inputs- market facing sent-out energy and peak demand- 100 point sampled load duration curves

Model outputs- Emissions (CO2-e)- Generation/capacity mix- Fuel use- Wholesale electricity prices- LGC prices- LRET market surrenders/shortfalls- New entry tehnology and build timing- Retirements and refurbishments- System resource costs

Retail electricity pricing model- network costs- retail costs and margin- hedging costs (based on load shape)- retail series for households- retail series for SMEs/large users

Off-grid database- detailed breakdown of existing generation ~ 16,000 GWh- curent renewable composition approximately 2%

Source: ACIL Allen

2.1.1 PowerMark LT

PowerMark LT simulates the electricity markets 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


5


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.

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.

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 configured to model all of Australia’s major electricity grids, namely:

the National Electricity Market (NEM), covering New South Wales, Australian Capital Territory, 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 geographic extent of these grids is shown in Figure 2.


6


Figure 2 Modelled electricity grids

Note: Grid extents are representative onlySource: ACIL Allen

As discussed above, PowerMark LT is designed to optimise outcomes over long model horizons and can model out to 2050. Under this assignment, the Department required outcomes to be modelled over the period 2008-09 to 2034-35 (6 years of history; 21 year projection). ACIL Allen extended the modelling period to 2039-40 (a total of 32 years) to ensure the outputs do not suffer from any end effects. This extended period also captures the potential impact of forward-looking investment behaviour toward the end of the results period leading up to 2034-35.

Generation from non-scheduled and off-grid sources were not be modelled with PowerMark LT, but estimated using ACIL Allen’s off-grid database. Both components have been calibrated to achieve alignment with emissions and fuel consumption data within the latest available NGGI (June 2014 Quarterly) and NGER (2011-12 and 2012-13) datasets.

PowerMark LT was supplemented by some of our other Excel-based models, namely our SRES projection model, our Retail price model and our off-grid/embedded database. The following section provides more information on the modelling process and how these models interact.

2.1.2 LRET implementation in PowerMark LT

The LRET is explicitly incorporated into PowerMark LT, inputs comprise of:

Certificate demand: mandated targets, voluntary demand through GreenPower and desalination and other voluntary surrenders

Supply from existing generators (both scheduled generators and non-scheduled generators)

Cumulative banked certificates at the start of the model horizon and certificate financial holding costs

Constraints on banking and borrowing, incorporating the three year “make-good” allowance for liable entity shortfalls


7


New entrants generators creating LGCs are those flagged as eligible technologies within PowerMark LT’s new entrant list. A certificate multiplier (typically one) can be applied for each generation technology per MWh of electrical output.

2.1.3 Retail price model

ACIL Allen’s retail price model develops projections of all components of retail electricity prices for each State and Territory for the purpose of translating wholesale electricity and LRET/SRES outcomes into average representative prices for residential, SME’s and large end users in each jurisdiction.

The retail price model takes these wholesale prices and applies projections of the other components of retail prices to arrive at a retail series applicable for various customer types. We note that there is a certain amount of circularity between the SRES and retail price models as changes in retail prices affects the paybacks for behind-the-meter solutions.

The cost components modelled include: network charges; wholesale energy; LRET/SRES; losses; FiT costs; other green scheme costs; metering charges; retail operating costs, NEM pool fees, ancillary service costs and retail margin.

2.1.4 SRES projection model

ACIL Allen’s forecasts for uptake of small-scale generation units (SGUs) are based on a regression model relating historic uptake to historic net financial returns to installing solar PV 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.

The model uses a quarterly resolution and separately estimates uptake for each state and territory. 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.

2.1.5 Off-grid and embedded generationACIL Allen has recently undertaken an extensive analysis of off-grid electricity generation for the Commonwealth Government, published as a Bureau of Resources and Energy


8


Economics (BREE) report in October 2013. This snapshot of off-grid generation was used as a starting point to help inform the projections of emissions from remote locations.

This was supplemented by detailed analysis of generation and emissions outcomes as reported from NGERs reports.

2.2 ScenariosThis section outlines the scenario design for the ‘Baseline’ and ‘No supply-measures’ scenarios as defined by the Department.

2.2.1 Baseline scenario

The Baseline scenario includes all current electricity demand and supply side measures currently in place continue for the length of the projection (or until the planned end date of a particular program). It includes the effects of a range of specific greenhouse gas abatement measures, including energy efficiency programs, the Large-scale Renewable Energy Target (LRET) and the Small-scale Renewable Energy Scheme (SRES).

At the request of the Department, the scenario’s design includes modifications to the LRET such that the annual GWh targets for compliance years 2016 to 2030 are adjusted such that the proportion of renewables in Australia’s energy mix in 2020 is 20 per cent.

It also includes the modification of the scheme so that the electricity used in all defined Energy Intensive Trade Exposed activities are 100 per cent exempt from RET liability (both LRET and SRES). As this reduces the denominator in the calculation of the RPP and STP values (the means by which aggregate liabilities are translated to individual customers) non-exempt customers will see larger RET cost imposts as a result of the EITE exemption.

The Department has requested ACIL Allen calculate the revised 2020 and interim targets for this purpose. The ‘Real 20%’ level has been calculated by ACIL Allen in accordance with the following formula which is similar in nature to that used during the 2014 RET Review:1

2020GWhtarget=E2020∗20%−(ℜSRES PV+ℜPre−RET )

Where:

E2020: Projected electricity generation in calendar year 2020 (in sent-out GWh) across all Australian grids, including estimates of embedded and off-grid energy.

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.

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.2

This results in a rounded 2020 target of 27,000 GWh, with interim targets (2016 to 2019) revised as a straight line trajectory as detailed in Table 1. The scenario does not involve and changes to the currently legislated SRES.

1 See ACIL Allen, RET Review Modelling: Market modelling of various policy options, August 2014, page 38. The 2014 RET Review formula contained a more detailed breakdown of demand and solar PV components, but had the same overall intent. The impact of electricity displacement from solar water heater installations is excluded in the calculation.

2 The RET Review settings for the existing baseline energy was set at a higher level of 16,148 GWh which ACIL Allen acknowledged was at the upper end of a reasonable range. For this exercise a more conservative value of 15,000 GWh has been adopted.


9


Table 1 Real 20% by 2020 annual LRET targets

Calendar year Current legislation Baseline scenario Change

2014 16,100 16,100 0

2015 18,000 18,000 0

2016 20,581 19,800 -781

2017 25,181 21,600 -3,581

2018 29,781 23,400 -6,381

2019 34,381 25,200 -9,181

2020 41,000 27,000 -14,000

2021-30 41,000 27,000 -14,000

Note: Values exclude 850 GWh allowance for WCMGSource: ACIL Allen

The Baseline scenario reflects the repeal of the carbon pricing mechanism from 1 July 2014 but does not include the impact of the Government’s Direct Action Plan.

The scenario includes a restriction on the development of new conventional coal-fired capacity (i.e. coal-based technologies which do not employ carbon capture and storage). This aligns with ACIL Allen’s standard assumption when undertaking market outlook studies and can be justified within a scenario which employs greenhouse gas abatement policies due to a range of factors including:

Community views and corporate sustainability policies

Potential difficulty obtaining generation licences from State and Territory governments

Long-term risk of explicit carbon pricing being reintroduced

Difficulty in securing financing on a commercial basis due to these risks.

While the scenario does not allow the development of new coal, it does permit existing stations to refurbish, through life extension capital projects, where it is economic to do so.

The scenario includes the impact of voluntary abatement provided through GreenPower. The scenario assumes 1,000 GWh of GreenPower annually throughout the projection period.

The impacts of other state-based schemes – for example the ACT renewables policy –have not been incorporated. Whilst the results of the latest wind auction have been made public3, to-date these projects remain uncommitted and may not proceed. As such the Department instructed ACIL Allen not to assume this additional generation necessarily occurs.

2.2.2 No supply-side measures scenario

The ‘No supply-side measures’ scenario removes the influence of supply-side policy interventions designed to reduce greenhouse gas emissions from the electricity generation sector. Upon review of the measures contained within the Baseline scenario, it was decided in conjunction with the Department that the only measures removed would be the LRET and SRES.

Consideration was given whether any remaining solar PV feed-in-tariff schemes were above cost reflective levels thereby providing additional incentives for the deployment of rooftop solar PV. However upon review, it was decided that those schemes in place have been scaled back to provide roughly the economic value of energy exported to the grid.

The ‘No supply-side measures’ scenario therefore reduced to a No RET scenario. In accordance with the scenario design, this was assumed to have occurred from 1 July 2014,

3 See http://www.environment.act.gov.au/energy/wind_power


10

http://www.environment.act.gov.au/energy/wind_power


with all renewable deployment which had occurred up until that time assumed to remain in place.4 Therefore, the results for the period 2008-09 to 2013-14 are unchanged from what actually occurred in the Baseline scenario and no historical counterfactual modelling was required.

The No supply-side measures scenario does not have any restrictions on the development of new conventional coal-fired generation capacity. It is assumed that under this scenario there is no opposition to development of these technologies and consideration of greenhouse gas emissions do not form part of developers’ technology evaluations. Existing stations are also permitted to refurbish to extend lives where it is economic to do so.

2.3 SensitivitiesThe project also required ACIL Allen to test the robustness of the modelling results by running sensitivity analysis on the Baseline scenario across a range of key input parameters. The sensitivity analysis was used to determine which assumptions have the greatest impact on projections of Australia’s electricity sector emissions.

The project included four pairs of sensitivity scenarios (high and low) on the baseline scenario – three individual variable elements and a single combination sensitivity which will include the modification of a number of input elements simultaneously. These comprised of combinations of deviations that yield the highest and lowest aggregate emissions.

2.4 Historical calibration with National Greenhouse Gas Inventory

Before modelling commenced, the Department provided ACIL Allen with a detailed breakdown of historical National Greenhouse Gas Inventory values from the June 2014 Quarterly report, by generator for the purposes of historical calibration. This revised NGGI series included the revised definition for fuel consumption by BREE where all fuel consumed for electricity generation activities is classified as ‘electricity generation’ whereas previously some of these values would have reported under Direct Combustion.

With the use of the detailed NGERs reports for 2011-12 and 2012-13 ACIL Allen examined emission outcomes at the facility level and determined emission intensity values. These were then applied to actual generation outcomes for years 2008-09 to 2013-14 (sourced from market operator data and other sources) and matched against emissions reported in the National Greenhouse Gas Inventory for electricity generation.

This process revealed some differences between the two data sets including instances where some emissions from electricity generation continued to be reported under Direct Combustion activities rather than electricity generation. This was a result of the method applied for the National Greenhouse Gas Inventory, which uses reported fuel consumption by BREE for larger facilities and groups the residual reported fuel consumption into an ‘other’ category. By comparison, ACIL Allen’s approach utilised reported generation volumes based on reported market operator dispatch data.

In order to achieve alignment between the two series, ACIL Allen was asked by the Department to exclude a number of generation facilities from the historical period and projections. For consistency these facilities were also excluded from the projections and emissions associated with them should be accounted for in Direct Combustion within the Department’s overall projections work. This ensured coverage of emissions across

4 Renewable projects committed and under construction at that point in time were assumed to be completed as per the Baseline scenario.


11


stationery energy in the Department’s projection match the National Greenhouse Gas Inventory.

Once these facilities were excluded from the historical series, close alignment between ACIL Allen and the National Greenhouse Gas Inventory was achieved as shown in Figure 3. The largest differential of the Australian totals was 0.3 Mt CO2-e in 2013-14 which this accounting for a 0.17 percent differential.

Figure 4 provides a comparison of the two series by year and by State showing close alignment even within this more disaggregated view of the data.

Figure 3 Model calibration against actual National Greenhouse Gas Inventory values

Note: Emissions based on IPCC AR4 GWPsSource: National Greenhouse Gas Inventory June 2014 Quarterly report, ACIL Allen analysis


12


Figure 4 Comparison with National Greenhouse Gas Inventory by year by State (Mt CO2-e)

Note: ET = Extra territoriesSource: National Greenhouse Gas Inventory, ACIL Allen analysis


13


3 Scenario results

3.1 Baseline scenarioFigure 5 and Table 2 presents the projected results of generation by fuel type under the Baseline scenario. The period presented includes the historical period of 2008-09 to 2013-14, with projections through to 2034-35.5

In the Baseline scenario, which includes a modified RET to a real 20% level as discussed in the previous section, coal-fired generation continues to supply the bulk of Australia’s energy requirements over the projection period. Whilst the scenario does not introduce any new coal-fired plant (new conventional coal is restricted from entry in this scenario), incumbent plant continue to operate through life extension programs.

A key feature of the scenario is the development of new renewables through the RET, particularly over the period to 2020. In the period to 2020 this new capacity is primarily wind, with around 4,000 MW of new capacity added to 2020-216, with smaller amounts of utility-scale solar PV in regional grids. On-going development of rooftop solar PV occurs throughout, encouraged by subsidies under the SRES.

Annual historical aggregate generation volumes have been relatively flat at around 240 TWh through 2008-09 to 2012-13 (declines in the NEM offset by growth in other regions), with declines more recently to 234 TWh in 2013-14. Based on the demand projection, aggregate generation volumes (on a sent-out basis) are expected to increase to 262 TWh by 2019-20 and to 317 TWh by 2034-35.

Figure 5 Australian generation by fuel type: Baseline scenario

Source: ACIL Allen

5 Note that the modelling extended beyond 2034-35 (to 2039-40) to ensure there were no end effects and to capture the impact of forward-looking investment behaviour.

6 Increase in capacity relative to 2012-13 level. This total includes capacity which is already committed or under construction.


14


Table 2 Australian generation by fuel type (sent-out TWh): Baseline scenario

2008-09 2012-13 2019-20 2034-35

Black coal 123.7 105.5 124.0 148.6

Brown coal 52.5 44.4 46.1 49.0

Peaking Gas 8.9 8.3 1.9 2.8

Baseload Gas 26.0 34.0 30.1 45.2

Cogen 5.9 7.4 3.8 2.8

Liquid fuel 3.7 4.4 4.0 4.4

Hydro 12.1 18.3 15.7 14.2

Wind 3.9 8.1 22.6 23.2

Solar 0.1 3.6 10.8 23.7

Biothermal 2.5 2.9 3.0 3.0

Geothermal 0.0 0.0 0.0 0.0

Total 239.3 236.7 262.1 317.1

Source: ACIL Allen

Figure 6 shows the projected annual emission outcomes for Australia, broken down by generation type in million tonnes CO2-e. Coal-fired generation accounts for around 85% of emissions in 2012-13 and it is the change in coal-fired output over time which is the principal determinant of the emissions trajectory. Natural gas, being much less emission intensive, has less influence, currently only accounting for around 14% of aggregate emissions.

In the Baseline scenario aggregate emissions are projected to increase from 187 Mt in 2012-13 to 200 Mt by 2016-17 (a 6.8% increase) driven by the decline in gas-fired generation and a corresponding increase in coal-fired output. The decline in gas-fired generation is a result of increasing wholesale gas prices over this period, with gas-fired peaking output particularly suffering due to the oversupplied wholesale market resulting in few generation opportunities. Emissions remain relatively flat though to 2019-20 at around 201 Mt despite the demand growth over the period due primarily to the introduction of renewables under the LRET.

By 2034-35, projected emissions from the generation sector reach 236 Mt, a 26% increase over 2012-13 levels.

Figure 6 Australian emissions by fuel type: Baseline scenario

Source: ACIL Allen


15


A range of summary figures for emissions are presented in Table 3. Aggregate emissions are projected to increase by 7.7% over the period 2012-13 to 2019-20 (equivalent to average annual growth rate of 1.1% per annum). A number of factors are driving this outcome:

The repeal of the carbon tax effective from 1 July 2014 results in a shift back to coal-fired generation

This is further exacerbated by the decrease in gas-fired generation owing to wholesale gas price increases and oversupplied wholesale markets

Projected demand growth under the scenario which results in some mothballed coal-fired generators returning to service

Limited development of renewables over the period.

Over the period 2019-20 to 2035-35 annual emissions increase by a further 17% (equivalent to average annual growth of 1.1% per annum).

Table 3 Summary of emission outcomes by fuel type (Mt CO2-e): Baseline scenario


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 117.2 140.9 -16.3% 18.7% 20.3% 855.6 1,892.5

Brown coal 69.0 59.6 61.1 65.4 -13.7% 2.5% 7.1% 506.4 995.2

Peaking Gas 5.5 5.0 1.0 1.6 -9.6% -78.9% 53.0% 17.0 15.3

Baseload Gas 13.6 17.4 17.9 24.4 28.3% 2.5% 36.4% 134.9 320.0

Cogen 3.3 3.9 1.9 1.4 18.9% -51.9% -27.8% 23.1 23.4

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -4.1% 9.1% 17.4 35.3

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 201.4 236.2 -11.4% 7.7% 17.3% 1,555.5 3,283.1

Source: ACIL Allen

Figure 7 provides generation results by fuel for the NEM, SWIS, a combined NWIS/DKIS view and for off-grid generation. Key features of the results include:

A resurgence in coal-fired generation in the NEM (particularly black coal) in the near-term, offset by steep reductions in gas-fired generation. Coal-fired generation accounted for 82% of generation in the NEM in 2008-09; this has since fallen to 72% in 2012-13. The projection sees coal output averaging 75% of total NEM generation through to 2034-35. Gas-fired output declines significantly over the next few years.

