Professional people with purpose
Making the people of New South Wales
proud of the work we do.
Level 15, 1 Margaret Street
Sydney NSW 2000 Australia
t +61 2 9275 7100
f +61 2 9275 7200
e mail@audit.nsw.gov.au
office hours 8.30 am–5.00 pm
audit.nsw.gov.au
New South Wales Auditor-General’s Report
Special Report
Solar Bonus Scheme
New South Wales Auditor-General’s Report | Special Report | Solar Bonus Scheme
Professional people with purpose
audit.nsw.gov.au
The role of the Auditor-General
The roles and responsibilities of the Auditor-
General, and hence the Audit Office, are set
out in the Public Finance and Audit Act 1983.
Our major responsibility is to conduct
financial or ‘attest’ audits of State public
sector agencies’ financial statements.
We also audit the Total State Sector Accounts,
a consolidation of all agencies’ accounts.
Financial audits are designed to add credibility
to financial statements, enhancing their value
to end-users. Also, the existence of such
audits provides a constant stimulus to agencies
to ensure sound financial management.
Following a financial audit the Office issues
a variety of reports to agencies and reports
periodically to parliament. In combination
these reports give opinions on the truth
and fairness of financial statements,
and comment on agency compliance with
certain laws, regulations and Government
directives. They may comment on financial
prudence, probity and waste, and recommend
operational improvements.
We also conduct performance audits. These
examine whether an agency is carrying out its
activities effectively and doing so economically
and efficiently and in compliance with relevant
laws. Audits may cover all or parts of an
agency’s operations, or consider particular
issues across a number of agencies.
Performance audits are reported separately,
with all other audits included in one of the
regular volumes of the Auditor-General’s
Reports to Parliament – Financial Audits.
audit.nsw.gov.au
GPO Box 12
Sydney NSW 2001
The Legislative Assembly
Parliament House
Sydney NSW 2000
Pursuant to section 194(1) of the Electricity Supply Act 1995,
I present my report entitled Solar Bonus Scheme.
Peter Achterstraat
Auditor-General
7 November 2011
© Copyright reserved by the Audit Office of New South
Wales. All rights reserved. No part of this publication may
be reproduced without prior consent of the Audit Office of
New South Wales.
The Audit Office does not accept responsibility for loss or
damage suffered by any person acting on or refraining from
action as a result of any of this material.
Our vision
To make the people of New South Wales
proud of the work we do.
Our mission
To perform high quality independent audits
of government in New South Wales.
Our values
Purpose – we have an impact, are
accountable, and work as a team.
People – we trust and respect others
and have a balanced approach to work.
Professionalism – we are recognised
for our independence and integrity
and the value we deliver.
The Legislative Council
Parliament House
Sydney NSW 2000
1
NSW Auditor-General's Report
NSW Solar Bonus Scheme
CONTENTS
Executive Summary 2
Background 2
Key Findings 3
Recommendations 4
Responses from Department of Premier and Cabinet 6
Introduction 14
NSW Solar Bonus Scheme 14
Key Findings 15
1. Number of Customers under the Scheme 15
2. Costs of the Scheme 18
2.1 Scheme costs incurred to 30 June 2011 18
2.2 Current projected Scheme costs 19
2.2.1 Projected costs 19
2.2.2 Uncertainty within projections 21
2.3 History of projected Scheme costs 23
2.3.1 Review of models underpinning the different projected costs of the Scheme 25
3. Other matters considered relevant 26
3.1 How the Scheme will be funded 27
3.2 Were specific and measurable program objectives set? 28
3.3 Were costs and benefits assessed? 29
3.4 Were relevant risks identified, assessed and addressed? 31
3.5 Was there an achievable implementation program? 32
3.6 Were there adequate control mechanisms? 33
3.7 Was information relevant and timely? 33
3.8 Was there an effective audit process to provide program assurance? 35
Appendices 36
Appendix 1: Chronology of the Key Scheme Events 36
Appendix 2: A Generous Scheme 38
Appendix 3: Uncertainty within Projections 39
Appendix 4: Sensitivity Analysis for Uncertainty within Projections 41
Appendix 5(a): Review of Models 42
Appendix 5(b): Comparison to DNSP#1 Scheme tariff payments 52
Appendix 5(c): Comparison to DNSP#2 Scheme tariff payments 53
Appendix 5(d): Comparison to DNSP#3 Scheme tariff payments 54
Appendix 6: Glossary 55
Appendix 7: About the Audit 56
Contents
2
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
Background
I have prepared this Report to Parliament as required by section 194(1) of the Electricity
Supply Act 1995, on the following aspects of the NSW Solar Bonus Scheme:
a. the number of small retail customers that have installed and connected complying
generators,
b. the costs of the scheme including the total amount credited to small retail customers
under the scheme,
c. any other matter that the Auditor-General considers to be relevant.
In relation to (c) above, I examined how the Scheme will be funded and whether:
• specific and measurable program objectives were set
• costs and benefits associated with achieving the objectives were assessed and reported
in the proposals put forward for decision making
• relevant risks to the achievement of each program objective were identified, assessed and
addressed
• there was an achievable implementation program
• there were control mechanisms to identify and react to the take-up of the scheme and
other changes that could affect the achievement of program objectives
• relevant and timely information was provided to Government decision makers, potential
applicants and other stakeholders to take informed action at each stage in the
development and operation of the scheme
• there was an effective audit process to provide program assurance.
The New South Wales Government’s Solar Bonus Scheme (the Scheme) was introduced
through legislation in 2009 and commenced on 1 January 2010. The Scheme is legislated to
run for seven years to 31 December 2016.
The Scheme is a gross feed-in tariff scheme. It provides support to people who produce
electricity through rooftop solar photovoltaic (PV) systems or mini wind turbines connected to
the electricity grid.
The Scheme originally announced by the Government in June 2009, was a net tariff scheme
over 20 years. The Government later switched to a gross tariff scheme over seven years. Net
tariff pay system owners only for the electricity they export to the grid in excess of that
consumed inside the property. Under gross tariff arrangements, system owners export and
receive payment for all the electricity generated by their system.
The NSW Department of Trade and Investment, Regional Infrastructure and Services
(DTIRIS), the Office of Environment and Heritage (OEH), Department of Premier and Cabinet
(DPC) and Treasury have been the major agencies involved in the Scheme. They, or their
predecessor agencies, were represented on the NSW Solar PV Feed-in Tariff Task Force that
designed the Scheme originally and administered its implementation. DTIRIS chaired the task
force; submitted the original proposals to Cabinet; drew up the implementation plan and
reported progress to the Minister. The electricity network businesses registered new
customers and reported statistics on number of application and installations to DTIRIS.
Whilst my report focuses on the actions of Government agencies, it is important to recognise
that the final decisions in relation to the Scheme rested with the New South Wales
Government and the New South Wales Parliament.
Executive Summary
3
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
Key Findings
Number of Customers under the Scheme
In relation to section 194(1)(a) of the Act, I found that:
• at 30 June 2011, 132,061 small retail customers had installed and connected complying
generators under the Scheme.
Additionally, I found the following with respect to connections:
• at 30 June 2011, the 300 MW cap had been broken and total capacity was forecasted to
reach 372MW if all approved customers were connected
• 28,733 (or 17.9 per cent of total applicants) applicants were not connected to the Scheme
at 30 June 2011
• 142,849 (88.8 per cent) applicants joined the Scheme when the tariff rate was 60 cents
• at 30 June 2011, Essential Energy had approximately 34.7 per cent of the Scheme
connections compared to its approximate 24 per cent share of all network customers
• there was no centralised process for households to apply to join the Scheme.
Costs of the Scheme
In relation to section 194(1)(b) of the Act, I found that $142 million had been credited to small
retail customers under the Scheme to 30 June 2011. An estimated $26.8 million of other
further capital and operating expenditure has been incurred by the Government to
30 June 2011.
Additionally, I found the following with respect to costs:
• the latest Government projection for the total tariffs to be paid under the Scheme is
$1.75 billion, including the $142 million paid to 30 June 2011
• I have reviewed the Scheme including modelling and, with the advice from my
consultants, I anticipate the total tariffs to be paid under the Scheme will be between
$1.05 billion and $1.75 billion. The Government’s projected Scheme cost is at the upper
end of this range. This is mainly because it has assumed very good energy output from
the Solar panels over seven years
• there are many variables that contribute to the uncertainty of projected costs of this seven
year Scheme. The two most significant variables are:
− the ‘real world’ average performance of solar systems within the Scheme could be
significantly less than typical ‘good’ solar system performance. Underperformance of
systems can be caused by many factors, including system orientation to the sun,
possible shading, quality of products and/or wiring of installations. Whilst there is little
available information at present, it is entirely possible that, on average, PV systems
within the Scheme might underperform very significantly. A 20 per cent reduction on
‘good’ typical PV generation is entirely possible, equating to an 18 per cent reduction
in projected Scheme costs .
− potentially changing weather patterns – total sunshine (i.e. solar insolation) can vary
+/-10 per cent year to year, although it would be expected that the average variation
over the remaining five years of the Scheme would be significantly less than this. A +/-
5 per cent variation in total sunshine (and hence PV generation) over the life of the
Scheme equates to approximately +/-4.5 per cent variability in projected Scheme
costs
• there are many models used to project the total cost of the Scheme. Some key models
have been reviewed by my consultant, the University of New South Wales (UNSW), and
overall I consider they are satisfactory given at times resource and time constraints,
although one DNSP model can be improved by reducing the assumed PV performance
• determining and reporting of Scheme costs varied between the DNSPs as no guidance or
direction was given.
4
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
Other Matters Considered Relevant
In relation to section 194(1)(c) of the Act, I found with respect to funding the Scheme:
the majority of the funds in the New South Wales Climate Change Fund (NSW CCF) will
be used to reimburse DNSPs for their tariff payments to retailers under the Scheme
I understand approximately $430 million of interest free loans will be made by the
Government to the NSW CCF until the DNSPs network tariffs charged to all customers
are sufficient to repay the loans
In relation to section 194(1)(c) of the Act, I found with respect to planning and management of
the Scheme:
the Scheme had three broadly stated objectives, with no specific targets against which
progress could be measured. These objectives do not include reducing emissions or
obtaining value for money
no cost-benefit analysis was undertaken before the Government’s decision in 2008 to
introduce a scheme. Likewise, no cost-benefit analysis was undertaken when changes
were made to the Scheme in 2009, or when changes were made to funding arrangements
early in 2011
no market research was undertaken (including about non-tariff options) to investigate
customer motivations in generating renewable energy
little was done early enough to identify and reduce relevant risks. I found no contingency
planning, analysis and assessment of options and exit strategies to address potential high
risk situations
no overall implementation program, including no clear definition of project roles and
responsibilities of those involved in implementing and delivering the Scheme
the Scheme lacked the most elementary operational controls. There was initially a poor
monitoring system. There was a time limit of 2016, but initially there was no cap on total
Scheme capacity and costs
there were significant shortcomings in the provision of information to Government
decision-makers
there were insufficient review points based on applications so that progress could be
assessed and the Scheme amended or even stopped, if necessary. The one legislated
review point was when capacity reached 50 MW. By the time that review was complete,
capacity had doubled to 100 MW.
at the outset, there was no audit process to provide program assurance.
Recommendations
Number of Customers under the Scheme
In relation to section 194(1)(a) of the Act, I recommend the DNSPs ascertain whether
applicants who have not been connected to the Scheme still intend to join and/or need
assistance with the process. A clear understanding of intentions will assist DNSPs to
anticipate future resource needs and improve/maintain good service delivery by the
Government.
Costs of the Scheme
In relation to section 194(1)(b) of the Act, I recommend:
The Government publish the total projected Scheme costs, within a range, based on a
sensitivity analysis of variables, including impact of weather patterns on the extent of
energy produced and paid for under the Scheme
DTIRIS seeks medium term forecasting of El Nino and La Nina weather patterns and
shares this information with DNSPs to assist them in forecasting Scheme costs
DNSPs continue to update total projected Scheme costs with actual data collected
DNSPs use a standard approach to estimating and reporting Scheme costs.
5
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
Other Matters Considered Relevant
In relation to section 194(1)(c) of the Act, I recommend:
• Schemes that involve significant recurrent expenditure and economic costs to consumers
should be assessed in a similar fashion to major Government infrastructure expenditure.
Prior to approval, major programs need:
− specific objectives that are measurable
− a cost-benefit analysis
− an economic analysis
− a business case including options, costs, time frames and risks
− a risk assessment and risk management plan
− a budget
− an implementation plan
− a performance monitoring framework
− an exit plan
• in the interests of being transparent, and ensuring that Government decision-makers have
access to adequate information on such schemes, such assessments should be made
public
• the performance of such schemes needs to be monitored more closely. Agencies need to
be prepared to step in and take corrective action more quickly if important implementation
risks begin to materialise
• there needs to be an ability and a willingness to alter the parameters of such schemes
more quickly, particularly if there is a risk that they will not meet their objectives.
6
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
Responses from Department of Premier and Cabinet
7
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
8
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
9
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
10
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
11
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
12
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
13
NSW Auditor-General's Report
NSW Solar Bonus Scheme
EXECUTIVE SUMMARY
14
NSW Auditor-General's Report
NSW Solar Bonus Scheme
INTRODUCTION
NSW Solar Bonus Scheme
The Government’s Solar Bonus Scheme (the Scheme) was introduced by way of legislation in
late 2009 and commenced on 1 January 2010.
The Scheme is legislated to run for seven years to December 2016, offering customers a
gross feed-in tariff of 60 cents per kilowatt hour (kWh) for electricity that was generated by
eligible roof-top solar photovoltaic (PV) systems and mini wind turbines connected to the
electricity grid.
In June 2009, the Government announced (but did not legislate) and commenced preparation
for a 60 cents net tariff scheme over 20 years. The relevant scenario from the 2009 NSW
Solar Feed-In Tariff Report (Tariff Report) to Ministers indicated the total nominal cost of tariff
payments over the life of this Scheme design would be $1.6 billion (net present value
$857 million). Later, in November 2009, the Government announced its final decision to
switch to a 60 cents gross tariff scheme over seven years. That Scheme design had a total
nominal cost of $449 million (net present value $355 million) based on the relevant scenario
from the Taskforce Report. These Scheme scenarios assumed a linear uptake rate of 15 MW
a year. Thus, at the end of a 20 year scheme the total installed capacity would be 300 MW
and 105 MW for a seven year scheme.
A chronology of events is included in Appendix 1 and a comparison of the NSW Scheme with
those in other jurisdictions is at Appendix 2.
How the Scheme works
Under a gross tariff scheme, all electricity produced by the solar PV system is paid for at the
feed-in tariff rate, regardless of the amount used by the household or business. Therefore, the
household or business’s own consumption is treated as a standard retail purchase of
electricity, the same as any other consumer.