According to the demand projection used, the SWIS sees significant demand growth over the period and high wholesale prices attract wind and solar PV developments. Coal generation increases marginally over the period as underutilised capacity is taken up. Gas-fired output increases in the longer-term, meeting growth in energy requirements.

The NWIS and DKIS are dominated by gas-fired generation with large-scale solar PV developments projected to occur from 2017-18 onwards due to the relatively high wholesale cost of energy projected in these grids

Off-grid generation is also dominated by gas-fired generation with liquid fuelled generation continuing at remote sites including some solar PV penetration. Much of the growth in gas-fired generation relates to generation units associated with new LNG plants in Queensland, WA and NT.


16


Figure 7 Generation by fuel type by grid: Baseline scenario

Note: Off-grid includes Mt IsaSource: ACIL Allen

Figure 8 shows a similar view of generation, but broken down by jurisdiction. Key points to note:

NSW, Victoria and Queensland are dominated by low cost coal-fired generation with gas and renewables only accounting for a relatively small proportion of generation output

South Australia sees a significant shift away from natural gas with renewables accounting for a large proportion of generation. Coal provides baseload energy, with natural gas only playing a back-up role to renewables

Western Australia’s supply mix largely remains constant with growing contributions from renewables

Tasmania and NT see some growth in renewables, but overall generation volumes are largely unchanged over the period.


17


Figure 8 Generation by fuel type by jurisdiction: Baseline scenario

Source: ACIL Allen

Figure 9 presents projected emission intensity trends (on a ‘sent out’ basis) for each state, for the Baseline scenario. Aside from the decline in South Australia, emission intensities are relatively stable, declining only very slightly over time. The Australian average intensity falls from 0.79 tonnes CO2-e/MWh sent-out in 2012-13 to 0.77 by 2019-20 and thereafter to 0.75 tonnes CO2-e/MWh sent-out by 2034-35.


18


Figure 9 Projected emission intensity: Baseline scenario

Source: ACIL Allen

Figure 10 shows the changes in generation capacity by fuel type over the period. Increases in generation capacity are almost entirely associated with renewable entry: wind (4,020 MW) and solar PV (14,850 MW). Large-scale wind developed ceases once sufficient capacity to meet the ‘real 20% RET’ is achieved whereas solar PV installations (primarily small-scale rooftop) continues to grow throughout due to the financial attractiveness of this as a distributed technology.

Figure 10 Projected generation capacity: Baseline scenario

Note: Solar PV capacity includes rooftop PVSource: ACIL Allen


19


3.2 No supply-side measuresFigure 11 and Table 2 presents the projected results of generation by fuel type under the No supply-side measures scenario.

Outcomes for this scenario are very similar to those under the Baseline scenario, with the exception that in the absence of the RET, no new wind, nor utility-scale solar PV developments occur within the projections as these technologies are not commercially viable without subsidies. Figure 12 shows the change in dispatch outcomes between the scenarios. The reduction in renewable generation is met by increased fossil-fuelled dispatch (primarily black and brown coal).

Figure 11 Australian generation by fuel type: No supply-side measures

Source: ACIL Allen

Figure 12 Australian generation by fuel type: No supply-side measures change from Baseline scenario

Note: Generation results from the scenario minus outcomes under the Baseline scenarioSource: ACIL Allen


20


Table 4 Australian generation by fuel type (sent-out TWh): No supply-side measures

2008-09 2012-13 2019-20 2034-35

Black coal 123.7 105.5 126.4 166.6

Brown coal 52.5 44.4 52.0 49.0

Peaking Gas 8.9 8.3 2.7 2.6

Baseload Gas 26.0 34.0 32.5 39.9

Cogen 5.9 7.4 3.9 2.7

Liquid fuel 3.7 4.4 4.1 4.4

Hydro 12.1 18.3 16.3 16.3

Wind 3.9 8.1 12.5 12.4

Solar 0.1 3.6 9.1 20.1

Biothermal 2.5 2.9 3.0 3.0

Geothermal 0.0 0.0 0.0 0.0

Total 239.3 236.7 262.4 317.0

Source: ACIL Allen

Figure 13 shows the projected annual emission outcomes for Australia under this scenario, broken down by generation type in million tonnes CO2-e. In the ‘No supply-side measures’ scenario, total annual emissions increase from 187 Mt in 2012-13 to 214 Mt by 2019-20 (a 14.4% increase). In aggregate, emissions over this period are 1,583 Mt which is around 27.2 Mt (1.7%) higher than the Baseline scenario. This is due to the lower level of renewables developed and increased utilisation of coal-fired capacity over this period.

The differential between the two scenarios grows over time. In the period 2020-21 to 2034-35 emissions total 3,457 Mt under this scenario compared with 3,283 Mt under the Baseline scenario. This represents an increase of around 174 Mt, or 5.3%.

By 2034-35, projected emissions from the generation sector reach 248 Mt, a 32.4% increase over 2012-13 levels.

Figure 13 Australian emissions by fuel type: No supply-side measures

Source: ACIL Allen

Figure 14 summarises the projected impact of the supply-side measures on emissions outcomes relative to the Baseline scenario and Figure 18 provides the change in emission by fuel type.


21


Figure 14 Impact of supply-side measures on emissions

Source: ACIL Allen

Figure 15 Australian emissions by fuel type: No supply-side measures change from Baseline scenario

Note: Emission results from the scenario minus outcomes under the Baseline scenarioSource: ACIL Allen

A range of summary figures for emissions are presented in Table 5 Aggregate emissions are projected to increase by 14.4% over the period 2012-13 to 2019-20 (equivalent to annual average growth rate of 1.9% per annum). Over the period 2019-20 to 2034-35 annual emissions increase by 15.7% (equivalent to annual growth of 1% per annum).

Figure 16 provides generation results by fuel by grid and Figure 17 provides results by jurisdiction. These results are broadly similar to those under the Baseline scenario. The key difference is the lower level of renewables developed in the absence of the RET. In the No measures scenario, entry of conventional coal is not restricted and the modelling projects a small amount of new conventional coal would be developed in the SWIS from around 2020 and in Queensland from 2025 onwards. By the end of the projection a total of around 3,350 MW of new conventional coal is developed in these markets.


22


Table 5 Summary of emission outcomes by fuel type (Mt CO2-e): No supply-side measures


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 119.5 154.6 -16.3% 21.1% 29.3% 854.9 2,021.0

Brown coal 69.0 59.6 69.7 65.4 -13.7% 16.9% -6.1% 531.7 1,035.5

Peaking Gas 5.5 5.0 1.5 1.4 -9.6% -69.0% -5.7% 18.2 18.0

Baseload Gas 13.6 17.4 19.0 22.3 28.3% 8.8% 17.6% 136.3 322.1

Cogen 3.3 3.9 2.0 1.3 18.9% -50.1% -33.4% 23.3 23.7

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -3.9% 9.1% 17.4 35.4

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 214.0 247.6 -11.4% 14.4% 15.7% 1,582.7 3,457.3

Source: ACIL Allen

Figure 16 Generation by fuel type by grid: No supply-side measures

Note: Off-grid includes Mt IsaSource: ACIL Allen


23


Figure 17 Generation by fuel type by jurisdiction: No supply-side measures

Note: Other = Tasmania and the Northern TerritorySource: ACIL Allen

Figure 18 presents projected emission intensity trends (on a ‘sent out’ basis) for the No supply-side measures scenario relative to the Baseline. The impact of the ‘real 20 per cent’ RET is to reduce the emissions intensity of the generation sector by around 0.04 tonnes CO2-e/MWh.

In the absence of supply-side measures the Australian average intensity increases from 0.79 tonnes CO2-e/MWh sent-out in 2012-13 to 0.82 by 2019-20 and 0.78 by 2034-35.


24


Figure 18 Projected emission intensity: No supply-side measures

Source: ACIL Allen

Figure 19 shows the changes in generation capacity by fuel type relative to the Baseline scenario. In the absence of the supply-side measures projected installed capacity of renewables is considerably lower (3,300 MW less wind, 2,400 MW less solar PV by 2025). This is offset by some increase in fossil fuel capacity, most of which is associated with the earlier return to service of mothballed plant and the development of some new coal-fired capacity in the SWIS and Queensland toward the latter part of the projection period.

Figure 19 Projected generation capacity: No supply-side measures change from Baseline scenario

Note: Solar PV capacity includes rooftop PVSource: ACIL Allen


25


4 Sensitivity resultsThis section provides the results from the sensitivities run against the baseline scenario. The design of the sensitivities were agreed with the Department prior to modelling with some input assumptions provided by the Department and some of the variations developed by ACIL Allen.

4.1 High/Low demandIn long-term electricity market projections, wholesale electricity demand is one of the key uncertainties. Within the demand projections provided by pitt&sherry, high and low cases were also developed and these formed the basis for the high and low demand sensitivities.

Figure 20 shows aggregate Australian electricity demand under the high and low demand sensitivities relative to the baseline assumption. Table 6 also provides figures for selected years and the percentage deviation from the baseline assumptions. All three cases include significant growth in electricity demand over the period. Even the low case shows electricity demand increasing by around 67 TWh from 2014-15 to 2034-35 (a 29% increase).

Figure 20 Demand sensitivity inputs: Australia

Note: TWh sent-out at source of generation (i.e. includes network losses). Includes behind-the-meter consumption from sources such as rooftop solar PVSource: ACIL Allen based on pitt&sherry demand projections

Table 6 Demand sensitivity inputs for selected years and change from baseline: Australia2014-15 2019-20 2024-25 2029-30 2034-35

TWh % change TWh % change TWh % change TWh % change TWh % change

High demand 238.1 0.4% 282.1 3.8% 313.5 7.4% 346.0 9.9% 379.1 11.6%

Baseline 237.0 271.9 292.0 314.7 339.7

Low demand 235.7 -0.6% 263.3 -3.2% 275.5 -5.7% 289.8 -7.9% 303.3 -10.7%

Note: TWh sent-out at source of generation (i.e. includes network losses). Includes behind-the-meter consumption from sources such as rooftop solar PVSource: ACIL Allen based on pitt&sherry demand projections


26


Table 7 and Table 8 summarise demand for Australia by grid under the High and Low demand sensitivities respectively.

Table 7 Electricity demand projection for selected years by grid (GWh): High demand sensitivity

2008-09 2012-13 2014-15 2019-20 2024-25 2029-30 2034-35

NEM 204,697 197,382 191,552 216,962 236,236 257,787 280,887

SWIS 21,088 22,745 23,790 29,050 32,970 37,215 41,969

NWIS 1,551 2,783 2,995 3,681 4,136 4,565 5,003

DKIS 1,437 1,547 1,510 1,675 2,032 2,399 2,890

Mt Isa 1,914 2,198 2,312 2,583 2,941 3,441 3,940

Off grid 8,571 10,014 9,922 17,312 18,950 19,688 19,495

Total Australia 239,258 236,668 232,081 271,263 297,265 325,094 354,184

Note: GWh sent-out at source of generation (i.e. includes network losses). Includes rooftop solar PV; excludes some generation for calibration purposesSource: ACIL Allen

Table 8 Electricity demand projection for selected years by grid (GWh): Low demand sensitivity

2008-09 2012-13 2014-15 2019-20 2024-25 2029-30 2034-35

NEM 204,697 197,382 190,029 203,059 207,625 215,460 223,808

SWIS 21,088 22,745 22,981 24,857 25,790 27,186 28,678

NWIS 1,551 2,783 2,971 3,445 3,647 3,845 4,033

DKIS 1,437 1,547 1,499 1,567 1,791 2,019 2,328

Mt Isa 1,914 2,198 2,294 2,417 2,592 2,898 3,176

Off grid 8,571 10,014 9,922 17,314 18,951 19,689 19,496

Total Australia 239,258 236,668 229,695 252,658 260,397 271,097 281,519


Figure 21 shows the resulting emission paths for the Australian generator sector under the High and Low demand sensitivities. Despite the change in demand being roughly symmetrical either side of the baseline assumption, emission outcomes do not rise as much under the High demand assumptions as they fall under the Low demand assumptions. This is due to the higher demand being met predominantly by development of new gas-fired capacity which has a lower than average emission intensity. Conversely, under the low demand assumptions, the reduction in emissions is caused by lower output from coal-fired power stations.

Emission outcomes in 2019-20 under the High case are 3% higher than the Baseline and 3.6% lower under the Low demand assumptions. The change in cumulative emissions over the period 2012-13 to 2019-20 exhibit less sensitivity: 1.3% higher under High demand; -1.4% under Low demand assumptions.


27


Figure 21 Demand sensitivity aggregate emission outcomes: Australia

Source: ACIL Allen

Table 9 provides summary figures for emissions by fuel type for selected years under the High demand sensitivity.

Table 9 Summary of emission outcomes by fuel type (Mt CO2-e): High demand sensitivity


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 121.4 144.0 -16.3% 23.0% 18.6% 869.8 1,979.0

Brown coal 69.0 59.6 61.6 65.4 -13.7% 3.3% 6.3% 507.5 999.0

Peaking Gas 5.5 5.0 1.3 2.0 -9.6% -74.3% 53.2% 17.9 21.4

Baseload Gas 13.6 17.4 19.5 36.7 28.3% 12.1% 88.1% 140.6 422.7

Cogen 3.3 3.9 2.0 1.4 18.9% -49.8% -28.4% 23.5 26.9

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -4.0% 9.1% 17.4 35.3

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 208.1 252.1 -11.4% 11.3% 21.2% 1,577.6 3,485.8

Source: ACIL Allen

As shown in Figure 22, the increased emissions under the High demand sensitivity initially come from increased output from coal-fired generation as excess capacity is taken up in the market. Around half to two thirds of the increase in emissions is due to increased coal output over the period to 2028-29. Relative to the Baseline scenario, emissions increase in a linear fashion from 2014-15 to 2024-25 before stabilising somewhat thereafter at between 12 Mt and 17 Mt CO2-e per annum.

The additional generation capacity required to meet the higher demands is comprised of 6,150 MW of gas-fired capacity (5,400 MW of baseload/intermediate CCGT; 750 MW of additional peaking plant) and 850 MW of additional solar PV capacity.

Australia’s emissions intensity is slightly lower under this case (0.71 tonnes CO2-e/MWh sent-out compared with 0.75 tonnes CO2-e/MWh sent-out by 2034-35 under the Baseline scenario) due to the increased gas-fired and solar PV generation in the long-term. One of the reasons for this outcome is the restriction on new coal development assumed under the scenario. If new conventional coal was allowed to enter, it is likely that conventional coal


28


would be selected as the lowest cost option over gas-fired technologies and a larger impact on emissions under high demand assumptions could be expected.

Figure 22 Australian emissions by fuel type: High demand sensitivity change from Baseline scenario

Note: Emission results from the sensitivity minus outcomes under the Baseline scenarioSource: ACIL Allen

Table 10 and Figure 23 provides the corresponding emission figures by fuel under the Low demand sensitivity. Under low demand conditions, the market develops very little additional capacity aside from that required to meet the Real 20% LRET and rooftop solar PV.

The change in emission outcomes shows an increasing reduction in output from coal and gas-fired generation. By 2034-35 coal-fired generation is around 17 TWh lower and gas 15 TWh lower than under the baseline scenario. By comparison, renewable generation in 2034-35 is slightly lower: wind output (-1.2 TWh) and solar PV output (-0.9 TWh). While the LRET is set as a fixed target in GWh, slightly earlier wind build under this scenario results in less overall wind capacity being required. Lower uptake of rooftop solar PV is due to lower retail prices under the low demand conditions.

Aggregate emissions under the Low demand sensitivity are 6.7 Mt CO2-e lower in 2019-20 and 22.3 Mt CO2-e lower in 2034-35 relative to the baseline assumptions. Cumulative emissions for the period 2012-13 to 2019-20 are 1.3% lower relative to the baseline.


29


Table 10 Summary of emission outcomes by fuel type (Mt CO2-e): Low demand sensitivity


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 111.6 125.1 -16.3% 13.1% 12.0% 842.0 1,721.7

Brown coal 69.0 59.6 61.0 65.4 -13.7% 2.4% 7.2% 506.4 994.2

Peaking Gas 5.5 5.0 0.9 0.7 -9.6% -82.7% -18.6% 17.0 10.2

Baseload Gas 13.6 17.4 16.9 19.0 28.3% -2.9% 12.0% 128.1 274.0

Cogen 3.3 3.9 1.9 1.3 18.9% -52.6% -31.8% 23.1 21.7

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -4.2% 9.1% 17.4 35.3

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 194.6 213.9 -11.4% 4.1% 9.9% 1,534.8 3,058.7

Source: ACIL Allen

Figure 23 Australian emissions by fuel type: Low demand sensitivity change from Baseline scenario


4.2 High/Low fuel pricesThe next set of sensitivities tested the impact of higher and lower fossil fuel prices upon emissions outcomes. Inputs for these sensitivities were developed in consultation with the Department with the key inputs being percentage deviations from the baseline assumptions as shown in Table 11.