Under a net tariff scheme, only the electricity produced that exceeds the requirements of the
household or business is paid for at the feed-in tariff rate. In other words, the household or
business substitutes the electricity generated by their solar PV system for the retail electricity
purchase they would otherwise have to make, and only when their generation is greater than
their consumption do they export this excess and receive payment for it at the feed-in tariff
rate.
Although the Scheme may encourage and support persons who want to generate renewable
energy, there is very little incentive under a gross tariff scheme for consumers to reduce their
electricity consumption. Under a net tariff scheme consumers benefit from energy they do not
consume. A net feed-in tariff thereby fosters improved energy efficiency in the household as
the less electricity used by the household, the greater the return in the form of a tariff
payment/credit they will receive.
The Government intends that the DNSPs’ tariff payments will be recovered from the NSW
CCF. The electricity retailers are currently not asked to pay for the energy gained as a result
of the Scheme.
Example
In the case of a 60 cents gross tariff, a household with a 1.5 kW system that generates
2,073 kWh (for example) per year would expect to receive a credit on their electricity bill or a
cash payment for $1,244 from their electricity retailer. If the household uses:
• only 2,500 kWh per year for which they may typically pay an amount which averages
around 26.4 cents per kWh to their electricity retailer, they will pay $660 and receive
$1,244, a net benefit of $584
• 7,500 kWh per year for which they pay an amount which averages around 30.1 cents per
KWh, they will pay $2,257 and their net cost of electricity use will be $1,013 which is
45 per cent less than the full cost they would ordinarily pay.
In both cases, the electricity retailer is reimbursed for the $1,244 paid to the household for
solar energy generated.
Introduction
15
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
1. Number of Customers under the Scheme
As required by section 194(1)(a) of the Act, I am to review and report to parliament on the
number of small retail customers that have installed and connected complying generators
under the Scheme.
At 30 June 2011, the number of small retail customers installed and connected under the
Scheme with one of three DNSPs was 132,061. The total amount credited to this group of
small retail customers was $142.1 million.
The table appearing below covers the 18 month period from the commencement of the
Scheme to 30 June 2011 and comprises data collated from information provided by the three
DNSPs. The data has been confirmed to records from customer and network systems
maintained by the DNSPs I note that there was no centralised process for households to
apply to join the Scheme. Households only interaction was with installers whose contact with
DNSPs was for the purpose of arranging for a system to be connected.
The table also includes data for both connected customers at 30 June 2011 and those still
awaiting connection and eligible for connection at that time. The data are updated by DNSPs
over the life of the Scheme and is available for use by Treasury regarding to future funding
commitments of the Scheme.
Solar Bonus Scheme Statistics
for the eighteen months to
30 June 2011
$0.60 kWh $0.20kWh Total
Existing connected customers
Total number connected
118,531
13,530
132,061
Total MW capacity connected 272.4 28.5 300.9
Total kWh generated under the Scheme 224,101,796 2,827,533 226,929,329
Total amount of subsidy paid ($ million) 141.2 0.9 142.1
Approved but waiting connection
Total number awaiting connection 24,318 4,415 28,733
Total MW capacity still to be connected 50.6 20.1 70.7
Total connected and awaiting connection
Total number of customers 142,849 17,945 160,794
Total MW capacity 323.0 48.6 371.6
Source: Table complied from data provided by the three DNSPs during July and August 2011
The data relating to outstanding connections at 30 June 2011 are subject to ongoing on-going
validation by the three DNSPs and as a consequence may vary over time.
Since the Scheme commenced there was strong growth in the number and rate of installation
of PV systems. By October 2010, it is understood that New South Wales had the largest
amount of installed small scale PV systems of any jurisdiction in Australia.
A reason for the high uptake and cost of the Scheme, as noted in the October 2010 NSW
Solar Bonus Scheme Statutory Review Report, was global solar panel prices had more than
halved since mid 2009. Another reason for the high take up rate was the upfront
Commonwealth subsidy. The take up rate also increased around the time the Government
introduced legislation reducing the tariff rate (from midnight on 27 October 2010) from
60 cents to 20 cents for new Scheme applicants.
Key Findings
16
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
On 29 April 2011, the Minister for Resources and Energy announced a two month hold on the
Scheme to new applications. On 13 May 2011, the New South Wales Government announced
no more applications would be received for the Scheme effective midnight on 28 April 2011.
The Scheme was formally closed on 1 July 2011 with the publication of a gazette notice.
Legislation allowed the Minister to close the Scheme once satisfied the total generating
capacity of all complying generators installed and connected to the grid reached 300 MW.
Number of PV Systems in New South Wales, as at 30 June 2011:
Capacity of PV Systems in New South Wales, as at 30 June 2011:
Sources: DTIRIS, May 2011and DNSPs, 30 June 2011
0
50
100
150
200
250
300
350
400
Capacity (megawatt)
Date
Growth in Scheme PV systems capacity
Capacity of Applications
Installed Capacity
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
180,000
Number of PV System
Growth in Scheme PV connections
No. of Connections
No. of Applications
17
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
At 30 June 2011, Essential Energy had approximately 34.7 per cent of the Scheme
connections compared to its approximately 24 per cent share of all network customers.
The graph below details market share of the three DNSPs at 30 June2011.
* Connections and pending connections
Recommendation
I recommend the DNSPs ascertain whether applicants who have not been connected to the
Scheme still intend to join and/or need assistance with the process. A clear understanding of
intentions will assist DNSPs to anticipate future resource needs and improve/maintain good
service delivery by the Government.
53,372
33%
55,875
35%
51,547
32%
DNSPs percentage share of Scheme connections*
Ausgrid
Essential Energy
Endeavour Energy
1.620 million
49%
0.800 million
24%
0.870 million
26%
DNSPs percentage share of network connections
Ausgrid Essential Energy Endeavour Energy
18
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
2. Costs of the Scheme
As required by section 194(1)(b) of the Act, I am to review and report to parliament the costs
of the Scheme including the total amount credited to small retail customers under the
Scheme.
2.1 Scheme costs incurred to 30 June 2011
Tariff Costs
The total amount credited to small retail customers was $142.1 million.
The total tariffs paid to customers over the 18 month life of the Scheme to 30 June 2011 has
been compiled from information provided by the three DNSPs during July 2011. The total
amount paid as subsidies under the Scheme comprises cash payments, accrued amounts
and estimates for unread meters. The unread meters amounts are estimates of the value of
electricity generated under the Scheme, where customer meters have not been read at
30 June 2011. These customers will have their meters read progressively over the four
months from 30 June 2011.
The DNSPs provide the Minister and Director-General figures on the number of customers
who have installed and connected to the network; the total generating capacity; and the
amount of electricity supplied by their systems, as required under the Electricity Supply Act
1995. In addition to this information, each DNSP also provides details of tariffs paid to date as
well as estimates of the total amount of tariffs expected to be paid over the life of the Scheme.
Other Costs
While it was a licence condition of each DNSP to maintain and provide both prescribed and
requested information to the Minister and Director-General, there was no requirement to
collect information on other costs incurred by each DNSP and other Government agencies
involved in the administration, policy direction and reporting obligations for the Scheme.
These have not been considered when assessing total cost over the life of the Scheme.
While not significant when compared to the total subsidy cost of the Scheme, these other
costs are still considerable and may be borne by all electricity customers and the community,
either through increased network charges or reduced dividend payments made to
Government and appropriations by Government.
18 months period to 30 June 2011 DNSP
$'000
Other
Government
$'000
Total
$'000
Operating expenditure 14,888 1,650 16,538
Capitalised costs 10,276 -- 10,276
Total operating and capital expenditure
25,164
1,650
26,814
Life of Scheme to 31 December 2016 DNSP
$'000
Other
Government
$'000
Total
$'000
Operating expenditure 19,050 3,150 22,200
Capitalised costs 10,407 -- 10,407
Total operating and capital expenditure 29,457 3,150 32,607
Notes: The table comprises a collection of both actual costs incurred and costs calculated from management
estimates. The significant costs for DNSP#3 were subject to independent external review as part of an AER pass
through application.
Operating expenditure will be reported annually by each agency through their profit and loss statements, while capital
cost are depreciated over the life of each asset added to the agencies’ balance sheets.
19
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
2.2 Current projected Scheme costs
2.2.1 Projected costs
I engaged UNSW to provide an estimate of the projected costs for the Scheme. The following
summarises UNSW’s possible and probable ranges of Scheme costs expected over the life of
the Scheme to December 2016. UNSW has provided a possible range of the Scheme costs
between $1.05 billion and $1.75 billion. The current Scheme cost estimate of $1.75 billion is
compiled from estimates supplied by the three DNSPs. It is at the top of this range and is
higher than the UNSW reference scenario value of $1.44 billion. Differences between the
UNSW and DNSPs’ estimates are discussed below.
The UNSW Consultants have developed a ‘reference scenario’ that incorporates their best
central estimates of key uncertainties. They term this a scenario because there are currently
significant uncertainties that mean this estimate of Scheme tariff payments will almost
certainly prove wrong. However, the upper generation scenario does provide a basis for
comparison with each DNSP’s data provided to Treasury for the 2011-12 NSW State Budget.
a Scheme cost that is unlikely to be exceeded
b Scheme cost that is unlikely to be lowered
1.75
1.44
1.25
1.75
1.05
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
DNSPs
UNSW reference
upper generation
UNSW reference
lower generation
UNSW
maxium (a)
UNSW
minimum (b)
Solar Bonus Scheme cost estimate cost - $billion (nominal)
probable cost
range
20
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
Cost Estimate
(nominal)
$ billion
Comments
DNSPs
1.75 Estimate at June 2011. UNSW assume that DNSPs will
continue to update these models and this estimate may
change.
UNSW reference upper
generation (‘reference’
scenario)
1.44 The most appropriate basis for comparison with alternative
modelled outcomes.
The reference scenario uses PV system generation estimates
that allow for some loss due to suboptimal orientation and
shading. It differs from some of the alternative models used
for Government in the assumed PV system performance,
allowance for net metering, different system sizes, inclusion
of some participant churn from 60 cents to 20 cents
and some
PV system performance degradation over the life of the
scheme.
UNSW reference lower
generation (‘reference –
low generation’ scenario)
1.25 There are significant concerns about actual average
performance of systems installed under the Scheme but little
firm data to make a formal assessment.
This estimate is based on average PV performance of
1,175 kWh/kW/yr, which is a 15 per cent reduction compared
with typical ‘good’ PV systems (correctly oriented, tilted and
largely unshaded).
This performance uncertainty can be significantly reduced
once installed PV systems under the Scheme have settled
and sufficient systems have provided a year or more of
metered data.
UNSW maximum (‘upper
bound’ scenario)
1.75 It is unlikely that the Scheme tariff costs will exceed this
unless there has been some significant misreporting of PV
system sizes, or NSW experiences highly clear weather.
UNSW minimum (‘lower
bound’ scenario)
1.05 It is unlikely that Scheme costs will be less than this unless
average PV system performance is significantly worse than
currently appreciated, or NSW experiences highly cloudy
weather.
Differences between the UNSW and DNSPs’ modelling
The main reasons for the differences between the UNSW and DNSPs’ modelled figures are
attributable to the following significant items:
• DNSPs’ assumed high level of annual generation. This has increased the cost estimate by
$152 million
• DNSPs’ use of constant cost data throughout the life of the Scheme added a further
$56.8 million
• one DNSP’s assumption that all connections are gross metered adds another
$45.9 million.
Together these items total $255 million and represent approximately 82 per cent of the
variance between the UNSW reference scenario and the DNSPs’ cost estimate reported
above. These confirm why differences occur and therefore confirm that UNSW has a model
that can be relied upon to produce a range within which the projected Scheme tariff payments
will be, namely $1.05 billion to $1.75 billion.
21
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
Total all DNSPs DNSP
#
1 DNSP
#
2 DNSP
#
3
Year UNSW
reference
($m)
DNSPs
($m)
UNSW
reference
($m)
DNSP#1
($m)
UNSW
reference
($m)
DNSP#2
($m)
UNSW
reference
($m)
DNSP#3
($m)
Jun 2011
142.1 142.1 57.6 57.6 46.8 46.8 37.7 37.7
Jun 2012
246.2 292.9 99.4 130.6 74.3 78.0 72.5 84.3
Jun 2013
247.5 292.9 100.1 130.6 72.4 78.0 75.0 84.3
Jun 2014
240.2 292.9 97.2 130.6 70.1 78.0 72.9 84.3
Jun 2015
233.2 292.9 94.4 130.6 68.0 78.0 70.8 84.3
Jun 2016
226.5 292.9 91.6 130.6 66.0 78.0 68.9 84.3
Dec 2016
103.5 146.5 41.9 65.3 30.1 39.0 31.5 42.2
Scheme
total 1,439.2 1753.1 582.2 775.9 427.7 475.8 429.3 501.4
Differences 313.9 193.7 48.1 72.1
2.2.2 Uncertainty within projections
There are a range of uncertainties associated with the Scheme that may have significant
implications for the projected and actual cost of the Scheme over each year to the end of the
calendar year 2016. Most of the uncertainty is reducing with the closure of the Scheme and
the settling out of eligible PV capacity. Other uncertainties might also be reduced with some
analysis. I detail some of the key uncertainties, their potential impact on current estimates of
the Scheme tariff payments and possibilities to reduce uncertainty as follows:
• climate variation in solar insolation
• real world performance of PV systems
• participant churn
• approval queue
• participant transition to gross feed-in tariff (FiT)
• participant transition to net FiT
• registration/data/metering errors
• PV system degradation
• percentage of generation exported.
0
50
100
150
200
250
300
350
Jun 2011
Jun 2012
Jun 2013
Jun 2014
Jun 2015
Jun 2016
Dec 2016
$m
Total Solar Bonus Scheme Tariff Costs Estimates
UNSW ‘reference’ estimate
DNSPs estimates
22
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
Uncertainty
Potential impacts on current Scheme estimates
Options to reduce
uncertainty
Highly Significant Variables
Panel
Performance -
Difference in
average
generation per
kW
Typical estimates of PV performance are for well installed
systems with reasonable orientation, tilts and no shading.
There are no available estimates for overall NSW
residential PV systems that the UNSW consultants are
aware of at present. Some limited assessment on Sydney
PV systems in DNSP#2 and DNSP#3 regions suggests
average performance may be significantly less than for
‘good’ systems. UNSW consultants suspect that average
performance may be up to 20 per cent less than for the
‘good’ systems used in modelling, including our reference
scenario. This would reduce Scheme costs by some
18 per cent against current estimates. There is also some
limited possibility that systems might perform better than
this typical ‘good’ system (for example, from improved
efficiency of inverters over the last few years).
As systems are
connected and a year of
metered performance
data is collected, actual
tariff
payments made by
DNSPs could provide
useful guidance on
actual average PV
performance. This could
also be ‘tuned’ with
respect to actual solar
insolation for that year
to provide better
estimates looking
forward.