30


Table 11 Fuel cost variations relative to baseline assumptions2014-15 2015-16 2016-17 2017-18 2018-19 2019-20 2024-25 2029-30 2034-35

High fuel costsGas and liquid fuel 8% 8% 15% 23% 30% 35% 35% 35% 35%

Black and brown coal 5% 10% 15% 15% 15% 15% 15% 15% 15%

Low fuel costsGas and liquid fuel -8% -8% -15% -23% -30% -35% -35% -35% -35%

Black and brown coal -5% -10% -15% -20% -25% -30% -30% -30% -30%

Source: ACIL Allen and Department of the Environment

These variations were translated through to international traded prices for natural gas and thermal coal and resulted in the price series shown in Figure 24. As fuel costs to some power stations are insulated from movements in international prices (for example, vertically integrated supply or no access to export facilities), domestic mining and gas development costs were also adjusted such that the percentage change in costs were the same.

Figure 24 Fuel cost variations: International fuel price markers

Source: ACIL Allen

Figure 21 shows that aggregate emission outcomes are relatively insensitive to the variations in fuel costs examined.

Figure 25 Fuel cost sensitivity aggregate emission outcomes: Australia

Source: ACIL Allen


31


Table 9 provides the emission outcomes by fuel for selected years under the High fuel costs sensitivity. Cumulative emissions (2012-13 to 2019-20) are virtually unchanged from the Baseline scenario, with aggregate emissions only 0.2 Mt CO2-e higher.

Table 12 Summary of emission outcomes by fuel type (Mt CO2-e): High fuel cost sensitivity


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 118.1 142.1 -16.3% 19.7% 20.3% 857.3 1,915.6

Brown coal 69.0 59.6 61.1 65.4 -13.7% 2.4% 7.2% 506.8 996.9

Peaking Gas 5.5 5.0 1.0 1.4 -9.6% -78.8% 38.0% 17.6 14.3

Baseload Gas 13.6 17.4 17.5 22.9 28.3% 0.5% 30.7% 134.2 303.3

Cogen 3.3 3.9 1.8 1.3 18.9% -53.1% -26.8% 23.1 22.1

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -4.1% 9.1% 17.4 35.3

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 201.9 235.8 -11.4% 8.0% 16.8% 1,557.2 3,289.1

Source: ACIL Allen

The change in emissions is a result of higher coal-fired emissions offset (to some extent) by lower gas-fired emissions. This is a result of the relative competitiveness of coal versus gas within the inputs as gas prices tend to move by a larger amount when expressed in $/GJ terms. The higher fuel prices results in more renewable generation being developed – an additional 4,400 GWh per annum of solar PV by 2034-35, however the higher prices remain insufficient to make large-scale renewables competitive without additional subsidies.

Figure 26 Australian emissions by fuel type: High fuel cost sensitivity change from Baseline scenario


The Low fuel cost sensitivity shows the opposite effect, with higher gas-fired emissions and lower coal-fired emissions as shown in Table 10 and Figure 23. Under these inputs, cumulative emissions (2012-13 to 2019-20) are only slightly lower by 0.7 Mt CO2-e relative to the Baseline scenario. Development of renewables is lower due to lower retail prices


32


lowering the incentive for rooftop PV installations. Aggregate solar PV output is around 3,100 GWh by 2034-35.

Table 13 Summary of emission outcomes by fuel type (Mt CO2-e): Low fuel cost sensitivity


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 116.2 139.7 -16.3% 17.7% 20.3% 856.5 1,862.8

Brown coal 69.0 59.6 61.1 65.4 -13.7% 2.4% 7.2% 506.6 994.9

Peaking Gas 5.5 5.0 0.8 1.7 -9.6% -83.8% 113.2% 15.9 17.1

Baseload Gas 13.6 17.4 18.5 26.1 28.3% 6.5% 40.8% 135.7 341.9

Cogen 3.3 3.9 1.9 1.4 18.9% -50.9% -25.2% 23.2 25.6

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -4.1% 9.1% 17.4 35.3

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 200.8 237.0 -11.4% 7.4% 18.0% 1,556.2 3,279.2

Source: ACIL Allen

Figure 27 Australian emissions by fuel type: Low fuel cost sensitivity change from Baseline scenario


4.3 High uptake of electric vehiclesWithin the demand projections undertaken by pitt&sherry, the uptake of Electric Vehicles (EVs) was specifically excluded from the analysis. As part of this project, the Department asked ACIL Allen to estimate an electricity vehicle uptake scenario and include this incremental electricity demand to the Baseline scenario as an additional sensitivity.

ACIL Allen estimated the impact of the take up of plug in electric vehicles on energy consumption across all Australian states through the development of a logistic model which was used to convert the underlying economic drivers of electric vehicles into an impact on market share and take-up of the technology. This was done by creating a model which values each of the attributes that drive the decision to adopt the technology and then to


33


apply an elasticity or measure of responsiveness of market share to each factor. More details of the process used is provided in section A.8.

Figure 28 provides the incremental demand added to the Baseline under the EV sensitivity. It was assumed that the additional energy would not affect the peak demand projection (assumes all charging occurs during off-peak periods or is time-shifted so as not to impact peak periods).

Figure 28 Incremental demand for EV sensitivity

Source: ACIL Allen

Table 14 and Figure 29 provide the emission outcomes under the EV sensitivity. Aggregate emissions increase by around 44 Mt CO2-e over the period to 2034-35. This equates to a marginal emissions intensity for the incremental demand of 0.57 tonnes CO2-e/MWh sent-out. This figure is well below the average intensity for the system as much of the incremental generation is met by baseload gas (76%). Only 25% of the incremental demand is met by coal-fired power as coal-fired generators are operating at close to maximum output by the time EV demand becomes material.

Table 14 Summary of emission outcomes by fuel type (Mt CO2-e): EV sensitivity


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 117.5 143.8 -16.3% 19.0% 22.4% 856.3 1,915.1

Brown coal 69.0 59.6 61.1 65.4 -13.7% 2.4% 7.2% 506.4 995.8

Peaking Gas 5.5 5.0 1.1 1.3 -9.6% -78.3% 24.6% 17.1 14.7

Baseload Gas 13.6 17.4 17.9 27.9 28.3% 2.9% 55.9% 135.3 339.5

Cogen 3.3 3.9 1.9 1.4 18.9% -51.7% -26.3% 23.2 24.3

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -4.2% 9.1% 17.4 35.3

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 201.8 242.5 -11.4% 7.9% 20.2% 1,556.7 3,326.3

Source: ACIL Allen


34


Figure 29 Australian emissions by fuel type: EV sensitivity change from Baseline scenario


4.4 Aggregate High/Low scenariosThe aggregate High/Low scenarios utilise a combination of the previous sensitivities to create the upper and lower bounds for emissions.

The Aggregate High sensitivity is comprised of: High demand, Low fuel prices and EV uptake. Conversely, the Aggregate Low sensitivity uses the Low demand and High fuel price inputs.

Figure 30 shows the emission outcomes under the Aggregate High/Low sensitivities in comparison with the Baseline scenario. The sensitivities provide an almost uniform spread around the central case, with emissions in 2019-20 being 6.7 Mt CO2-e (3.3%) higher under the High case and 6.7 Mt CO2-e (3.3%) lower under the Low case. Cumulative emissions (2012-13 to 2019-20) are 1.5% higher and 1.3% lower respectively.

Figure 30 Aggregate High/Low sensitivity emission outcomes: Australia

Source: ACIL Allen


35


Table 15 provides summary figures for emissions by fuel type for selected years under the Aggregate High sensitivity.

Table 15 Summary of emission outcomes by fuel type (Mt CO2-e): Aggregate High sensitivity


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 120.7 142.0 -16.3% 22.3% 17.6% 871.6 1,959.8

Brown coal 69.0 59.6 61.6 65.4 -13.7% 3.4% 6.2% 507.7 997.9

Peaking Gas 5.5 5.0 1.1 1.9 -9.6% -77.6% 70.6% 17.2 30.1

Baseload Gas 13.6 17.4 20.3 43.7 28.3% 16.4% 115.6% 141.1 467.3

Cogen 3.3 3.9 2.0 1.7 18.9% -48.7% -14.8% 23.6 30.8

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -4.0% 9.1% 17.5 35.3

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 208.0 257.3 -11.4% 11.2% 23.7% 1,579.4 3,522.8

Source: ACIL Allen

As shown in Figure 31, the increased emissions under the Aggregate High sensitivity come from increased output from coal-fired generation as excess capacity is taken up in the market, supplemented by additional gas-fired generation to meet the additional demand growth.

Relative to the Baseline scenario, emissions increase in a linear fashion from 2014-15 to 2034-35 reaching around 21 Mt CO2-e per annum in the final year

The additional generation capacity required to meet the higher demands is comprised of 6,625 MW of gas-fired capacity (6,720 MW of baseload/intermediate CCGT, offset by a reduction of 95 MW of peaking plant). The scenario results in less rooftop solar PV being developed (approximately 2,100 MW less by 2034-35) due to the scenario having generally lower retail prices for consumers. This is caused by a combination of the low fuel prices flowing through to wholesale price outcomes7 and the higher demand reducing unit network charges.

Australia’s emissions intensity is slightly lower under this case (0.70 tonnes CO2-e/MWh sent-out compared with 0.75 tonnes CO2-e/MWh sent-out by 2034-35 under the Baseline scenario) due to the increased gas-fired generation in the long-term.

7 Some of the impact of the lower fuel costs on wholesale prices are negated by the higher demand, but the overall impact is negative.


36


Figure 31 Australian emissions by fuel type: Aggregate High sensitivity change from Baseline scenario


Table 16 and Figure 32 provides the corresponding emission figures by fuel under the Aggregate Low sensitivity. Under low demand conditions, the market develops very little additional capacity aside from that required to meet the Real 20% LRET and rooftop solar PV.

The change in emission outcomes shows an increasing reduction in output from coal and gas-fired generation. By 2034-35 coal-fired generation is around 17 TWh lower and gas 16 TWh lower than under the baseline scenario. By comparison, renewable generation in 2034-35 is slightly lower overall: a reduction in wind output (-1.5 TWh) is partially offset by higher solar PV output (+0.5 TWh). This is due to the LRET being set as a fixed target in GWh and uptake of rooftop solar PV being largely unaffected by wholesale demand conditions.8

Emissions under the Aggregate Low sensitivity are 6.7 Mt CO2-e lower in 2019-20 and 22.8 Mt CO2-e lower in 2034-35 relative to the baseline assumptions. Cumulative emissions for the period 2012-13 to 2019-20 are 1.3% lower relative to the baseline.

8 The lower wholesale electricity demand increases the network tariff within retail prices (as regulated revenues are spread across fewer GWh) and therefore improves the attractiveness of rooftop solar PV for residential and commercial customers.


37


Table 16 Summary of emission outcomes by fuel type (Mt CO2-e): Aggregate Low sensitivity


2008-09 2012-13 2019-20 2034-35 2008-09 to 2012-13

2012-13 to 2019-20

2019-20 to 2034-35

2012-13 to 2019-20

2020-21 to 2034-35

Black coal 117.9 98.7 112.1 124.9 -16.3% 13.6% 11.4% 841.7 1,729.5

Brown coal 69.0 59.6 61.0 65.4 -13.7% 2.4% 7.2% 506.3 995.1

Peaking Gas 5.5 5.0 0.8 0.6 -9.6% -83.1% -26.9% 16.8 10.0

Baseload Gas 13.6 17.4 16.5 18.7 28.3% -5.1% 12.9% 128.4 268.2

Cogen 3.3 3.9 1.8 1.2 18.9% -53.9% -30.8% 23.0 20.7

Liquid fuel 1.8 2.3 2.2 2.4 27.5% -4.2% 9.2% 17.4 35.3

Hydro 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Wind 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Solar 0.0 0.0 0.0 0.0 0.0% 0.0% 0.0% 0.0 0.0

Biothermal 0.1 0.1 0.1 0.1 11.6% -0.1% 0.0% 0.9 1.6

Geothermal 0.0 0.0 0.0 0.0 - - - 0.0 0.0

Total 211.2 187.0 194.7 213.4 -11.4% 4.1% 9.6% 1,534.5 3,060.5

Source: ACIL Allen

Figure 32 Australian emissions by fuel type: Aggregate Low sensitivity change from Baseline scenario



38


Appendix A Input assumptions

A.1 Macro inputsThe following sections outline the key assumptions used in the Baseline scenario. Variations to the key assumptions were also undertaken through the modelling of sensitivities (see section 2.3 for more information of these assumptions).

A.1.1 Electricity demand

Under a separate consultancy, pitt&sherry were engaged to provide the Department with electricity demand forecasts to be used within this project. This study focused primarily on projected electricity demand on main grids.

The pitt&sherry work provided the following demand components:

NEM native energy (as defined by AEMO) for each NEM region including network losses

SWIS energy (in accordance with the definition used by the IMO)

Energy delivered from the DKIS (as defined by the NT Utilities Commission)

Energy forecasts covering a number of for mini grids/off-grid users which covered only residential, business and community consumption.

ACIL Allen provided pitt&sherry with projections of rooftop solar PV for each jurisdiction under the Baseline scenario assumptions and these were incorporated to arrive a grid-level energy values.

Peak demand forecasts were also provided by pitt&sherry on a P50 basis for NSW, QLD, VIC, SA and the SWIS.

ACIL Allen was also tasked with providing the balance of the demand assumptions for those demand components not covered by the pitt&sherry work, primarily additional large industrial, mini-grid, off-grid and grid-exempt demand’, which includes approximately 24 TWh of additional demand as shown in Table A1. ACIL Allen’s analysis was verified through interrogation of NGER data for electricity generation in 2011-12 and 2012-13.

Table A17 Breakdown of electricity demand comparison for 2012-13ACIL Allen Pitt&Sherry

NEM AEMO Native Energy 189,663 187,511

NEM Scheduled and semi-scheduled 183,440

NEM Significant non-scheduled 3,105

NEM Small non-scheduled 3,118

NEM Market exempt 4,661

SWIS 22,352 18,104

SWIS market 17,706

SWIS Non-market 4,646

DKIS 1,540 1,585

NWIS 2,761 496

Mt Isa 2,188 370

Off-grid 9,984 1,204

Rooftop solar PV 3,520 3,502

Total 236,668 212,773

Note: GWh sent-out at source of generation (i.e. includes network losses). Includes rooftop solar PV; excludes some generation for calibration purposesSource: ACIL Allen analysis


A-1


Table A18 and Table A19 summarise the Baseline demand projection used in the modelling broken down by jurisdiction and grid respectively.

Table A18 Baseline electricity demand projection for selected years by jurisdiction (GWh)

2008-09 2012-13 2014-15 2019-20 2024-25 2029-30 2034-35

NSW/ACT 71,597 64,900 68,474 70,730 74,114 78,900 84,854

VIC 53,738 51,278 44,248 47,098 50,901 55,374 61,010

QLD 59,780 58,591 57,731 72,089 75,676 80,830 85,956

SA 14,180 13,159 13,632 13,938 14,029 14,324 15,094

TAS 8,683 13,054 10,650 10,574 10,543 10,500 10,484

WA 28,091 32,333 33,628 41,550 45,448 49,041 52,434

NT 3,189 3,353 2,668 5,192 5,619 5,974 6,307

Total Australia 239,258 236,668 231,032 261,170 276,329 294,943 316,141


Table A19 Baseline electricity demand projection for selected years by grid (GWh)

2008-09 2012-13 2014-15 2019-20 2024-25 2029-30 2034-35

NEM 204,697 197,382 190,881 209,101 219,472 233,512 250,538

SWIS 21,088 22,745 23,434 27,109 29,434 32,283 35,493

NWIS 1,551 2,783 2,985 3,546 3,848 4,151 4,488

DKIS 1,437 1,547 1,505 1,613 1,890 2,180 2,592

Mt Isa 1,914 2,198 2,305 2,488 2,735 3,129 3,535

Off grid 8,571 10,014 9,922 17,313 18,951 19,689 19,495

Total Australia 239,258 236,668 231,032 261,170 276,329 294,943 316,141


Various methods were used to project the balance of demand components, including relying on industry output projections from the Department for alumina, iron ore, coal and LNG (developed in consultation with Treasury and the Department of Industry) and various other projections from BREE.

A key component for off-grid energy is electricity generated associated with liquefaction facilities of LNG developments. These were undertaken with reference to electricity intensity indices provided by the Department for each new project.

In aggregate the demand projections under the Baseline scenario aligned reasonably closely with those used as the Core assumptions from the 2014 RET Review as shown in Figure A33. The Baseline demand is around 4.7 TWh higher by 2019-20. This gap widens in the longer-term to be 13.1 TWh higher by 2034-35. Owing to a range of other differences in input assumptions and scenario definitions, we would not expect outcomes to be the same as the RET Review process.