Weather -
Variation in solar
insolation
Research suggests an approximate 20 per cent annual
range of solar insolation in Australia (Trocolli and Davy,
2010). UNSW Consultancy modelling using SAM for PV
system generation over 1998-2010, similarly, found a +/–
10 per cent range around approximately
400
kWh/kW/year. Uncertainty hence with respect to both
increased and decreased the Scheme tariff payments of
perhaps +/–10 per cent for each year of the scheme.
Likely less an issue over full the Scheme life unless
annual variability has significant longer (multi-year)
patterns.
Long-term climatic
forecasting may be able
to assist (El Nino and
La Nina patterns may
be relevant).
Less Significant Variables
Churn from
60 cents net to
60 cents gross
The 60 cents gross tariff is far more financially attractive
than a net 60 cents tariff for all scheme participants with
significant loads (that reduce exports under a net tariff).
Nevertheless, as of June 2011, three to four per cent of
60 cents systems are still on net metering. It is unclear
how many and at what rate these might tr
ansition to gross
tariff, or the impact of this (those on net may have virtually
no load at that metering point hence little change in the
Scheme tariff payments received). Based on DNSP#1’s
advice, I have assumed 10 per cent per year. Current
models generally assume that all eligible generation is, or
will transition to, gross so the impact would be to reduce
Scheme costs.
Actual data will become
available as the
Scheme progresses.
Churn from
60 cents to
20 cents
PV systems on houses that are sold over the period of the
Scheme transition from 60 cents to 20 cents tariff. Typical
estimates of three to five per
cent annual household churn
suggest impact might be significant,
although the Scheme
participants may not represent typical households
because they made a long-term investment decision
installing PV. Note that churn rates of two to five
per cent/year might see some nine to 21 per cent of
scheme participants transition to 20 cents by 2016
reducing scheme costs in that final year by some seven
to15 per cent.
Actual data will become
available as the
Scheme progresses.
Percentage of
generation
exported
The per cent of generation exported will vary between
households and is influenced by both system size and the
nature of the load profile. This is only relevant to net
tariffs.
Actual data will become
available as the
Scheme progresses.
23
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
Sensitivity analysis applying UNSW Consultant’s judgments of reasonable high/low variations
to key uncertainties has highlighted which of these uncertainties are likely to be the most
significant for estimates of overall system costs.
Uncertainty UNSW
Consultants
reference
scenario
Sensitivity range assessed (+%/-%) Impact on ‘reference’
scenario costs
($)
Panel
Performance -
Difference in
average
generation per
kW
0%
(assumed
average
NSW PV
performs at
level of a
‘good
system’
+ 5/–20% around 1382 kWh/kW/year + $65m to-$259m
Weather -
Variation in
solar
insolation
0% (typical
year)
+/-5% around ’Typical Mean Year’ values
(note that modeling for overall scheme tariff
payments representing six years – year to
year variability of +/–10% previously noted
likely to overestimate variability over six
years)
+/–$65m
Churn from
60 cents net to
60 cents gross
10%/per
year
0% to 75% of those on 60 cents net
transition to 60 cents gross each year
+$94.4 to–$42.9m
Churn from
60 cents to
20 cents
5% 1% to 7%/year –$10.5m to+$28.5m
Percentage of
generation
exported
50% 10% to 60% of PV generation exported and
hence eligible for net Feed-in tariff.
–$36.8m to+$9.2m
Annual churn
from 20 cents
net to 20 cents
gross
0%/per year 0% to 25% of current net 20 cents tariff
participants transition to gross 20 cents tariff
each year
$0m to +$5.9m
PV system
degradation
0.8%/per
year
0% to 1.6% per year (reflecting PV panel
degradation but also potential other reasons
for system performance to degrade such as
dirty panels, growth of trees that shade
system)
+/– $0.9m
Annual churn
from 20 cents
gross to 20
cents net
0%/per year
0% to 25% of current gross 20 cents tariff
participants transition to net 20 cents tariff
each year
$0m to -$22.6m
Approval queue
conversion to
connections
100% UNSW Consultant model assumes that all
currently approved but unconnected
systems are connected by November 2011
2.3 History of projected Scheme costs
The following graph summarises different cost estimates for the total nominal tariff payments
from when the Scheme was established and noting the changes in Scheme features.
The forecast assumptions and variables used for each cost projection differ, but demonstrate
the significant fluctuation in cost projections for the Scheme within the period of two years.
24
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
a NSW Government, NSW Feed-in Tariff Taskforce Report to Ministers, February 2009. (being Taskforce
scenarios, not forecasts)
http://www.industry.nsw.gov.au/energy/sustainable/renewable/solar/solar-scheme/established
b External consultant, NSW Solar Bonus Scheme Advice Forecast Scheme Capacity and Costs, prepared for
Industry and Investment NSW, 5 November 2009 (not publicly available)
c Industry and Investment NSW, NSW Solar Bonus Scheme Statutory Review Report to the Minister for Energy,
October 2010. The cost estimate of $3.989 bilion was provided by an external consultant.
http://www.industry.nsw.gov.au/__data/assets/pdf_file/0015/360141/Solar-Bonus-Scheme-Review-Report.pdf
d External consultant, Solar Bonus Scheme, Forecast NSW PV Capacity and Tariff Payments, prepared for
Industry and Investment NSW, 25 October 2010
http://www.industry.nsw.gov.au/__data/assets/pdf_file/0016/360142/AECOM-REPORT-for-Solar-Bonus-Scheme
-Review.pdf
e Industry and Investment NSW, NSW Electricity Network and Prices Inquiry, Final Report, December 2010
http://www.dpc.nsw.gov.au/__data/assets/pdf_file/0005/118904/NSW_Electricity_Network_and_Prices_Inquiry_
Report.pdf
f DNSPs’ cost estimates underpinning the 2011–12 NSW State Budget
None of the capital and working capital costs associated with the Scheme (for example
meters, administration and other implementation costs) have been included in the total tariff
cost of the Scheme outlined above.
$449
$318
$362
$3,989
$1,954
$1,830
$1,750
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
Feb 2009
7 yr 60c
Gross
a
Nov 2009
20 yr 60c
Net
b
Nov 2009
7 yr 60c
Gross
b
Oct 2010
7 yr 60c
Gross
c
Oct 2010
7 yr 60 & 20c
300MW cap
Gross
d
Dec 2010
7 yr 60 & 20c
300 MW cap
Gross
e
Sept 2010
7yr 60 and
20c
300 MW cap
Gross
f
Total Scheme Payments ($ million)
Estimated Nominal Scheme Costs
25
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
2.3.1 Review of models underpinning the different projected costs of the
Scheme
UNSW reviewed the models that were used to forecast the different projected costs of the
Scheme over its life. Overall, I consider these models to be satisfactory. Comments on the
individual models follow (Refer Appendix 5 (a) for complete table and Appendices 5 (b) to (d)
for comment on each DNSP’s model):
Report Model UNSW Consultant comments
a
NSW
Feed-in
taskforce
(2009)
Estimated the Scheme costs for different
tariff settings given two assumed
scenarios of deployment as part of
scheme design.
A particularly challenging modelling task at
the early stages of Scheme design. PV
deployment not modelled as such, instead
assumed. PV generation appears to have
been estimated in house.
b
AECOM
Advice
(2009)
Estimated future deployment and the
Scheme costs for two tariff options as part
of Scheme design.
Review of this modelling limited and based
on discussions with stakeholders that were
intended to better understand the AECOM
(2010) modelling. PV uptake model was
an earlier version of AECOM (2010) model
and incorporated 2009 PV system price
estimates. PV performance model
apparently the same as AECOM (2010),
as was eligible generation estimate.
c & d
AECOM
(2010)
Estimated future deployment and the
Scheme costs given early experience with
scheme and 50 MW review trigger.
Scenarios of possible tariff changes
investigated.
Estimated Scheme costs of $2.7 billion
(60 cents gross, seven years).
PV deployment model based on modified
NPV function using statistical tariff from
historical Australian PV uptake.
Incorporated much lower installed PV
system costs than earlier models given
2010 data. PV performance model
provided by AECOM in-house research
group.
e& f
DNSP#1
(ongoing)
Two separate budget impacts models:
one used for input to Parry Duffy and one
used for ongoing reporting to DTIRIS.
In-house budgeting models based on
system application and connection data.
PV performance based on DNSP#1’s
consultant advice. Ongoing model
development
and tuning given the Scheme
closure and system application and
connection data.
e & f
DNSP#2
(ongoing)
Three separate budget impacts models:
one used for input to Parry Duffy, one
used for ongoing reporting to DTIRIS and
one provided as including the latest data.
In-house budgeting models based on
system application and connection data.
PV generation based on Clean Energy
Council data. Ongoing model development
and tuning given the Scheme closure and
system application and connection data
e & f
DNSP#3
(ongoing)
A budget impacts model provided by
Oakley Greenwood, that includes some
scope for modelling deployment.
Model used for projecting deployment and
internal budgeting and AER pass through
request. PV generation based on ORER
performance estimates. Ongoing model
development and tuning given the Scheme
closure and system installation and
connection data.
f
Combined
DNSPs’
(ongoing)
Estimated budget impacts spreadsheet
consists of three main components.
June 2011 spreadsheet suggests total
the Scheme tariff costs of $1.75 billion.
Component 1 sums ongoing DNSP
reporting of estimated tariff costs.
Component 2 summarises the Scheme
applications and connections with respect
to net and gross 60 cents and 20 cents
Scheme participants.
Component 3 presents various scenarios
based on former Scheme options.
Reviewed by DNSPs’ consultant.
26
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
Recommendations
• The Government publish the total projected Scheme costs, within a range, based on a
sensitivity analysis of variables, including impact of weather patterns on the extent of
energy produced and paid for under the Scheme
• DTIRIS seeks medium term forecasting of El Nino and La Nina weather patterns and
shares this information with DNSPs to assist them in forecasting Scheme costs
• DNSPs continue to update total projected Scheme costs with actual data collected
• DNSPs use a standard approach to estimating and reporting Scheme costs.
3. Other matters considered relevant
As required by section 194(1)(c) of the Act, I am to review and report to parliament any other
matter I consider to be relevant.
This section looks at how the Scheme will be funded, and key aspects of the establishment
and administration of the Scheme. In particular, I examined whether:
• specific and measurable program objectives were set
• costs and benefits associated with achieving the objectives were assessed and reported
in the proposals put forward for decision making
• relevant risks to the achievement of each program objective were identified, assessed and
addressed
• there was an achievable implementation program
• there were control mechanisms to identify and react to the take-up of the scheme and
other changes that could affect the achievement of program objectives
• relevant and timely information was provided to Government decision-makers, potential
applicants and other stakeholders to take informed action at each stage in the
development and operation of the scheme
• there was an effective audit process to provide program assurance.
I did not seek to:
• question the merits of Government policy objectives
• duplicate any reviews already being conducted in relation to this topic.
My findings are based on discussions and documentation from the NSW Department of Trade
and Investment, Regional Infrastructure and Services (DTIRIS) (formerly the Department of
Industry and Investment NSW), the Office of Environment and Heritage (OEH)(formerly the
Department of Environment, Climate Change and Water), and Treasury. The then Industry
and Investment NSW chaired the NSW Solar PV Feed-in Tariff taskforce that originally
designed the Scheme, and continued to administer its implementation in conjunction with the
network businesses. I found that the roles and responsibilities of other agencies were less
clearly defined.
To guide my work, I have referred to principles of better practice including the NSW
Government’s:
• Guide to Better Regulation, 2009
• Gateway Review System
• Internal Audit and Risk Management Policy for the NSW Public Sector, 2009.
Whilst my report necessarily focuses on the actions of Government agencies, it is important to
recognise that the final decision making in relation to the Solar Bonus Scheme rested with the
New South Wales Government and the New South Wales Parliament.
27
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
3.1 How the Scheme will be funded
While the DNSPs have been making tariff payments for Scheme participants since
1 January 2010, they have not been reimbursed from the NSW CCF.
Electricity customers are likely to bear the costs of the Scheme via the NSW CCF levy
through higher electricity prices.
Sufficient funding to meet the costs of the Scheme will not be available through the NSW CCF
at the time reimbursements are due to the DNSPs. The Treasury has advised that the
difference is expected to be covered, in the interim, through the Government’s Consolidated
Fund as interest free repayable advances to the NSW CCF from the Crown. This will have a
short term impact on the Budget. The NSW CCF will then reimburse the Crown over a period
of time when positive cashflows are available. The actual time period is not certain due to the
relationship between future contributions into the NSW CCF and the actual cost of tariff
payments in the future.
The costs of the NSW CCF are ultimately funded by contributions from electricity and water
utilities and recovered by these agencies through network tariffs which apply to all customers.
In 2010-11, the DNSPs are required to contribute a total of $150 million to the NSW CCF. The
2011-12 NSW State Budget detailed the NSW CCF will require DNSPs to contribute
$250 million in 2012-13 and $400 million per annum from 2013-14 to 2016-17. The Parry
Duffy report states that this would be equivalent to about a one per cent increase in regulated
retail tariffs for electricity customers.
Sources: NSW Treasury April and June 2011; Combined estimates distribution businesses July 2011
Details of the operations of the NSW CCF will be reviewed in volume six of my 2011 Report to
Parliament.
62
33
162
294 294 294
147
72
261
131
0 0
0
0
0
50
100
150
200
250
300
350
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17
Funding ($ million)
Likely funding sources for the cost of total tariff payments
Amount funded from Climate Change Fund
Budget
28
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
3.2 Were specific and measurable program objectives set?
I found that the Scheme had three broadly stated objectives, with no specific targets against
which progress could be measured. These objectives do not include reducing emissions or
obtaining value for money.
The design taskforce was guided by three New South Wales Government policy objectives for
renewable energy. These were later incorporated in the legislation that introduced the
Scheme as follows:
• encourage and support persons who want to generate renewable energy as a response to
climate change
• develop jobs in the renewable energy sector by assisting renewable energy generation to
compete with non-renewable energy generation
• increase public exposure to renewable energy technology in order to encourage the whole
community to respond to climate change.
Source: NSW Electricity Supply Act 1995 – s.15A
While the number of jobs in the renewable energy sector could be estimated, the Scheme was
set no specific targets against which progress could be measured. Nor could levels of
encouragement and increased public exposure be measured.
The New South Wales Government’s Guide to Better Regulation, 2009 states that the
objectives of a regulatory proposal should:
• be clear, concise and specific
• directly target the root cause of the problem
• where possible, be measurable (e.g. by specifying an outcome and a time period over
which the objective is to be achieved)
• be consistent with existing Government objectives or policies.
The Scheme was not a regulation, but the principles are pertinent. It had no specific targets
against which progress could be measured.
There is a range of alternative measures available to address the policy objectives of
renewable energy, and the Scheme should have been assessed on its merits as one such
measure.
No objectives were set for:
• reducing emissions (as required by the NSW State Plan)
• reducing electricity consumption (as required by the NSW State Plan)
• minimising costs to Government or the consumer
• obtaining value for money.