A-2


Figure A33Comparison of Australian energy demand to 2020

Source: ACIL Allen

A.1.2 Foreign exchange rates

The assumed exchange rate series used was provided in confidence by the Department based on a Treasury series. All foreign equipment is assumed to be priced in US dollars.

A.1.3 Commodity prices and wage indices

ACIL Allen utilise commodity prices and wage indices as modifiers to capital cost components. These typically include ‘Metals’ and domestic wage price indices. For this work, the Department provided a wage price index (sourced from the 2014-15 Budget) as shown in Figure A34. This is applied to domestic labour components. Due to the inability to provide an update for the metals index, its influence was removed from the capital cost projections.

Figure A34Real wage index used

Note: 2009-10 = 100Source: ACIL Allen based on 2013 Emissions Projections Study


A-3


A.1.4 Fuel costs

The gas prices used in this work were based on the International Energy Agency’s (IEA) 2014 World Energy Outlook. Within this publication the IEA produces two gas price series: the New Policies Scenario (baseline scenario) and a Gas Price Convergence Scenario in which global prices converged on the basis of increased exports from the United States.

The Department has provided equivalent domestic gas price based on a transition to export netback prices for each of these series as shown in Figure A35. 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 next 2-3 years.

Figure A35Calculated netback gas price for Australian gas producers

Source: ACIL Allen based on data from the Department of the Environment and WEO 2014

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). Forecast thermal coal price used for this purpose as shown in Figure A36. These prices are converted back into Australian dollars and adjusted as discussed above for use in the modelling.

Figure A36Assumed new entrant coal costs

Source: Department of the Environment


A-4


A.1.5 Carbon pricing

The Clean Energy Future Act 2011 (Commonwealth) implemented a multi-sectoral, nation-wide carbon price, with the core effect of imposing a cost on emitting greenhouse gases from a range of activities including fossil fuel combustion (e.g. in electricity generation) and fugitive emissions from gas production and coal mining. The scheme was in operation from 2012-13 (nominal $23/tonne CO2-e) through to 2013-14 (nominal $24.15/tonne CO2-e). The legislation has since been repealed, effective from 1 July 2014. The scenarios do not incorporate any other explicit carbon pricing policies in the future.

The Government’s Direct Action Plan, including the Emissions Reduction Fund and safeguard mechanism have not been modelled as policy settings have not been finalised.

A.1.6 Conventional coal entry

ACIL Allen’s standard assumption when undertaking market outlook studies is to restrict conventional coal (i.e. coal-based technologies which do not employ carbon capture and storage) from entry. Restricting coal from entering can be justified due to:

Community views and corporate sustainability policies

Potential difficulty obtaining generation licenses from State and Territory governments

Long-term risk of explicit carbon pricing being reintroduced

Difficulty in securing financing on a commercial basis due to these risks.

Under the Baseline scenario is it assumed that the above considerations make new conventional coal developments commercial unviable and hence are constrained from entry within the modelling.

Conventional coal entry will be permitted in scenarios which do not include supply-side policy measures based purely on the economics of the technology within the market. That is, the modelling assumes that under a no measures scenario, there is also no risk of measures being introduced.

A.2 Existing generatorsThe modelling incorporates a total of around 190 existing generators across the nine regions modelled as shown in Table A20. For the NEM, these generators represent those that are scheduled and semi-scheduled (i.e. those that report and participate in AEMO’s central dispatch functions). These are generally those generators with a nameplate capacity above 30 MW in the NEM and above 1 MW in the SWIS.

Non-scheduled, embedded ‘behind the meter’ and off-grid generation are handled outside of PowerMark LT. For the most part, non-scheduled and embedded generators (aside from rooftop solar PV – see section A.6) are held constant at current levels.

For the SWIS, the generators and their capacity corresponds with capacity offered to the Independent Market Operator (IMO) as part of the wholesale markets net pool functions. This means that capacity and energy related to own-use consumption (most notably from cogeneration projects) is not included explicitly and is handled outside the modelling.

For NWIS, DKIS and Mt Isa regions no formal market structure exists and generators include all major grid connected plants.


A-5


Table A20 Existing and committed generators: type, capacity and life

Region Generator Plant type Fuel type CommissionedTechnical

Life (Years)

Technical

Retirement

Year

Capacity

(gross MW)

NSW

AGL SF PV Broken Hill Solar PV Solar 2014 30 2044 53

AGL SF PV Nyngan Solar PV Solar 2014 30 2044 106

Bayswater Subcritical pf Black coal 1983 53 2036 2,720

Bendeela Pumps Pump n/a 1977 150 2127 240

Blowering Hydro Hydro 1969 150 2119 80

Boco Rock WF Wind turbine Wind 2014 25 2036 113

Colongra OCGT Natural gas 2009 30 2039 664

Eraring Subcritical pf Black coal 1983 50 2033 2,880

Gullen Range WF Wind turbine Wind 2014 25 2036 165.5

Gunning WF Wind turbine Wind 2011 25 2036 47

Guthega Hydro Hydro 1955 150 2105 60

Hume NSW Hydro Hydro 1957 150 2107 29

Hunter Valley GT OCGT Liquid fuel 1988 30 2018 51

Liddell Subcritical pf Black coal 1972 60 2032 2,100

Mt Piper Subcritical pf Black coal 1993 50 2043 1,340

Redbank Subcritical pf Black coal 2001 50 2051 150

Shoalhaven Bendeela Hydro Hydro 1977 150 2127 240

Smithfield CCGT Natural gas 1997 30 2027 176

Tallawarra CCGT Natural gas 2009 30 2039 430

Taralga WF Wind turbine Wind 2014 25 2036 106.8

Tumut 1 Hydro Hydro 1959 150 2109 616

Tumut 3 Hydro Hydro 1973 150 2123 1,500

Tumut 3 Pumps Pump n/a 1973 150 2123 400

Uranquinty OCGT Natural gas 2009 30 2039 664

Vales Point B Subcritical pf Black coal 1978 50 2028 1,320

Wallerawang C Subcritical pf Black coal 1978 45 2023 960

Woodlawn WF Wind turbine Wind 2011 25 2036 48

QLD

Barcaldine CCGT Natural gas 1996 30 2026 55

Barron Gorge Hydro Hydro 1963 150 2113 60

Braemar 1 OCGT Natural gas 2006 30 2036 504

Braemar 2 OCGT Natural gas 2009 30 2039 459

Callide B Subcritical pf Black coal 1989 50 2039 700

Callide C Supercritical pf Black coal 2001 50 2051 810

Collinsville Subcritical pf Black coal 1998 30 2028 190

Condamine CCGT Natural gas 2009 30 2039 140

Darling Downs CCGT Natural gas 2010 30 2040 630

Gladstone Subcritical pf Black coal 1980 50 2030 1,680

Kareeya Hydro Hydro 1958 150 2108 81

Kogan Creek Supercritical pf Black coal 2007 50 2057 750

Mackay GT OCGT Liquid fuel 1975 45 2020 34

Millmerran Supercritical pf Black coal 2002 50 2052 851

Mt Stuart OCGT Liquid fuel 1998 40 2038 418

Oakey OCGT Natural gas 2000 30 2030 282

Roma OCGT Natural gas 1999 30 2029 80

Stanwell Subcritical pf Black coal 1995 50 2045 1,440

Swanbank E CCGT Natural gas 2002 30 2032 385

Tarong Subcritical pf Black coal 1985 50 2035 1,400

Tarong North Supercritical pf Black coal 2002 50 2052 443

Townsville CCGT Natural gas 2005 30 2035 240

Wivenhoe Hydro Hydro 1984 150 2134 500

Wivenhoe Pump Pump n/a 1984 150 2134 480

Yarwun Cogeneration Natural gas 2010 30 2040 168

SA

Bluff WF Wind turbine Wind 2011 25 2036 53

Clements Gap WF Wind turbine Wind 2008 25 2033 57

Dry Creek OCGT Natural gas 1973 45 2018 156

Hallett OCGT Natural gas 2002 30 2032 200

Hallett 2 WF Wind turbine Wind 2008 25 2033 71

Hallett WF Wind turbine Wind 2007 25 2032 95


A-6



Life (Years)

Technical

Retirement

Year

Capacity

(gross MW)

Ladbroke Grove OCGT Natural gas 2000 30 2030 80

Lake Bonney 2 WF Wind turbine Wind 2008 25 2033 159

Lake Bonney 3 WF Wind turbine Wind 2010 25 2035 39

Mintaro OCGT Natural gas 1984 30 2014 90

North Brown Hill WF Wind turbine Wind 2011 25 2036 132

Northern Subcritical pf Brown coal 1985 50 2035 530

Osborne CCGT Natural gas 1998 30 2028 180

Pelican Point CCGT Natural gas 2000 35 2035 485

Playford B Subcritical pf Brown coal 1960 60 2020 231

Port Lincoln OCGT Liquid fuel 1999 30 2029 74

Quarantine OCGT Natural gas 2002 30 2032 216

Snowtown 2 North WF Wind turbine Wind 2014 25 2039 144

Snowtown 2 South WF Wind turbine Wind 2014 25 2039 126

Snowtown WF Wind turbine Wind 2008 25 2033 99

Snuggery OCGT Liquid fuel 1997 30 2027 63

Torrens Island A Steam turbbine Natural gas 1967 52 2019 480

Torrens Island B Steam turbine Natural gas 1977 50 2027 800

Waterloo WF Wind turbine Wind 2011 25 2036 111

TAS

Bastyan Hydro Hydro 1983 150 2133 80

Bell Bay Subcritical pf Natural gas 1971 38 2009 240

Bell Bay Three OCGT Natural gas 2006 30 2036 120

Cethana Hydro Hydro 1971 150 2121 85

Devils Gate Hydro Hydro 1969 150 2119 60

Fisher Hydro Hydro 1973 150 2123 43

Gordon Hydro Hydro 1978 150 2128 432

John Butters Hydro Hydro 1992 150 2142 144

Lake Echo Hydro Hydro 1956 150 2106 32

Lemonthyme_Wilmot Hydro Hydro 1970 150 2120 82

Liapootah_Wayatinah_Catagunya Hydro Hydro 1960 150 2110 170

Mackintosh Hydro Hydro 1982 150 2132 80

Meadowbank Hydro Hydro 1967 150 2117 40

Musselroe WF Wind turbine Wind 2013 25 2038 168

Poatina Hydro Hydro 1964 150 2114 300

Reece Hydro Hydro 1986 150 2136 231

Tamar Valley CCGT Natural gas 2010 30 2040 200

Tamar Valley GT OCGT Natural gas 2009 30 2039 58

Tarraleah Hydro Hydro 1938 150 2088 90

Trevallyn Hydro Hydro 1955 150 2105 80

Tribute Hydro Hydro 1994 150 2144 83

Tungatinah Hydro Hydro 1953 150 2103 125

VIC

Anglesea Subcritical pf Brown coal 1969 52 2021 160

Bairnsdale OCGT Natural gas 2001 30 2031 92

Dartmouth Hydro Hydro 1960 150 2110 158

Eildon Hydro Hydro 1957 150 2107 120

Energy Brix Subcritical pf Brown coal 1960 58 2018 195

Hazelwood Subcritical pf Brown coal 1968 63 2031 1,640

Hume VIC Hydro Hydro 1957 150 2107 29

Jeeralang A OCGT Natural gas 1979 50 2029 228

Jeeralang B OCGT Natural gas 1980 50 2030 255

Laverton North OCGT Natural gas 2006 30 2036 312

Loy Yang A Subcritical pf Brown coal 1986 50 2036 2,180

Loy Yang B Subcritical pf Brown coal 1995 50 2045 1,050

Macarthur WF Wind turbine Wind 2013 25 2038 420

McKay Hydro Hydro 1980 150 2130 300

Mortlake OCGT Natural gas 2011 40 2051 566

Mt Mercer WF Wind turbine Wind 2014 25 2039 131

Murray Hydro Hydro 1968 150 2118 1,500

Newport Steam turbine Natural gas 1980 50 2030 500

Oaklands Hill WF Wind turbine Wind 2011 25 2036 63

Somerton OCGT Natural gas 2002 30 2032 160


A-7



Life (Years)

Technical

Retirement

Year

Capacity

(gross MW)

Valley Power OCGT Natural gas 2002 30 2032 300

West Kiewa Hydro Hydro 1956 150 2106 62

Yallourn Subcritical pf Brown coal 1980 55 2035 1,538

SWIS

Albany Wind turbine Wind 2001 25 2026 22

Alcoa Kwinana Cogen Cogeneration Natural gas 1998 30 2028 5

Alcoa Pinjarra Cogen Cogeneration Natural gas 1985 35 2020 10

Alcoa Wagerup Cogen Cogeneration Natural gas 1990 30 2020 25

Bluewaters Subcritical pf Black coal 2009 40 2049 441

BP Cogen Cogeneration Natural gas 1996 30 2026 81

Canning/Melville LFG Reciprocating engine Landfill gas 2007 15 2022 9

Cockburn CCGT Natural gas 2003 30 2033 246

Collgar WF Wind turbine Wind 2012 25 2037 206

Collie Subcritical pf Black coal 1999 40 2039 333

Emu downs Wind turbine Wind 2006 25 2031 80

Geraldton OCGT Distillate 1973 40 2013 21

Grasmere Wind turbine Wind 2012 25 2037 14

Greenough River Solar PV Solar 2012 30 2042 10

Kalgoorlie OCGT Distillate 1990 30 2020 63

Kalgoorlie Nickel OCGT Natural gas 1996 30 2026 10

Kemerton OCGT Natural gas 2005 30 2035 310

Kwinana A Steam turbine Natural gas 1970 41 2011 245

Kwinana B Steam turbine Natural gas 1974 34 2008 0

Kwinana C Steam turbine Natural gas 1976 39 2015 385

Kwinana GT OCGT Distillate 1975 40 2015 21

Kwinana HEGT OCGT Natural gas 2011 30 2041 201

Muja A&B Subcritical pf Black coal 1968 40 2008 240

Muja C Subcritical pf Black coal 1981 40 2021 398

Muja D Subcritical pf Black coal 1986 40 2026 454

Mumbida Wind turbine Wind 2012 25 2037 55

Mungarra OCGT Natural gas 1991 30 2021 113

Namarkkon OCGT Distillate 2012 30 2042 70

Neerabup Peaker OCGT Natural gas 2009 30 2039 330

Newgen Power CCGT Natural gas 2007 30 2037 314

Parkeston SCE OCGT Natural gas 1996 30 2026 68

Pinjar A B OCGT Natural gas 1990 30 2020 228

Pinjar C OCGT Natural gas 1992 30 2022 233

Pinjar D OCGT Natural gas 1996 30 2026 124

Pinjarra Alinta Cogen Cogeneration Natural gas 2007 30 2037 280

Tesla (various sites) OCGT Distillate 2012 30 2042 40

Tiwest Cogen Cogeneration Natural gas 1999 30 2029 37

Wagerup Alinta Peaker OCGT Distillate 2007 30 2037 323

Walkaway Wind turbine Wind 2005 25 2030 89

Western Energy Peaker OCGT Natural gas 2011 30 2041 106

Worsley Cogeneration Black coal 1990 40 2030 0

Worsley SWCJV Cogeneration Natural gas 2000 25 2025 116

NWIS

Burrup Peninsula OCGT Natural gas 2006 30 2036 74

Cape Lambert CCGT Natural gas 1996 30 2026 105

Cape Preston CCGT Natural gas 2009 30 2039 450

Dampier Steam Turbine Natural gas 2000 50 2050 120

Dampier C Steam Turbine Natural gas 1970 50 2020 120

Karratha Steam turbine Natural gas 2005 50 2055 44

Karratha ATCO OCGT Natural gas 2010 30 2040 86

Paraburdoo Reciprocating Engine Liquid fuel 1985 30 2015 20

Port Hedland OCGT Natural gas 1997 30 2027 180

South Hedland CCGT Natural gas 2017 30 2047 150

DKIS

Berrimah OCGT Liquid fuel 1979 30 2009 30

Channel Island u1-3 OCGT Natural gas 1986 30 2016 95

Channel Island u4-6 CCGT Natural gas 1998 30 2028 95

Channel Island u7 OCGT Natural gas 2006 30 2036 42

Channel Island u8-9 OCGT Natural gas 2012 30 2042 90


A-8



Life (Years)

Technical

Retirement

Year

Capacity

(gross MW)

Katherine OCGT Natural gas 1987 30 2017 34

LMS Shoal Bay Reciprocating engine Landfill gas 2005 15 2020 1

Pine Creek CCGT CCGT Natural gas 1989 30 2019 27

Weddell OCGT Natural gas 2008 30 2038 128

Mt Isa

APA Xstrata OCGT OCGT Natural gas 2008 30 2038 30

Diamantina CCGT CCGT Natural gas 2014 30 2044 242

Diamantina OCGT OCGT Natural gas 2014 30 2044 60

Ernest Henry Reciprocating Engine Liquid fuel 1997 30 2027 32

Mica Creek A CCGT CCGT Natural gas 2000 30 2030 103

Mica Creek A GT OCGT Natural gas 2000 30 2030 132

Mica Creek B OCGT Natural gas 2000 30 2030 35

Mica Creek C CCGT Natural gas 2000 30 2030 55

Mt Isa Mines Station Steam turbine Natural gas 1974 50 2024 38

Phosphate Hill OCGT Natural gas 1999 30 2029 42

Source: ACIL Allen

Table A21 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.