Further to this:
• the Scheme was designed when the Commonwealth Government’s Carbon Pollution
Reduction Scheme (CPRS) was well advanced and expected to commence on
1 July 2011 and the taskforce was required to prepare its report taking into consideration
COAG’s Complementarity Principles for Climate Change Mitigation Measures (principles
designed to ensure jurisdictions’ mitigation measures complemented the CPRS). In this
context, the taskforce in its February 2009 Report to Ministers found that the Feed-in
Tariff scheme would not result in a net reduction in greenhouse gas emissions once the
CPRS was introduced
• the Scheme was designed to encourage generation of renewable electricity, rather than
cut consumption of electricity
• the taskforce considered that the Scheme would in fact increase costs to the consumer.
No specific
targets for
measuring
progress
29
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
3.3 Were costs and benefits assessed?
I was concerned to establish if costs and benefits associated with achieving the objectives
were assessed and reported in the proposals put forward for decision making.
There was some costing of scenarios and some assessment of benefits but, overall I found
that no cost-benefit analysis was undertaken before the Government’s decision in 2008 to
introduce a scheme. Likewise, no cost-benefit analysis was undertaken when changes were
made to the Scheme in 2009, or when changes were made to funding arrangements early in
2011.
The New South Wales Government’s Guide to Better Regulation, 2009 states that the impact
of Government action should be properly understood by considering the costs and benefits of
a range of options, including non-regulatory options.
The Government announcement that it would establish a feed-in tariff for New South Wales
was made without seeking preparation of a business case. While a taskforce was established
by the Government to consider options for the design of the Scheme, there was no
requirement to consider the merits of establishing a Scheme at all. It is not evident that there
was an adequate examination of the options, costs, time frames and risks associated with the
Scheme.
Major Government infrastructure projects are required to go through an independent Gateway
Review process which requires, amongst other things, economic and financial appraisals and
risk management plans. For infrastructure projects undergoing a Gateway Review, these
documents would be prepared by the policy agency and would have been submitted to
Treasury for assessment as part of the Budget Committee approval process. However, as the
Solar Bonus Scheme was not an infrastructure project, this was not a requirement.
In the context of its announcement that a feed-in tariff would be established, the Government
sought no additional economic appraisal or cost-benefit analysis. Had an equivalent process
to Gateway Review existed for major recurrent programs and assuming such a process was
complied with, the requirements would have been more stringent. With the advantage of
hindsight it is unfortunate that this level of scrutiny did not occur, considering the size and
impact of the Scheme.
In 2008, the Council of Australian Governments (COAG) had agreed to a set of national
principles for feed-in tariff schemes. This included a requirement to undertake analysis to
establish the benefits and costs of any subsidy against the objectives of that subsidy. The
difficulty in assessing the performance of the Scheme more closely was that, as discussed
earlier, there were three broadly stated objectives set for the Scheme but no specific targets
against which progress could be measured. No cost-benefit analysis was undertaken before
the Government’s decision on 18 November 2008 to introduce a feed-in tariff. With the
decision to have a feed-in tariff taking place without a cost-benefit analysis, there was no
assessment of whether the installation of small solar panels on the rooftops of houses was an
efficient or effective way for New South Wales to ensure an adequate and reliable energy
supply; whether it would reduce reliance on coal fired generation and also meet its climate
change objectives in a cost effective way. The taskforce did consider common reasons for
implementing feed-in tariff Schemes, including greenhouse abatement and ‘green job’ support
and, in their report to Government, noted the inefficiency of this mechanism to support either
of these goals in the New South Wales context.
No cost-benefit analysis was undertaken when changes were made to the scheme on
9 November 2009, or when changes were made to funding arrangements on
1 February 2011. For example, the NSW CCF provides for a range of different programs. The
programs are focused on reducing greenhouse gas emissions and the impact of climate
change. Examples include the Home Power Savings Program, Fridge BuyBack Program, a
home insulation program, and an Energy Efficiency for Small Business Program. Treasury
advised that, while the programs were due to wind down, the decision to use them to meet the
costs of the Scheme, rather than passing them directly through to electricity consumers, may
prevent or delay the establishment of equivalent new programs.
No business
case to
consider
merits of
Scheme
No cost-benefit
analysis
30
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
The New South Wales Government’s Guide to Better Regulation, 2009 states that
Government action is commonly justified on the basis of responding to market failures or
imbalances. It is important to determine whether there is a need for Government to be
involved, or whether the problem will be solved through market forces or by existing
regulations at the State or Commonwealth level.
I found that no market research was undertaken (including about non-tariff options) to
investigate customer motivations in generating renewable energy. The taskforce sought
submissions on the design of the tariff scheme and held a stakeholder forum.
The tariff was in addition to generous incentives provided by the Commonwealth’s Mandatory
Renewable Energy Target scheme, which also supported the installation of small scale solar
panel systems through the Solar Credits Scheme. The taskforce report acknowledged the
incentives that were in place through Commonwealth programs, and included consideration of
these in their scenarios costings. The taskforce also considered the national feed-in tariff
principles agreed by all levels of Government at COAG. But I found no evidence of State
Government agencies liaising closely with Australian Government agencies on the design of
the Scheme.
As there was no clear assessment of the market need and the possible options and the costs
and benefits of addressing it, it is more difficult to assess how much of the Scheme’s
unexpectedly high take-up can be attributed to the financial support being offered. The NSW
Solar Bonus Scheme Statutory Review identified that the major reason for the rapid take-up
under the Scheme was that customers were receiving high returns due to a rapid decline in
the costs of purchasing solar PV generators since 2009. The Scheme may have been seen as
a business opportunity by many small investors, rather than just a means of tackling climate
change.
Costs were examined under differing scenarios, and there was some assessment of the
benefits of the Scheme in terms of the NSW Government’s policy objectives for renewable
energy.
Financial Assessments
Three major assessments were undertaken, as outlined below:
• The first assessment was in February 2009, when the NSW Solar PV Feed-in Tariff
taskforce of representatives from DTIRIS, OEH, Treasury and the Department of Premier
and Cabinet presented a financial analysis of ten scenarios based on alternative scheme
designs including net and gross tariff rates over seven year and 20 year periods. Given
limited time in which to report it assumed simple linear take-up rates, noting that this
represented a significant area of uncertainty. On 22 June 2009, on the basis of the
different scenarios presented in the taskforce report, the Government decided on a
60 cents/kWh net tariff over 20 years. The taskforce had presented a scenario which
could attract up to 202,900 participants and cost up to $856 million (calculated using net
present value at a five per cent discount rate). On 9 November 2009, again relying on the
scenarios in the taskforce report, the Government switched to a 60 cents/kWh gross tariff
over seven years. The taskforce report had included a scenario in which the Scheme
could attract up to 72,900 participants and cost up to $354 million (calculated using net
present value at a five per cent discount rate).
No market
research
Initial estimate
up to $354m
31
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
• A second assessment was commissioned by DTIRIS from specialist consultants and
presented to them on 5 November 2009, entitled NSW Solar Bonus Scheme Advice
Forecast Scheme Capacity and Costs. The report indicated that, rather than the seven
years gross scheme having a lower cost as in the previous assessment, the reverse
would more likely be the case. The report forecast the likely cost of a 60 cents/kWh gross
tariff over seven years to be $289 million, while a 60 cents/kWh net tariff over 20 years
was forecast to only cost $155 million (both calculated using net present value at a
seven per cent discount rate). This is because the report showed that under the net
scheme scenario, take-up was likely to be at the lower end of the range considered by the
taskforce, while under the gross scheme, take-up would be higher. The report also
identified the likelihood that solar panel installation costs would fall and indicated that
under a gross feed-in tariff scheme, a reduction in PV installation costs of 20 per cent
would result in an increase in system installations of approximately 50 per cent by 2016
respectively. Total premium feed-in tariff costs would increase by a similar proportion (at a
total cost of $433 million). It is apparent that not all agencies were aware of this report at
the time the Government made its decision to redesign the scheme from a net to a gross
tariff.