Thermal efficiency and Scope 1 emission factors have been calibrated based on Clean Energy Regulator publicly released emissions data. This was undertaken on the context of the assumptions development work for AEMO in mid-2014.9

Table A21 Existing and committed generators: efficiency, emissions and O&M costs

Region Generator

Thermal

efficiency

HHV (%)

sent-out

Auxiliaries

(%)

Scope 1

emission

factor (kg

CO2-e/GJ of

fuel)

Scope 1

emission

intensity

(tonnes

CO2-e/MWh

sent-out)

FOM

($/MW

gross/year)

VOM

($/MWh

sent-out)

Forced

outage rate

(%)

Planned

outage rate

(%)

MLF

NSW

AGL SF PV Broken Hill 0.0% 0 0.000 34,833 0 0.00% 0.00% 1.1026

AGL SF PV Nyngan 0.0% 0 0.000 34,833 0 0.00% 0.00% 1.1026

Bayswater 35.9% 6.0% 91.8 0.921 46,039 1.11 3.00% 4.00% 0.9653

Bendeela Pumps 0.0% 0 0.000 48,858 8.67 0.00% 0.00% 0.9877

Blowering 0.0% 0 0.000 48,858 4.82 0.00% 4.00% 0.9368

Boco Rock WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 1.0156

Colongra 32.0% 1.0% 58.9 0.663 12,214 9.38 1.50% 0.00% 0.9855

Eraring 35.4% 6.0% 90.2 0.917 46,039 1.11 3.00% 4.00% 0.9850

Gullen Range WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9667

Gunning WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9609

Guthega 0.0% 0 0.000 48,858 6.74 0.00% 4.00% 0.8987

Hume NSW 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9232

Hunter Valley GT 28.0% 1.0% 69.7 0.896 12,214 8.93 2.50% 0.00% 0.9755

Liddell 33.8% 5.0% 90.9 0.968 48,858 1.11 3.00% 8.00% 0.9663

Mt Piper 37.0% 5.0% 93.3 0.908 46,039 1.23 3.00% 4.00% 0.9698

Redbank 29.3% 8.0% 105.0 1.290 46,509 1.11 4.00% 4.00% 0.9744

Shoalhaven Bendeela 0.0% 0 0.000 48,858 8.67 0.00% 4.00% 0.9737

Smithfield 41.0% 5.0% 57.7 0.506 23,489 2.23 2.50% 2.00% 1.0026

Tallawarra 50.0% 3.0% 51.3 0.369 30,249 1.1 3.00% 2.00% 0.9839

Taralga WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9860

Tumut 1 0.0% 0 0.000 48,858 6.74 0.00% 4.00% 0.9363

Tumut 3 0.0% 0 0.000 48,858 10.6 0.00% 4.00% 0.9309

Tumut 3 Pumps 0.0% 0 0.000 48,858 0 0.00% 0.00% 0.9490

Uranquinty 32.0% 1.0% 52.2 0.588 12,214 9.38 1.50% 0.00% 0.8675

Vales Point B 35.4% 5.0% 87.3 0.887 46,039 1.11 3.00% 8.00% 0.9865

9 See http://www.aemo.com.au/Electricity/Planning/Related-Information/~/media/Files/Other/planning/Emissions%202014/20140411_Emissions_report_V2.ashx


A-9

http://www.aemo.com.au/Electricity/Planning/Related-Information/~/media/Files/Other/planning/Emissions%202014/20140411_Emissions_report_V2.ashx

http://www.aemo.com.au/Electricity/Planning/Related-Information/~/media/Files/Other/planning/Emissions%202014/20140411_Emissions_report_V2.ashx


Region Generator

Thermal

efficiency

HHV (%)

sent-out

Auxiliaries

(%)

Scope 1

emission

factor (kg

CO2-e/GJ of

fuel)

Scope 1

emission

intensity

(tonnes

CO2-e/MWh

sent-out)

FOM

($/MW

gross/year)

VOM

($/MWh

sent-out)

Forced

outage rate

(%)

Planned

outage rate

(%)

MLF

Wallerawang C 33.1% 7.0% 85.5 0.930 48,858 1.23 3.00% 8.00% 0.9729

Woodlawn WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9618

QLD

Barcaldine 28.0% 1.0% 78.6 1.010 23,489 2.23 2.50% 4.00% 0.9871

Barron Gorge 0.0% 0 0.000 48,858 10.6 0.00% 4.00% 1.0883

Braemar 1 30.0% 1.0% 48.0 0.576 12,214 7.33 1.50% 0.00% 0.9480

Braemar 2 30.0% 1.0% 49.8 0.598 12,214 7.33 1.50% 0.00% 0.9480

Callide B 36.1% 9.0% 94.2 0.939 46,509 1.12 4.00% 4.00% 0.9525

Callide C 36.5% 6.0% 96.6 0.952 46,509 2.54 6.00% 5.00% 0.9525

Collinsville 27.7% 10.0% 89.4 1.162 61,072 1.23 4.00% 2.00% 1.0389

Condamine 48.0% 3.0% 62.5 0.468 30,249 1.1 1.50% 4.00% 0.9519

Darling Downs 46.0% 6.0% 57.1 0.447 30,249 1.1 3.00% 4.00% 0.9480

Gladstone 35.2% 8.0% 90.4 0.924 48,858 1.11 4.00% 4.00% 0.9797

Kareeya 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 1.0870

Kogan Creek 37.5% 9.0% 94 0.902 45,099 1.17 4.00% 4.00% 0.9493

Mackay GT 28.0% 1.0% 69.7 0.896 12,214 8.4 1.50% 0.00% 1.0305

Millmerran 36.9% 6.0% 90.5 0.883 45,099 2.64 5.00% 8.00% 0.9532

Mt Stuart 30.0% 1.0% 80.5 0.966 12,214 8.4 2.50% 2.00% 1.0041

Oakey 32.6% 1.0% 82.1 0.907 12,214 8.93 2.00% 0.00% 0.9519

Roma 30.0% 1.0% 55.9 0.671 12,214 8.93 3.00% 0.00% 0.9404

Stanwell 36.4% 9.0% 95.7 0.946 46,039 2.99 2.50% 4.00% 0.9693

Swanbank E 47.0% 3.0% 51.9 0.397 30,249 1.1 3.00% 2.00% 0.9934

Tarong 36.2% 8.0% 94.0 0.935 46,509 6.98 3.00% 4.00% 0.9677

Tarong North 39.2% 6.0% 96.5 0.886 45,099 1.33 3.00% 4.00% 0.9678

Townsville 46.0% 3.0% 56.1 0.439 30,249 1.1 3.00% 2.00% 1.0292

Wivenhoe 0.0% 0 0.000 48,858 0 0.00% 4.00% 0.9891

Wivenhoe Pump 0.0% 0 0.000 28,187 0 0.00% 0.00% 0.9930

Yarwun 34.0% 2.0% 51.3 0.543 23,489 0 3.00% 0.00% 0.9837

SA

Bluff WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9795

Clements Gap WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9658

Dry Creek 26.0% 1.0% 58.4 0.808 12,214 8.93 3.00% 0.00% 1.0022

Hallett 24.0% 1.0% 57.2 0.859 12,214 8.93 1.50% 0.00% 0.9869

Hallett 2 WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9843

Hallett WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9869

Ladbroke Grove 30.0% 1.0% 46.4 0.557 12,214 3.34 3.00% 4.00% 0.9884

Lake Bonney 2 WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9665

Lake Bonney 3 WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9665

Mintaro 28.0% 1.0% 67.4 0.866 12,214 8.93 1.50% 0.00% 0.9879

North Brown Hill WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9795

Northern 34.9% 10.0% 108.8 1.122 51,676 1.11 5.00% 8.00% 0.9744

Osborne 42.0% 5.0% 61.7 0.529 23,489 4.72 3.00% 2.00% 0.9990

Pelican Point 48.0% 2.0% 54.0 0.405 30,249 1.1 3.00% 4.00% 0.9994

Playford B 21.9% 8.0% 91 1.496 65,770 2.79 10.00% 8.00% 0.9882

Port Lincoln 26.0% 8.0% 67.9 0.940 12,214 8.93 1.50% 0.00% 0.9108

Quarantine 32.0% 5.0% 64.4 0.725 12,214 8.93 2.50% 0.00% 0.9939

Snowtown 2 North WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9861

Snowtown 2 South WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9861

Snowtown WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9136

Snuggery 26.0% 3.0% 67.9 0.940 12,214 8.93 2.00% 0.00% 0.9944

Torrens Island A 27.6% 5.0% 50.0 0.653 36,666 2.05 4.50% 4.00% 1.0004

Torrens Island B 30.0% 5.0% 49.1 0.589 36,666 2.05 4.50% 4.00% 1.0004

Waterloo WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9819

TAS

Bastyan 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9396

Bell Bay 29.0% 2.5% 51.3 0.637 36,666 2.05 12.00% 8.00% 0.9999

Bell Bay Three 29.0% 1.0% 51.3 0.637 12,214 7.33 3.00% 0.00% 0.9999

Cethana 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9593

Devils Gate 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9639


A-10


Region Generator

Thermal

efficiency

HHV (%)

sent-out

Auxiliaries

(%)

Scope 1

emission

factor (kg

CO2-e/GJ of

fuel)

Scope 1

emission

intensity

(tonnes

CO2-e/MWh

sent-out)

FOM

($/MW

gross/year)

VOM

($/MWh

sent-out)

Forced

outage rate

(%)

Planned

outage rate

(%)

MLF

Fisher 0.0% 0 0.000 48,858 4.82 0.00% 4.00% 0.9645

Gordon 0.0% 0 0.000 48,858 4.82 0.00% 4.00% 0.9046

John Butters 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9405

Lake Echo 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9028

Lemonthyme_Wilmot 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9683

Liapootah_Wayatinah_Catagunya 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9702

Mackintosh 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9296

Meadowbank 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9644

Musselroe WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.8957

Poatina 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9681

Reece 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9317

Tamar Valley 48.0% 3.0% 56.4 0.423 30,249 1.1 3.00% 2.00% 0.9990

Tamar Valley GT 28.0% 1.0% 51.3 0.660 12,214 8.93 3.00% 2.00% 0.9999

Tarraleah 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9195

Trevallyn 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9745

Tribute 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9339

Tungatinah 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9122

VIC

Anglesea 27.2% 8.0% 92.4 1.223 124,962 1.11 3.00% 2.00% 0.9849

Bairnsdale 34.0% 1.0% 50.0 0.530 12,214 2.09 2.50% 0.00% 0.9745

Dartmouth 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 1.0097

Eildon 0.0% 0 0.000 48,858 8.67 0.00% 4.00% 1.0078

Energy Brix 24.0% 15.0% 109.5 1.643 93,957 2.05 2.50% 4.00% 0.9682

Hazelwood 22.0% 10.0% 94.8 1.551 131,539 1.11 3.50% 8.00% 0.9691

Hume VIC 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 0.9912

Jeeralang A 22.9% 3.0% 53.8 0.845 12,214 8.4 2.50% 0.00% 0.9647

Jeeralang B 22.9% 3.0% 53.8 0.845 12,214 8.4 2.50% 0.00% 0.9647

Laverton North 30.4% 1.0% 64.9 0.768 12,214 7.33 1.50% 2.00% 0.9972

Loy Yang A 27.2% 9.0% 96.3 1.274 122,144 1.11 3.00% 2.00% 0.9704

Loy Yang B 26.6% 8.0% 91.2 1.235 87,738 1.11 4.00% 2.00% 0.9704

Macarthur WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 0.9946

McKay 0.0% 0 0.000 48,858 6.74 0.00% 4.00% 1.0083

Mortlake 32.0% 1.0% 46.1 0.519 12,214 7.73 2.50% 0.00% 0.9922

Mt Mercer WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 1.0064

Murray 0.0% 0 0.000 48,858 5.78 0.00% 4.00% 1.0178

Newport 33.3% 5.0% 50.9 0.550 37,583 2.09 2.00% 4.00% 0.9953

Oaklands Hill WF 0.0% 0 0.000 32,083 0 0.00% 0.00% 1.0262

Somerton 24.0% 1.0% 49.1 0.736 12,214 8.93 1.50% 0.00% 0.9927

Valley Power 24.0% 1.0% 54.8 0.822 12,214 8.93 1.50% 0.00% 0.9704

West Kiewa 0.0% 0 0.000 48,858 6.74 0.00% 4.00% 1.0416

Yallourn 23.5% 10.0% 97.0 1.485 126,842 1.11 4.00% 4.00% 0.9538

SWIS

Albany 0.0% 0 0.000 42,000 1.05 0.00% 0.00% 1.0699

Alcoa Kwinana Cogen 30.0% 1.0% 51.3 0.616 25,000 0 3.80% 5.20% 1.0214

Alcoa Pinjarra Cogen 30.0% 1.0% 51.3 0.616 25,000 0 3.80% 5.20% 0.9951

Alcoa Wagerup Cogen 30.0% 1.0% 51.3 0.616 25,000 0 3.80% 5.20% 0.9868

Bluewaters 36.1% 7.5% 93.1 0.928 52,000 1.58 3.00% 4.90% 0.9992

BP Cogen 33.0% 2.0% 51.3 0.560 23,489 0 5.00% 4.10% 1.0246

Canning/Melville LFG 30.0% 0.0% 0 0.000 50,000 3.68 5.00% 0.00% 1.0296

Cockburn 48.0% 2.4% 51.3 0.385 30,249 4.73 4.20% 10.10% 1.0277

Collgar WF 0.0% 0 0.000 42,000 1.05 0.00% 0.00% 1.0146

Collie 36.0% 7.9% 93.1 0.931 52,000 1.58 3.20% 8.50% 0.9956

Emu downs 0.0% 0 0.000 42,000 1.05 0.00% 0.00% 0.9937

Geraldton 29.0% 0.5% 67.9 0.843 12,214 9.46 5.90% 9.00% 1.0507

Grasmere 0.0% 0 0.000 42,000 1.05 0.00% 0.00% 1.0699

Greenough River 0.1% 0 0.000 50,000 0 0.00% 0.00% 1.0227

Kalgoorlie 33.0% 0.5% 67.9 0.741 12,214 9.46 5.90% 4.10% 1.0782

Kalgoorlie Nickel 33.0% 0.5% 51.3 0.560 12,214 9.46 5.20% 4.70% 1.2253


A-11


Region Generator

Thermal

efficiency

HHV (%)

sent-out

Auxiliaries

(%)

Scope 1

emission

factor (kg

CO2-e/GJ of

fuel)

Scope 1

emission

intensity

(tonnes

CO2-e/MWh

sent-out)

FOM

($/MW

gross/year)

VOM

($/MWh

sent-out)

Forced

outage rate

(%)

Planned

outage rate

(%)

MLF

Kemerton 34.0% 0.5% 51.3 0.543 12,214 9.46 6.00% 7.90% 1.0079

Kwinana A 32.0% 9.0% 51.3 0.577 40,000 8.41 5.40% 14.80% 1.0201

Kwinana B 32.0% 9.0% 51.3 0.577 40,000 8.41 5.40% 14.80% 1.0201

Kwinana C 33.0% 4.0% 51.3 0.560 40,000 7.35 5.20% 9.90% 1.0201

Kwinana GT 32.0% 0.5% 67.9 0.764 12,214 9.46 5.20% 9.90% 1.0201

Kwinana HEGT 40.0% 0.5% 51.3 0.462 12,214 1.31 5.20% 4.10% 1.0201

Muja A&B 26.4% 8.5% 93.1 1.270 60,000 1.58 4.20% 10.00% 1.0000

Muja C 34.6% 8.0% 93.1 0.969 52,000 1.58 4.20% 9.90% 1.0000

Muja D 35.6% 8.0% 93.1 0.941 52,000 1.58 4.90% 9.90% 1.0000

Mumbida 0.0% 0 0.000 42,000 1.05 0.00% 0.00% 1.0353

Mungarra 29.0% 0.5% 51.3 0.637 12,214 9.46 5.20% 9.90% 1.0353

Namarkkon 30.0% 1.0% 67.9 0.815 12,214 9.46 4.00% 4.00% 1.0405

Neerabup Peaker 32.0% 2.0% 51.3 0.577 12,214 9.46 3.90% 2.20% 1.0379

Newgen Power 48.0% 2.0% 51.3 0.385 30,249 1.1 4.00% 3.30% 1.0224

Parkeston SCE 33.0% 0.5% 51.3 0.560 12,214 9.46 5.20% 4.90% 1.2012

Pinjar A B 29.0% 0.5% 51.3 0.637 12,214 9.46 5.20% 9.90% 1.0312

Pinjar C 29.0% 0.5% 51.3 0.637 12,214 9.46 5.20% 9.90% 1.0312

Pinjar D 29.0% 0.5% 51.3 0.637 12,214 9.46 5.20% 9.90% 1.0312

Pinjarra Alinta Cogen 34.1% 2.4% 51.3 0.542 25,000 0 3.90% 4.10% 1.0184

Tesla (various sites) 28.0% 1.0% 67.9 0.873 12,214 9.46 4.00% 4.00% 1.1229

Tiwest Cogen 32.0% 1.5% 51.3 0.577 25,000 0 5.90% 4.10% 1.0177

Wagerup Alinta Peaker 34.1% 0.5% 67.9 0.717 12,214 9.46 3.90% 4.10% 0.9868

Walkaway 0.0% 0 0.000 42,000 1.05 0.00% 0.00% 0.9560

Western Energy Peaker 32.0% 0.5% 51.3 0.577 12,214 9.46 5.20% 4.10% 1.0204

Worsley 28.0% 0.0% 93.1 1.197 25,000 0 4.80% 4.10% 0.9886

Worsley SWCJV 33.0% 2.0% 51.3 0.560 25,000 0 5.00% 4.10% 0.9846

NWIS

Burrup Peninsula 29.0% 2.0% 51.3 0.637 12,214 9.61 3.00% 8.00% 1.0000

Cape Lambert 45.0% 2.0% 51.3 0.410 30,249 1.1 3.00% 8.00% 1.0000

Cape Preston 50.0% 3.0% 51.3 0.369 30,249 1.1 3.00% 8.00% 1.0000

Dampier 30.0% 5.0% 51.3 0.616 40,000 2.25 3.00% 4.00% 1.0000

Dampier C 30.0% 5.0% 51.3 0.616 40,000 2.25 3.00% 4.00% 1.0000

Karratha 30.0% 5.0% 51.3 0.616 40,000 2.25 3.00% 4.00% 1.0000

Karratha ATCO 40.0% 2.0% 51.3 0.462 12,214 9.61 3.00% 8.00% 1.0000

Paraburdoo 29.0% 2.0% 67.9 0.843 13,000 9.61 3.00% 4.00% 1.0000

Port Hedland 29.0% 2.0% 51.3 0.637 12,214 9.61 3.00% 8.00% 1.0000

South Hedland 46.0% 2.0% 51.3 0.401 40,000 2.25 3.00% 4.00% 1.0000

DKIS

Berrimah 24.0% 1.0% 67.9 1.019 12,214 9.61 3.00% 8.00% 1.0000

Channel Island u1-3 27.0% 1.0% 51.3 0.684 12,214 9.61 3.00% 8.00% 1.0000

Channel Island u4-6 48.0% 2.0% 51.3 0.385 30,249 1.1 3.00% 8.00% 1.0000

Channel Island u7 37.0% 1.0% 51.3 0.499 12,214 9.61 3.00% 8.00% 1.0000

Channel Island u8-9 37.0% 1.0% 51.3 0.499 12,214 9.61 3.00% 8.00% 1.0000

Katherine 25.0% 1.0% 51.3 0.739 12,214 9.61 3.00% 8.00% 1.0000

LMS Shoal Bay 40.0% 2.0% 0 0.000 80,000 4 3.00% 5.00% 1.0000

Pine Creek CCGT 47.0% 2.0% 51.3 0.393 30,249 1.1 3.00% 8.00% 1.0000

Weddell 35.0% 1.0% 51.3 0.528 12,214 9.61 3.00% 4.00% 1.0000

Mt Isa

APA Xstrata OCGT 36.0% 1.0% 51.3 0.513 12,214 9.61 3.00% 8.00% 1.0000

Diamantina CCGT 48.0% 2.0% 51.3 0.385 30,249 1.05 3.00% 4.00% 1.0000

Diamantina OCGT 32.0% 2.0% 51.3 0.577 12,214 9.61 3.00% 5.00% 1.0000

Ernest Henry 29.0% 2.0% 67.9 0.843 13,000 9.61 3.00% 4.00% 1.0000

Mica Creek A CCGT 43.0% 2.0% 51.3 0.429 30,249 1.05 3.00% 8.00% 1.0000

Mica Creek A GT 27.0% 3.0% 51.3 0.684 12,214 9.61 3.00% 8.00% 1.0000

Mica Creek B 27.0% 3.0% 51.3 0.684 12,214 9.61 3.00% 8.00% 1.0000

Mica Creek C 43.0% 2.0% 51.3 0.429 30,249 9.61 3.00% 8.00% 1.0000

Mt Isa Mines Station 25.0% 1.0% 51.3 0.739 40,000 9.61 3.00% 8.00% 1.0000

Phosphate Hill 27.0% 3.0% 51.3 0.684 12,214 1.05 3.00% 8.00% 1.0000

Source: ACIL Allen


A-12


A.3 Energy constrained and intermittent generation

Hydro

Within PowerMark LT the annual output of hydro stations can be constrained explicitly to desired levels.10 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.

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.11

ACIL Allen incorporates representative solar PV output profiles for these projects which vary by time of day and month.

A.4 Mothballing, end of life and refurbishment

A.4.1 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

10 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.

11 Other smaller existing solar developments are treated as non-scheduled or embedded generation and are therefore handled outside of the PowerMark LT modelling.


A-13


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 generators 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 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 license 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.12

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 A22 summarises the plant that have had capacity withdrawn from the market in recent years.13

12 It has been suggested that some generators who have mothballed capacity are effectively deferring costs associated with rehabilitation.

13 It should be noted that after the finalisation of modelling inputs, the announced retirement of Redbank (NSW) and mothballing of Torren Island A (SA) have occurred, but these changes have not been incorporated into the input assumptions.


A-14


Table A22 Recently mothballed or retired units by NEM region

Portfolio Power Station Fuel type Capacity (MW) Public comments/assumptions on period affected

Queensland

Stanwell Tarong (2 units) Coal 700 Mothballed Oct 2012, back online late 2014, early 2015. I unit to return in Oct 2014 to replace Swanbank E

Stanwell Swanbank E Gas 385 Currently operating but due to be mothballed Oct 2014 due to fuel source being on-sold

Stanwell Swanbank B Coal 480Permanently retired, units decommissioned in April 2010, June 2010, 2011, and May 2012, due to the plant reaching the end of its operational life

RATCH Collinsville Coal 190 Mothballed December 2012, not expected to return to service

New South Wales

Delta Electricity Munmorah Coal 600 Retired July 2012

EnergyAustralia Wallerawang Coal 1000 Mothballed whole station March 2014 until viable for return to service. One unit permanently retired

Victoria

EnergyBrix Morwell (2 units) Coal 95 From July 2012 until viable

EnergyAustralia Yallourn (1 unit) Coal 350 From mid-2012 until viable

South Australia

Alinta Energy Northern Coal 540 Mothballed Apr-Sept until 2015

Alinta Energy Playford Coal 200 From March 2012 until viable; not expected to return to service

Source: ACIL Allen analysis

A.4.2 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.

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 A23 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.


A-15


Table A23 Refurbishment costs for incumbent plant

TechnologyEconomic life of

new plant (years)Refurbishment cost (% of new)

Additional life (% of original life)

Additional life from refurb (years)

CCGT 30 70% 100% 30

Cogeneration 30 70% 100% 30

OCGT 30 85% 100% 30

Solar PV 35 75% 100% 35

Steam turbine 50 25% 30% 15

Subcritical pf 50 25% 30% 15

Supercritical pf 50 25% 30% 15

Wind turbine 25 30% 100% 25

Source: ACIL Allen

A.4.3 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.14 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.

A.5 New entrant generatorsA range of new entrant generating technologies are made available within the modelling over the period to 2040. PowerMark LT determines a least cost plant mix for each modelled region on a dynamic inter-temporal basis.

New capacity is introduced to each region through the use of continuous capacity variables, that is, generation increments are not set to predetermined sizes and the model allows entry of any optimal increment.15

A range of cost and generation characteristics are required for each new entrant technology to solve the model in a way that minimises overall resource costs on a net present value basis. The key proposed inputs for each of these elements is discussed in the following sections.

Costs are presented in this section in real 2011-12 dollars to enable comparisons against the AETA 2012 work. Capital costs will be escalated to today’s dollars prior to modelling using published CPI.

A.5.1 Capital costs

Capital costs comprise one of the key inputs for long-term electricity sector modelling as capital is the largest cost component for most generation technologies.

The methodology employed for this study is to commence with a starting capital cost value (termed the ‘base’ capital cost) and break this down into its component parts: local labour; local equipment and commodities; and foreign equipment and commodities.

14 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.

15 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.


A-16


These component parts are then projected forward individually before being recombined into a final capital cost estimate. This process allows for the influences of learning rates (both foreign and local), labour costs, and exchange rates to be properly incorporated into the final cost estimates.

For the most part, the base capital cost estimates for most technologies were taken from the 2012 Australian Energy Technology Assessment (AETA) published by the Bureau of Resource and Energy Economics (BREE), with adjustments made in the 2013 update to the study.16

ACIL Allen has selected a sub-set of 29 of the 40 technologies examined within the AETA study. Technologies excluded include exotic coal-based technologies that do not employ carbon capture and storage (IGCC, oxy-fuel and direct injection), solar hybrids, offshore wind, landfill gas, bagasse and nuclear options.

Table A24 presents the proposed capital costs for each of the technologies considered within the modelling. The table also includes the headline splits for the cost components taken from the AETA study.

These capital costs are presented on an ‘overnight’ basis – interest during construction (IDC) and financing costs are excluded.17 For plants that employ carbon capture, the capital costs include capture and compression of CO2, but exclude transport and storage costs.

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 A37 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).

16 Note that the 2013 update only updated operating and maintenance costs for a select few technologies and no update to capital costs.

17 Interest during construction is added within the modelling.


A-17


Figure A37Base capital cost comparison with AETA

Source: ACIL Allen, BREE (AETA 2012)

Hydro-electric generation is not included as a model as a new entrant technology. This reflects the fact that few commercially viable large-scale hydro-electric sites remain in Australia for exploitation.


A-18

ACIL ALLEN CONSULTING

Table A24 Base capital costs and cost component splits

Category Technology2011-12 Base capital cost (A$/kW installed)

2011-12 Base capital cost (A$/kW net)

LabourForeign equipment and

commoditiesLocal equipment and

commodities

Coal PC Supercritical – Brown Coal 3,451 3,788 29% 38% 33%

PC Supercritical Black Coal 2,974 3,124 30% 39% 31%

PC Supercritical Black Coal (SWIS Scale) 3,192 3,381 31% 40% 29%

Natural gas CCGT 1,100 1,127 26% 56% 18%

CCGT SWIS Scale 1,078 1,111 26% 56% 18%

OCGT 800 808 11% 79% 10%

Solar CLFR 4,802 5,220 20% 55% 25%

CLFR with storage 8,550 9,500 25% 55% 20%

Parabolic trough 4,526 4,920 20% 55% 25%

Parabolic trough with storage 8,055 8,950 25% 55% 20%

Central Receiver 5,570 5,900 30% 55% 15%

Central Receiver with storage 7,477 8,308 25% 55% 20%

Solar PV Solar PV fixed 2,700 2,700 15% 70% 15%

Solar PV single axis tracking 3,180 3,180 15% 70% 15%

Solar PV dual axis tracking 4,730 4,730 15% 70% 15%

Wind On-shore Wind Farm 2,300 2,312 15% 72% 13%

Ocean/Wave 5,900 5,900 30% 40% 30%

Biomass Other Biomass Waste 4,400 5,000 18% 27% 55%

Geothermal Geothermal HSA 6,300 7,000 34% 23% 43%

Geothermal EGS 9,646 10,600 37% 17% 46%

CCS PC Supercritical with CCS – Brown Coal 5,902 7,766 29% 35% 36%

PC Supercritical with CCS – Bituminous Coal 4,559 5,434 29% 35% 36%

PC Oxy Combustion Supercritical with CCS 4,274 5,776 33% 35% 32%

CCGT with CCS 2,495 2,772 19% 67% 14%

IGCC with CCS – Bituminous Coal 4,984 7,330 27% 52% 21%

IGCC with CCS – Brown Coal 5,083 8,616 27% 52% 21%

CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 2,493 3,945 30% 30% 40%

PC Subcritical Black Coal - Retrofit CCS 1,611 2,244 30% 30% 40%

Existing CCGT with retrofit CCS 1,392 1,547 12% 78% 10%

Note: CCS capital costs are inclusive of capture, but exclude transport and storage costs as these are treated separately.Source: ACIL Allen, BREE (AETA 2012).


A-19


Learning rates

Learning rates are applied to the Base capital costs to reflect cost changes over time through technology and manufacturing improvements and learning by doing.

Learning rates for each major technology have been taken from CSIRO’s Global and Local Learning Model (GALLM) as part of the AETA 2012 study. For some technologies differential learning rates were provided for foreign and local content components and these have been applied to the respective foreign equipment and local equipment/local labour components respectively. Learning rates in the GALLM model are endogenous and respond to the rate of deployment of each technology both locally and internationally.

A complication in this process is the adjustments made by ACIL Allen to the base capital costs for solar PV and wind technologies from the AETA figures. As these represent a reduction in the starting base capital cost, it was decided that the learning rates should be reduced in the early years such that the capital cost for 2020 remained unchanged from the AETA work. The reported learning rates for these technologies in the period to 2020 will therefore differ due to the lower starting value.

Other adjustments

Adjustments are made to the some of the capital cost components based on macro assumptions as reported in section A.1. These include:

local labour costs are modified through the application of the real labour cost index

the composite metals price index is used to adjust 25% and 40% of the local and foreign equipment cost component respectively.

exchange rates are used to convert the foreign equipment and commodities cost component (which are projected in US dollars) back into Australian dollars.

A.5.2 Final capital costs

Table A25 presents the final capital costs for each of the technologies after all adjustments for learning, labour, metals and exchange rates are made. These are also shown graphically in Figure A38. Table A26 shows the average year-on-year percentage change in capital costs for each decade of the projection.


1


Table A25 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 3,826 3,797 3,820 3,896

PC Supercritical Black Coal 2,974 2,914 2,904 2,934 3,006

PC Supercritical Black Coal (SWIS Scale) 3,192 3,138 3,129 3,165 3,247

Natural gas CCGT 1,100 1,285 1,294 1,319 1,372

CCGT SWIS Scale 1,078 1,258 1,268 1,292 1,344

OCGT 800 948 931 919 912

Solar CLFR 4,802 2,849 2,318 2,347 2,413

CLFR with storage 8,550 5,102 4,166 4,241 4,394

Parabolic trough 4,526 2,685 2,185 2,212 2,275

Parabolic trough with storage 8,055 4,807 3,925 3,995 4,139

Central Receiver 5,570 3,344 2,740 2,803 2,925

Central Receiver with storage 7,477 4,462 3,643 3,709 3,842

Solar PV Solar PV fixed 2,700 1,995 1,580 1,010 961

Solar PV single axis tracking 3,180 2,350 1,861 1,190 1,132

Solar PV dual axis tracking 4,730 3,495 2,768 1,770 1,684

Wind On-shore Wind Farm 2,300 1,967 1,971 1,992 1,973

Ocean/Wave 5,900 6,735 3,227 3,322 3,288

Biomass Other Biomass Waste 4,400 4,806 4,794 4,646 4,794

Geothermal Geothermal HSA 6,300 6,962 7,086 7,379 7,809

Geothermal EGS 9,646 10,558 10,807 11,298 12,015

CCS PC Supercritical with CCS – Brown Coal 5,902 6,560 4,828 4,832 4,956

PC Supercritical with CCS – Bituminous Coal 4,559 5,067 3,729 3,733 3,828

PC Oxy Combustion Supercritical with CCS 4,274 4,774 3,527 3,547 3,661

CCGT with CCS 2,495 2,935 2,097 2,063 2,069

IGCC with CCS – Bituminous Coal 4,984 5,747 3,987 3,979 4,073

IGCC with CCS – Brown Coal 5,083 5,861 4,066 4,058 4,154

CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 2,493 2,742 2,055 2,059 2,114

PC Subcritical Black Coal - Retrofit CCS 1,611 1,772 1,328 1,331 1,366

Existing CCGT with retrofit CCS 1,392 1,663 1,178 1,150 1,141

Small CCGT Small Scale 1,886 2,202 2,218 2,261 2,352

Note: CCS capital costs are inclusive of capture, but exclude transport and storage costs which are treated separately.Source: ACIL Allen


2


Figure A38Final capital costs for new entrant technologies for selected years

Source: ACIL Allen


3


Table A26 Average real year-on-year capital cost change for each decade

Technology2011-12 to 2019-20

2019-20 to 2029-30

2029-30 to 2039-40

2039-40 to 2049-50

Coal PC Supercritical – Brown Coal 1.3% -0.1% 0.1% 0.2%

PC Supercritical Black Coal -0.3% 0.0% 0.1% 0.2%

PC Supercritical Black Coal (SWIS Scale) -0.2% 0.0% 0.1% 0.3%

Natural gas CCGT 2.0% 0.1% 0.2% 0.4%

CCGT SWIS Scale 2.0% 0.1% 0.2% 0.4%

OCGT 2.2% -0.2% -0.1% -0.1%

Solar CLFR -6.3% -2.0% 0.1% 0.3%

CLFR with storage -6.2% -2.0% 0.2% 0.4%

Parabolic trough -6.3% -2.0% 0.1% 0.3%

Parabolic trough with storage -6.2% -2.0% 0.2% 0.4%

Central Receiver -6.2% -2.0% 0.2% 0.4%

Central Receiver with storage -6.2% -2.0% 0.2% 0.4%

Solar PV Solar PV fixed -3.7% -2.3% -4.4% -0.5%

Solar PV single axis tracking -3.7% -2.3% -4.4% -0.5%

Solar PV dual axis tracking -3.7% -2.3% -4.4% -0.5%

Wind On-shore Wind Farm -1.9% 0.0% 0.1% -0.1%

Ocean/Wave 1.7% -7.1% 0.3% -0.1%

Biomass Other Biomass Waste 1.1% 0.0% -0.3% 0.3%

Geothermal Geothermal HSA 1.3% 0.2% 0.4% 0.6%

Geothermal EGS 1.1% 0.2% 0.4% 0.6%

CCS PC Supercritical with CCS – Brown Coal 1.3% -3.0% 0.0% 0.3%

PC Supercritical with CCS – Bituminous Coal 1.3% -3.0% 0.0% 0.3%

PC Oxy Combustion Supercritical with CCS 1.4% -3.0% 0.1% 0.3%

CCGT with CCS 2.1% -3.3% -0.2% 0.0%

IGCC with CCS – Bituminous Coal 1.8% -3.6% 0.0% 0.2%

IGCC with CCS – Brown Coal 1.8% -3.6% 0.0% 0.2%

CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 1.2% -2.8% 0.0% 0.3%

PC Subcritical Black Coal - Retrofit CCS 1.2% -2.8% 0.0% 0.3%

Existing CCGT with retrofit CCS 2.2% -3.4% -0.2% -0.1%

Small CCGT Small Scale 2.0% 0.1% 0.2% 0.4%

Source: ACIL Allen

A.5.3 Other new entrant parameters

Table A27 provides other technical parameters and cost assumptions for the new entrant technologies. For the most part these are aligned with the AETA 2012 and 2013 studies, with a few modifications.

Table A27 New entrant parameters

Category TechnologyThermal

efficiency (% HHV sent-out)

Auxiliary load (%)

FOM (Real 2011-12

$/MW/year)

VOM (Real 2011-12 $/MWh)

Coal PC Supercritical – Brown Coal 32.3% 8.9% 85,000 1.30

PC Supercritical Black Coal 41.9% 4.8% 52,000 1.30

PC Supercritical Black Coal (SWIS Scale) 41.4% 5.6% 55,500 8.00

Natural gas CCGT 49.5% 2.4% 33,000 1.20

CCGT SWIS Scale 49.3% 3.0% 10,000 4.00

OCGT 32.0% 1.0% 14,000 8.00

Solar CLFR 100.0% 8.0% 64,107 15.19

CLFR with storage 100.0% 10.0% 72,381 11.39


4


Category TechnologyThermal

efficiency (% HHV sent-out)

Auxiliary load (%)

FOM (Real 2011-12

$/MW/year)

VOM (Real 2011-12 $/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 maintenanceSource: ACIL Allen AETA (2012, 2013)

Both fixed and variable O&M charges are assumed to escalate at CPI (constant in real terms).

Table A28 shows the availability and construction profiles for each of the technologies. It is assumed that CCS based plant would not be available prior to 2030 based on slow international progress on demonstration plants.

Table A28 Technology availability and construction profiles

Category TechnologyFirst year

available

Construction period (years)

Yr1 Yr2 Yr3 Yr4

Coal PC Supercritical – Brown Coal 2018 4 35% 35% 20% 10%

PC Supercritical Black Coal 2018 4 35% 35% 20% 10%

PC Supercritical Black Coal (SWIS Scale) 2018 4 35% 35% 20% 10%

Natural gas CCGT 2017 2 60% 40% 0% 0%

CCGT SWIS Scale 2017 2 60% 40% 0% 0%

OCGT 2016 1 100% 0% 0% 0%

Solar CLFR 2018 3 50% 30% 20% 0%

CLFR with storage 2018 3 50% 30% 20% 0%

Parabolic trough 2018 3 50% 30% 20% 0%

Parabolic trough with storage 2018 3 50% 30% 20% 0%

Central Receiver 2018 3 20% 60% 20% 0%

Central Receiver with storage 2018 3 50% 30% 20% 0%

Solar PV Solar PV fixed 2016 2 70% 30% 0% 0%

Solar PV single axis tracking 2016 2 70% 30% 0% 0%

Solar PV dual axis tracking 2016 2 70% 30% 0% 0%


5


Category TechnologyFirst year

available

Construction period (years)

Yr1 Yr2 Yr3 Yr4

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 A29 shows the assumed economic life for each technology taken from AETA. As with incumbent generation, refurbishments are also applied to new entrants with the refurbishment capital cost expressed as a percentage of a new facility and resulting in a life extension expressed as a percentage of the original life. Installations can undergo multiple refurbishments within the projection horizon.

Table A29 Technology life and refurbishment costs

Category TechnologyEconomic life

(years)

Refurbishment cost (% of

new)

Additional life (% of original

life)

Additional life from refurb

(years)

Coal PC Supercritical – Brown Coal 50 25% 30% 15

PC Supercritical Black Coal 50 25% 30% 15

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

Natural gas CCGT 30 70% 100% 30

CCGT SWIS Scale 30 70% 100% 30

OCGT 30 85% 100% 30

Solar CLFR 40 75% 100% 40

CLFR with storage 40 75% 100% 40

Parabolic trough 35 75% 100% 35

Parabolic trough with storage 35 75% 100% 35

Central Receiver 35 75% 100% 35

Central Receiver with storage 40 75% 100% 40

Solar PV Solar PV fixed 35 75% 100% 35

Solar PV single axis tracking 35 75% 100% 35

Solar PV dual axis tracking 35 75% 100% 35

Wind On-shore Wind Farm 25 50% 100% 25

Ocean/Wave 25 75% 100% 25

Biomass Other Biomass Waste 30 75% 100% 30

Geothermal Geothermal HSA 40 75% 100% 40

Geothermal EGS 40 75% 100% 40


6


Category TechnologyEconomic life

(years)

Refurbishment cost (% of

new)

Additional life (% of original

life)

Additional life from refurb

(years)

CCS PC Supercritical with CCS – Brown Coal 50 25% 30% 15

PC Supercritical with CCS – Bituminous Coal 50 25% 30% 15

PC Oxy Combustion Supercritical with CCS 50 25% 30% 15

CCGT with CCS 45 50% 50% 23

IGCC with CCS – Bituminous Coal 30 50% 50% 15

IGCC with CCS – Brown Coal 30 50% 50% 15

CCS retrofit PC Subcritical Brown Coal - Retrofit CCS 30 25% 30% 9

PC Subcritical Black Coal - Retrofit CCS 30 25% 30% 9

Existing CCGT with retrofit CCS 30 50% 50% 15

Source: ACIL Allen, AETA (2012)

A.5.4 Carbon transport and storage costs

For plant that utilise carbon capture, transport and storage costs are applied separately. As the majority of costs related to transport and storage of CO2 are large upfront fixed costs (pipeline construction and drilling costs), it is appropriate for these to be levied to new entrant technologies as a fixed charge rather than through variable charges. This can be done either through an addition to the capital cost or through an addition charge to the fixed O&M cost. ACIL Allen adds the cost to fixed O&M values.

Costs for CO2 transport and storage are uncertain and highly dependent upon the scale of the development for both transmission pipelines and injection infrastructure. A larger CO2 pipeline grid would result in significant economies of scale over a single coal-fired power station development.

The assumed transport and storage costs are presented in Table A30. These assumptions have been informed by the AETA 2012 study. Costs are assumed to remain constant in real terms over the modelling period.

Table A30 Assumed CO2 transport and storage costsRegion $/tonne CO2-e

NSW 72

QLD 23

SA n/a

TAS n/a

VIC 22

SWIS 14

NWIS 19

DKIS n/a

Mt Isa n/a

Source: AETA (2012)

A.6 SRES model assumptions

A.6.1 Model overview

ACIL Allen’s forecasts for uptake of small-scale generation units (SGUs) are based on a regression model relating historic uptake to historic net financial returns to installing solar PV


7


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.6.8.

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.6.2 Historic uptake

Historic uptake is based on Clean Energy Regulator (CER) data on STC creation by solar installations. This data was up to date as of the end of April 2014. As STC creation can occur up to 12 months after installation, the CER data was adjusted for lag to estimate the likely ‘underlying’ level of installations in the period April 2013 to March 2014 based on raw STC creation data.

Table A31 shows the lag factors used to adjust installed capacity in each month based on data on installations or the period April 2012 to March 2013.


8


Table A31 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 occurredSource: 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.6.3 PV system costs

Average cost per kW

The cost of installing a PV system has varied over time. ACIL Allen’s estimates of historic system cost were derived by taking a national average system cost which was scaled to


9


account for differences in cost due to system size and to account for differences in system 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.18

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 A39.

Figure A39National average historic PV installation cost (2011$/kW)

$0

$1,000

$2,000

$3,000

$4,000

$5,000

$6,000

$7,000

$8,000

$9,000

$10,000

2009 2010 2011 2012 2013

$/kW

inst

alle

d (2

011)

Note: Cost excludes rebates, subsidies, and GSTSource: AECOM; ACIL Allen; SolarChoice

Small-scale costs were projected by:

taking the latest data on system cost reported by Solar Choice and adjusted for STC rebates as a starting point

18 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


10

http://www.businessspectator.com.au/article/2013/12/13/solar-energy/solar-pv-price-check-%E2%80%93-december

http://www.businessspectator.com.au/article/2013/12/13/solar-energy/solar-pv-price-check-%E2%80%93-december

http://www.solarchoice.net.au/


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 A40 shows the average cost of installing a residential and commercial roof-top solar system projected on this basis and in caparison to large-scale utilise PV systems on a per kW basis.

Figure A40National average projected PV installation cost

Note: Cost excludes rebates, subsidies, and GSTSource: ACIL Allen based on Solar Choice, and AETA

Variation by system size and location

Solar Choice’s PV Price Check data were also used to estimate a cost premium or discount for each state and territory based on averaged variations across the period. Similarly, smaller and larger systems were given a premium or discount based on observed variation from the average.

The relative premia/discounts associated with different sized systems are set out in Table A32.

Table A32 PV installation premium/discount by system sizeSystem size (kW) <1.5 1.5-2.5 2.5-3.5 3.5-4.5 4.5-7.5 >7.5

Premium/discount 12.5% 5.2% -3.0% -5.2% -9.5% -15%

Note: the 4.5 to 7.5 kW category incorporates the 4.5 to 5 kW and 5 to 7.5 kW categories. Similarly, the >7.5 kW category incorporates the 7.5 to 10 kW category and the >10 kW category.Source: ACIL Allen analysis of SolarChoice data

The premia/discounts associated with installations in different states and territories are set out in Table A33.


11


Table A33 State/territory variation in system costState/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 valueSource: 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.6.4 Rebates and subsidies

Two sources of upfront rebates and subsidies for PV installations were taken into account:

the former Solar Homes and Communities Program (SHCP), which provided an upfront cash rebate

the indirect subsidy provided by the creation of STCs under the SRES, including the creation of additional STCs through the ‘Solar Credits multiplier’.

Under SHCP, customers who installed PV systems received an upfront rebate of $8,000. SHCP was in place at the beginning of 2009, and was closed during June 2009. However, as systems installed in the second half of that based on prior applications for the rebate, it is analysed as having an effect on some installations in the second half of 2009.

In addition to the upfront payment through SHCP, PV systems were eligible to create certificates for the renewable electricity they generate during the historic period. The value of these certificates (initially RECs created under the Renewable Energy Target and then STCs created under the SRES) provides an upfront subsidy to installation of PV systems.

The value of this subsidy is depended on system size and certificate price. From June 2009 until 31 December 2012, it also depended on the ‘solar credit multiplier’, which was established under the Solar Credits scheme and allowed eligible customers who installed PV systems were deemed to create additional RECs/STCs, thereby increasing the amount of the subsidy. The multiplier was originally 5, meaning that a PV system would create 5 solar credits for every MWh of electricity it was deemed to generate, for the first 1.5 kW of capacity installed. The multiplier then declined over time.

The SHCP was phased out in favour of Solar Credits during 2009. Customers could benefit from either the SHCP or the Solar Credits multiplier, but not both. To address the overlap between these two policies, 50% of PV installations in quarter 3 2009, and 20% in quarter 4 2009 were assumed to receive the SHCP rebate. The remainder were assumed to use the Solar Credits multiplier to generate extra STCs (then RECs).

The solar multiplier and certificate values factored into the analysis are shown in Table A34. In effect, a PV system installed in 2009 was assumed to receive part of the SHCP grant and part of its entitlement through Solar Credits.


12


Table A34 Solar Credits multiplierUntil July

2009 Q3 2009 Q4 2009Q1 2010

– Q2 2011

Q3 2011 – Q2 2012

Q3 & Q4 2012

From January

2013

Solar Credits multiplier

1 3.0 4.2 5 3 2 1

SHCP value $8,000 $4,000 $1,600 $0 $0 $0 $0

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 A41. Beyond 2014, the certificate price and multiplier were assumed to remain constant (in nominal terms), at $38 per certificate, which is just below the legislated maximum.

Figure A41 REC/STC prices (nominal $/certificate)

$0

$10

$20

$30

$40

$50

$60

2009 2010 2011 2012 2013

nom

inal

$/c

ertif

icat

e

Note: REC prices prior to Q1 2011, STC prices subsequently. Source: AFMA; ACIL Allen analysis

All systems are assumed to create 15 years of ‘deemed’ RECs/STCs at the time on installation, and then cease to be eligible for further certificates after 15 years. The eligibility timeframe for deeming is limited to 2030 such that as we approach 2030, the number of years ‘deemed’ declines.

A.6.5 Retail electricity prices

Retail electricity prices are important to the financial return on solar PV as every kWh of solar output that is consumed by the owner of the system avoids the variable component of the retail electricity price.

For Tasmania, Western Australia, the Northern Territory and the Australian Capital Territory, historic retail prices were adopted based on published regulated retail prices in the relevant periods. This approach was adopted because, for the former three jurisdictions, there was no retail competition in that period, and so no other prices were available. In the case of the


13


ACT retail competition was in place, but the regulated prices were sufficiently low and 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.6.6 Feed-in tariffs and buyback rates

When solar PV systems produce more power than is required at the premises at which they are installed, the electricity is exported to the grid and on-sold to other customers. The value of this exported electricity is another important component of the financial return to PV installation.

For clarity, this report distinguishes between ‘feed-in tariffs’ and ‘buyback rates’. Within this categorisation, ‘feed-in tariff’ refers to a premium rate determined by legislation that must be paid for exported electricity from eligible PV systems.

In general, exported PV output always displaces electricity that would otherwise be purchased from the wholesale market, and therefore provides some value to the retailer that on-sells this electricity. Accordingly, retailers that supply power to owners of PV systems are generally willing to pay some amount for exported PV output that is separate from, and additional to, any premium feed-in tariff that might be imposed by legislation. The term ‘buyback rate’ refers to these payments by retailers that reflect the value of exported PV output to the retailer, and which, whilst sometimes regulated, are not intended to offer a premium rate or purposefully subsidise PV systems.

Within this categorisation, it is necessary to distinguish between three types of feed-in tariffs:

A ‘net’ feed-in tariff is the most common form, and pays a premium rate for all exported PV output

A ‘gross’ feed-in tariff meters PV output in such a way that all PV output is effectively exported, earning whatever premium rate is available, and then all of the customer’s electricity is then imported at the prevailing retail rate

A ‘one-for-one’ feed-in tariff establishes that the payment for exported PV output must be equal to the prevailing retail rate.

With this nomenclature established, Table A35 sets out the various feed-in tariffs that have been or are in operation in Australia.


14


Table A35 Feed-in tariffs by jurisdiction

Initial feed-in tariff regime

Comments

Subsequent

feed-in tariff

CommentsCurrent

arrangements

New South Wales 60c gross

Commenced early 2010, phased out

through early 2011. Paid to end

2016

20c gross

Breached cap in Q3 2011. Paid to end

2016

Buyback rate regulated to be

between 6.6 and 11.2 c

Victoria 60c netOctober 2009 to September 2011. Paid to end 2024

25c net

Breached cap in mid-2012. Paid to end

2016

Regulated minimum buyback rate of 8c

Queensland 44c net 2008 to mid-2013.

Paid to June 2028 8c net Feed-in tariff of 8c until 30 June 2014

South Australia 44c net

2008 to September 2011. Paid to June 2028

16c net

October 2011 to September 2012. Paid to

June 2016

Regulated minimum buyback rate of

7.6c

Western Australia 40c net

Mid 2010 to mid-2011. Paid for 10

years from installation

20c net

In place July and August

2011. Paid for 10 years from

installation

Synergy offers a buyback rate of

8.8529c; Horizon offers location-

specific rates of 10-50c

Tasmania One-for-one Still in place One-for-one feed-in

tariff

Northern Territory

One-for-one Still in place One-for-one feed-in

tariff

Australian Capital Territory

50.05c gross, then

45.7c gross from

October 2010

April 2009 to mid-2011. Paid for 20

years from installation

One-for-one

Mid 2011 to mid-2013

ActewAGL offers a buyback rate of

7.5c

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.6.7 System output and export rates

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

The solar zone ratings are set out in Table A36, and the historic share of installed capacity by zone, and implied average output per kW of installed capacity for each state and territory, are set out in Table A37.

Table A36 Solar zone ratingsZone 1 Zone 2 Zone 3 Zone 4

Output (MWh per kW of installed capacity)

1.622 1.536 1.382 1.185

Source: Clean Energy Regulations 2001, Schedule 5


15


Table A37 Installed capacity by solar zone

Zone 1 Zone 2 Zone 3 Zone 4

Implied average output

(MWh/kW)

New South Wales 0.0% 5.1% 93.2% 1.7% 1.386

Victoria 0.0% 0.0% 8.6% 91.4% 1.202

Queensland 0.0% 1.1% 98.9% 0.0% 1.384

South Australia 0.0% 1.3% 95.6% 3.1% 1.378

Western Australia 1.9% 5.5% 89.5% 3.1% 1.389

Tasmania 0.0% 0.0% 0.0% 100.0% 1.185


0.0% 0.0% 100.0% 0.0% 1.382

Northern Territory 33.2% 65.6% 1.3% 0.0% 1.563

Source: ACIL Allen analysis of CER data

Assumed export rates by state and by system size are set out in Table A38. Export rates are assumed to be 10% for all systems above 7.5 kW on the basis that most larger systems will be commercial or industrial systems, and will be sized to minimise exports and thereby maximise avoided network tariffs.

Table A38 Export rates<1.5 1.5-2.5 2.5-3.5 3.5-4.5 4.5-7.5 >7.5

New South Wales 35% 40% 45% 50% 55% 10%

Victoria 35% 40% 45% 50% 55% 10%

Queensland 35% 40% 45% 50% 55% 10%

South Australia 40% 45% 50% 55% 60% 10%

Western Australia 40% 45% 50% 55% 60% 10%

Tasmania 30% 35% 40% 40% 45% 10%


35% 40% 45% 50% 55% 10%

Northern Territory 35% 40% 45% 50% 55% 10%

Source: ACIL Allen assumptions

A.6.8 System size trends

Data on PV uptake illustrates clearly a substantial increase in average system size over time, in particular as the incentives created by the Solar Credits formula to install smaller systems has dissipated and been overwhelmed by the attraction of lower system costs. Further, although CER data does not distinguish between residential and non-residential systems, ACIL Allen analysis for various distribution businesses indicates recent growth in the share of larger systems being proposed by non-residential customers. This trend is widely anticipated within the PV industry, and reflects the financial opportunity from low


16


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.

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 A42 shows the share of installed capacity across Australia by system size class.


17


Figure A42Share of installed capacity by system size class

Note: 10-kW includes systems of 10-100kWSource: ACIL Allen Analysis of CER data

ACIL Allen has estimated the growth in the market share of larger commercial systems based on a logistic growth function. The logistic growth function was fitted to quarterly data on the share of installed PV capacity that came from systems larger than 7.5kW by determining the value for the parameter a in the equation below for each state:

S ( t )=e ln (S (0) )∗e (−a∗t )

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.7 Retail price modelRetail prices are modelled using a ‘cost-stack’ approach accounting for the major components of retail prices, namely:

Wholesale energy costs

Network costs

RET costs

Costs associated with other ‘green schemes’

Costs attributable to losses


18


Metering costs, retail operating costs and retail margins.

Each of these component are discussed further in subsequent sections.

Retail prices are 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 are 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.7.3.

Three separate retail series are 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.7.5. 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.7.1 Wholesale energy costs

Slightly different estimates for wholesale energy costs were made based on available information. For the NEM, the following process was followed:

Historic NEM demand for the years 2008 to 2012 was disaggregated into residential, commercial, industrial and EITE components by:

Allocating the net system load profile (NSLP) load to the residential category

Allocating the demand component identified as ‘industrial’ by AEMO in its 2013 National Electricity Forecasting report to the EITE category, on the basis that this load is constant over the course of each year

Separating the remainder into commercial and industrial components on the basis of an assumed industrial load shape involving demand that is 25% lower in off-peak times than peak times.

Calculating ‘uplift’ factors based on this historic decomposition for the non-residential categories, which are then held constant throughout the projection

Calculating uplift factors for the residential category that changes over time in response to modelling outcomes by:

Regressing historic NSLP load against historic NEM load and time-of-day and seasonal characteristics for each NEM region and the ACT


19


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 A39. For the DKIS, simpler factors were assumed due to a lack of detailed historic data, also shown in Table A39, and were estimated based on analysis by the AEMC in 2012 and 2013 of trends in residential retail prices.

Table A39 Uplift factors by customer classResidential Commercial Industrial EITE

NSW Time varying 8% 5% 0%

Queensland Time varying 4% 4% 0%

South Australia Time varying 15% 7% 0%

Tasmania Time varying 5% 5% 0%

Victoria Time varying 10% 5% 0%

ACT Time varying 8% 5% 0%

DKIS 60% 30% 15% 0%

Source: ACIL Allen analysis of AEMO data and AEMC analysis

Additional hedging factors were assumed for each NEM region to reflect the cost of hedging against higher price events (e.g. in the event of high demand events or combinations of outages that cause price spikes). These hedging factors were based on 2011 ACIL Allen (then ACIL Tasman) analysis for the AEMC on retail price trends, which found that hedging costs were a relatively stable uplift from load-weighted residential prices. In the SWIS, constant real capacity credit costs were assumed for each customer class. Assumed hedging cost factors and capacity credit costs are shown in Table A40.

Table A40 Hedging factors and capacity credit costs by customer classResidential Commercial Industrial EITE

Hedging factors expressed as a percentage

NSW 13% 13% 10% 10%

Queensland 16% 16% 10% 10%

South Australia 17% 17% 10% 10%

Tasmania 14% 14% 10% 10%

Victoria 15% 15% 10% 10%

ACT 10% 10% 8% 8%

DKIS 10% 10% 8% 8%

Capacity credit costs expressed as real $2014 per MWh

SWIS $61 $44 $33 $22

Source: ACIL Allen analysis

Carbon costs for the first half of 2014 (before the assumed repeal effective from 1 July 2014) are included within the wholesale energy component.


20


A.7.2 Network costs

Network costs can be analysed 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 A41 summarises the historic, regulatory determination based projection and modelled projection of network costs for each network.

Table A41 Basis of network cost calculation

Element Networks HistoricProjection based on

regulatory determinations

Projection based on modelled

assumptions

NSW distribution and sub-transmission

Ausgrid, Endeavour Energy, Essential

EnergyTo 2013-14 2014-15 to 2018-19* From 2019-20

NSW transmission TransGrid To 2013-14 2014-15 to 2018-19* From 2019-20

Victoria distribution

CitiPower, Jemena, Powercor, SP

AusNet, United Energy

To 2014 2015 From 2016

Victoria transmission SP AusNet To 2014 2015 to March 2017 From March 2017

Queensland distribution

Energex, Ergon Energy To 2013-14 2014-15 From 2015-16

Queensland transmission Powerlink To 2013-14 2014-15 to 2016-17 From 2017-18

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 distribution Aurora To 2013-14 2014-15 to 2016-17 From 2017-28

Tasmania transmission Transend To 2013-14 2014-15 to 2018-19* From 2019-20

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 AERSource: AER, ERAWA, NT Utilities Commission.

Overall cost trends were divided by energy growth trends to give a cost per unit energy trend (i.e. price) trend.


21


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. Similarly, these charges were assumed to linearly trend to 75% of industrial and EITE network charges from present levels by 2030.

A.7.3 RET costs

RET costs for non-EITE customers (i.e. customers without access to partial exemptions from the RET) are calculated as the simple multiplication of the prevailing LGC or STC price by the ‘renewable power percentage’ (RPP) or ‘small-scale technology percentage’ (STP).

The RPP is calculated directly for each scenario based on exogenous demand targets and assumed levels of ‘reduced acquisitions’. Reduced acquisitions are equal to ‘relevant acquisitions’ under the scheme, less the volume of partial exemption certificates (PECs) issued.

Relevant acquisitions were calibrated to the CER’s estimate of 2014 relevant acquisitions (approximately 205 TWh), and then grown in line with modelled electricity demand. PECs were calculated based on analysis of PEC creation from 2010 to 2013, and the implied level of EITE load underlying this level of PEC creation. EITE load was assumed to reduce by 2,500 GWh in 2015 to reflect the closure of the Point Henry aluminium smelter, and then grow in line with total demand from that point. The calibration of historic EITE load based on historic PEC creation is shown in Table A42.

Table A42 PEC creation and EITE load2011 2012 2013 2014

Actual Actual Actual CER estimate

PEC creation (000s)

Highly (actual) 26,515 32,455 30,437

Moderately (actual) 647 996 955

Total 27,162 33,451 31,392 27,200

Exemption rateHighly 68.4% 77.5% 74.7% 67.6%

Moderately 45.6% 51.6% 49.8% 45.0%

Implied EITE load (GWh)

Highly 38,767 41,889 40,729 39,029

Moderately 1,420 1,928 1,917 1,837

Total 40,187 43,827 42,646 40,867

Note: Partial exemptions differentiate between activities that are ‘highly’ or ‘moderately’ emissions intensive. The calculation above addresses these two categories separately. Source: ACIL Allen analysis of CER data

The STP was taken as the pre-determined STP for 2014. The 2015 STP is adjusted for over/under achievement of the implied STC creation target in 2014 relative to ACIL Allen’s modelled rate of STC creation. From 2016 onwards the STP is calculated as the modelled rate of STC creation divided by reduced acquisitions.

LGC costs for the LRET cost component are taken as the certificate price modelled by ACIL Allen. The market (as opposed to legislated) STC price paid by customers was assumed to be $38 (constant nominal) in all scenarios other than where the SRES was repealed.

A.7.4 Green scheme costs, losses and retail costs

These relatively small cost components include:


22


Various green schemes

The Energy Savings Scheme in NSW

The Victorian Energy Efficiency Target

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)

Smart metering costs in 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.7.5 Treatment of solar costs

As discussed above, the wholesale component of the retail cost series that is avoided by own consumption of PV output, and for ‘buyback rates’ for solar exports, were adjusted to transition over time towards the wholesale market value of solar generation.

This transition was assumed to occur such that:

Residential customers would avoid the wholesale cost of serving their typical energy consumption until 2025, after which time this rate would transition to the point where 50% of customers would avoid only the solar dispatch-weighted price by 2030.

Commercial customers would avoid the wholesale cost of serving their typical energy consumption until 2025, after which time this rate would transition to the point where 100% of customers would avoid only the solar dispatch-weighted price by 2035.

Industrial and EITE customers would transition to a rate 100% reflective of the wholesale value of solar between 2025 and 2030.

Though approximate, this transition implies that as solar penetration increases, and the correlated solar output depresses wholesale market prices during sunny periods, typical wholesale costs of supply and the wholesale market value of solar will diverge over time. In response, we assume that policy-makers and retailers will implement a regulatory and/or market response that sees rooftop solar systems metered in a way that allows the value of their output to be more accurately credited, rather than simply credited as avoided a ‘typical’ unit of energy consumption. This transition could be implemented through metering requirements for new solar installations, or widespread adoption of time-of-use metering for retail load allowing differentiation between the cost of serving customers with and without rooftop solar.

The commercial imperative for such a change is substantial, as our modelling suggests that the value of solar output declines to a value of around 80% of the average market price by 2030 in most NEM regions, and to around 60% by 2040, as solar penetration increases.

A.8 Electric vehicle energy projectionsWithin the demand projections undertaken by pitt&sherry, the uptake of Electric vehicles was specifically excluded from the analysis. As part of this project, the Department asked


23


ACIL Allen to estimate an electricity vehicle uptake scenario and include this incremental electricity demand to the Baseline scenario as an additional sensitivity.

ACIL Allen estimated the impact of the take up of plug in electric vehicles on energy consumption across all Australian states by:

1. Obtaining ABS population projections for each state

2. Projecting the number of passenger and light commercial vehicles in each state by applying historical vehicle ownership rates to the projected population

3. Using a LOGIT model specification to calculate the share of new vehicle sales that are captured by plug in electric vehicles

4. Making some assumptions concerning driving patterns and vehicle charging patterns to estimate the total impact of electric vehicles on energy consumption.

Definition of vehicles

There are four specific vehicle types used within the analysis. These are:

ICE (Internal Combustion Engine)

HEV (Hybrid Electric Vehicle)

PHEV (Plug in Hybrid Electric Vehicle)

EV (Electric Vehicle).

While HEVs are classified as electric vehicles, we do not consider these as part of the projections as they do not connect to the electricity grid.

The two categories of vehicles which are relevant to this study are PHEVs and EVs. Both are charged by connecting to the electricity grid. While PHEVs also run on conventional fossil fuels, EVs are completely powered by electricity.

Separate projections are produced for passenger and light commercial vehicles. Moreover, the passenger vehicle categories are further disaggregated into Small, Medium and Large vehicle categories.

Economics of electric vehicles

The key economic factors that are likely to play a significant role in the potential take up of electric vehicles are:

Vehicle prices

Petrol and electricity prices

Vehicle fuel efficiency

Running costs

Range

Charging convenience

A logistic modelling framework is used to convert the underlying economic drivers of electric vehicles into an impact on market share and take-up of the technology. This is done by creating a model which values each of the attributes that drive the decision to adopt the technology and then to apply an elasticity or measure of responsiveness of market share to each factor.

Input assumptions

Three separate scenarios are presented, a low, medium and high scenario.


24


Under the low scenario, population growth corresponds to series C of the ABS publication 3222 Population Projections, Australia, 2012 to 2101. Series B is used in the medium scenario while Series A corresponds to the high growth scenario.

The other factor that varies under the three scenarios is the relative rate of decline in plug-in electric vehicle costs. Under the medium scenario, the real cost of plug in electric vehicles is projected to decline at a rate of 4% per annum. Under the high scenario the real rate of price decline is 6% per annum, while it is 2% under the low scenario.

Under all three scenarios, the real price of conventional ICE vehicles is projected to remain unchanged.

A.8.1 Impact on energy consumption

To determine the energy consumption of the stock of electric vehicles we assume the average distance travelled per day is 40 km. This is equivalent to 14,610 kilometres per annum.

The average driving distance is then used in conjunction with the actual and projected fuel efficiency of plug in electric vehicles. The energy consumption of each electric vehicle is determined in this way. The total energy consumption of the entire stock of electric vehicles is then calculated by multiplying this measure by the number of projected electric vehicles.

The projected energy impact of electric vehicles across Australia is shown in Figure A43. Energy consumption of PHEV/EVs is projected to reach 12,310 GWh by 2035.

Figure A43Energy impact of electric vehicles by State (medium scenario)

Source: ACIL Allen

Figure A44 compares the aggregate energy impact across the three uptake scenarios. Under the ‘high uptake’ scenario, energy demand is 67% higher in 2040; whereas the ‘low uptake’ scenario projects energy consumption to be around 70% lower in 2040 when compared with the medium case. This highlights the level of uncertainty within the projections.


25


Figure A44Aggregate energy impact of electric vehicles across scenarios

Source: ACIL Allen


26

ACIL ALLEN CONSULTING PTY LTDABN 68 102 652 148

161 WAKEFIELD STREETADELAIDE SA 5000AUSTRALIAT+61 0412 089 043

LEVEL FIFTEEN127 CREEK STREETBRISBANE QLD 4000AUSTRALIAT+61 7 3009 8700F+61 7 3009 8799

LEVEL TWO33 AINSLIE PLACECANBERRA ACT 2600AUSTRALIAT+61 2 6103 8200F+61 2 6103 8233

LEVEL NINE60 COLLINS STREETMELBOURNE VIC 3000AUSTRALIAT+61 3 8650 6000F+61 3 9654 6363

LEVEL TWELVE, BGC CENTRE28 THE ESPLANADEPERTH WA 6000AUSTRALIAT+61 8 9449 9600F+61 8 9322 3955

LEVEL ONE50 PITT STREETSYDNEY NSW 2000AUSTRALIAT+61 2 8272 5100F+61 2 9247 2455

ACILALLEN.COM.AU

consultants report_acil allen_with alt text · web viewprojected demand growth under the scenario...

Documents