• The third assessment was in October 2010 when the DTIRIS NSW Solar Bonus Scheme
Statutory Review report reviewed the progress of the Scheme with the support of a
second report from the specialist consultant. The figures were revised sharply upward.
Installed capacity was forecast to grow to around 250 MW by June 2011 and to nearly
1,000 MW by the planned conclusion of the Solar Bonus Scheme in December 2016.
Unless changes were made, the Scheme was forecast to cost $2,701 million (calculated
using net present value at a seven per cent discount rate).
These cost assessments were based on the tariffs paid. They did not include associated
costs, such as network costs, customer costs, administration costs, avoided infrastructure
costs, network electricity savings.
Benefit Assessments
The reviews endeavoured to broadly assess the benefits of the scheme, but with varying
success. For example, in relation to the creation of new jobs:
• The 2009 NSW Feed-in Tariff Taskforce Report stated that the number of new jobs
created by the Scheme would be dependent upon the rate of take-up. It noted that the
Commonwealth’s Carbon Pollution Reduction Scheme and the expanded Renewable
Energy Target would also be providing incentives and support for the expanded take-up of
solar PV. So any new jobs created would result from a combination of these three
separate incentives. As a result, identifying the relative contribution of each of these
policies to the number of jobs created was not possible. Furthermore, the taskforce
identified that the subsidy cost per job created was likely to be very high, demonstrating to
the Government that a feed-in tariff was unlikely to be an efficient mechanism for
supporting industry development.
• The 2010 statutory review reported that benefits to the renewable energy industry and
jobs growth were reported to be mixed. On the one hand, there was strong growth in
installation jobs. On the other hand, some large generation companies, including those
with renewable energy portfolios, had raised concerns. They considered that the amount
of new generation when combined with the Commonwealth’s renewable energy multiplier
for small generation had delayed investment in large-scale renewable projects.
3.4 Were relevant risks identified, assessed and addressed?
I was concerned to establish whether relevant risks to the achievement of each program
objective were identified, assessed and addressed.
I found that little was done early enough to identify and reduce relevant risks. I found no
contingency planning, analysis and assessment of options and exit strategies to
address potential high risk situations. A risk management plan was not finalised until
30 September 2010, ten months after the Scheme had commenced.
Not all agencies
were aware of a
second report
Little was done,
early enough,
to identify and
reduce risks
32
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
Significant external factors contributed to a faster than anticipated take-up of the scheme.
These included a rapid decline in PV system prices since 2009, the strong Australian dollar,
and the compounding effect of Commonwealth Government schemes.
Early cost assessments had identified considerable uncertainty in relation to take-up rates,
quoting a wide range of possibilities. The specialist consultants’ report prepared for DTIRIS
NSW Solar Bonus Scheme Advice Forecast Scheme Capacity and Costs 2009 indicated that
PV costs were expected to continue to decrease. PV cost reductions in excess of 50 per cent
from 2007 levels had been predicted by individuals within the electricity industry
and that
within three to seven years, solar energy's unsubsidised cost to end users could approach the
cost of conventional electricity in a number of markets.
DTIRIS discussed risks arising from the high take-up with the Government. A Scheme risk
management workshop was held on 1 July 2010, and the Scheme was included on the
Industry and Investment NSW Risk Register on 17 August 2010. A risk management plan was
developed and risks advised to the Government, but not until 30 September 2010. It identified
a range of risks including the high rate of take-up, limited controls and accountability,
customers’ lack of understanding, industry capacity and safety. It pointed to the need for the
statutory review.
Had more stringent policy development protocols, such as those applied to infrastructure
projects, been in place and been complied with, a risk management plan would have been
required and submitted to Treasury for assessment as part of the Budget Committee of
Cabinet approval process. For infrastructure projects, such plans are required to contain
schedules that identify and allocate the risks associated with a particular project and detail
how each risk is to be managed.
3.5 Was there an achievable implementation program?
I found no overall implementation program, including no clear definition of project roles and
responsibilities of those involved in implementing and delivering the Scheme.
Interest in the Scheme was much greater than expected. DTIRIS had to respond to policy
decisions at short notice, and I have been advised that a restructure and staffing levels limited
its ability to respond quickly.
The Scheme was announced on 23 November 2008 and the taskforce was asked to produce
a report in January 2009 with recommendations on how best to implement a NSW feed-in
tariff scheme. It was anticipated that the new feed-in tariff would be introduced in mid 2009.
The Government did not announce the parameters for the Scheme until late June 2009.
By September 2009, DTIRIS had developed a Draft Implementation Framework. It described
the elements of the Scheme, but did not address the likely activity levels, performance
measures, time frames, or resources required.
While the design of a feed-in tariff was a policy matter, implementing the program rested with
the DNSP, which already had frameworks for connecting solar systems to the grid and
reviewing the metering arrangements for these installations. This commenced in June 2009.
As the Government substantially redesigned the Scheme on 9 November 2009, including
switching from a net 20 year scheme to a gross seven year scheme, there was little time
before the Scheme’s planned commencement on 1 January 2010. Aside from the
administrative and operational changes needed, a gross scheme would provide higher tariff
payments and was expected to lead to a greater demand from the public. Agency advice
cautioned that such a substantial redesign so close to the planned commencement date was
administratively complex, and proposed delaying the scheme’s start for 6 months to allow the
implications of this decision to be better understood.
By 11 December 2009, DTIRIS had consulted with Department of Premier and Cabinet,
Treasury, OEH, the electricity distributors and the Clean Energy Council to develop a draft
regulation.
No overall
implementation
program
33
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
Electricity retailers were given six months to make the required billing system changes to put
tariff payment information on customer bills. These were to take effect by 1 July 2010.
Transitional metering arrangements were also put in place to 1 July 2010 to allow the
networks sufficient time to source gross meters. However, as not all customers were able to
install gross meters, the legislation passed in October 2010 provided for customers to join the
Scheme with either net or gross metering arrangements.
3.6 Were there adequate control mechanisms?
I was concerned to establish whether there were control mechanisms to identify and react to
the take-up of the Scheme and other changes that could affect the achievement of program
objectives.
I found that the Scheme lacked the most elementary operational controls.
There was initially a poor monitoring system. There was a time limit of 2016, but initially there
was no cap on total Scheme capacity and costs. At least one agency expressed concern that
the gross tariff option presented an increased risk that Scheme costs would blow out, and that
the total cost of the Scheme should be capped.
The New South Wales Government’s Guide to Better Regulation, 2009 states that it is
important that agencies develop performance indicators based on the objectives of
regulations.
There were no overarching performance measures or evaluation framework established for
the Scheme. As there was no budget, there were no budgetary controls. For example, there
was no table summarising the committed and remaining budgets as take-up progressed. Such
a tool helps control against the possibility of overspending. Without such control mechanisms
DTIRIS had limited ability to effectively administer the Scheme.
DTIRIS attributed the strong take-up of PV to the generosity of the Scheme, coupled with
Commonwealth Solar Credits, and a significant drop in system prices as the Scheme
commenced.
The design of the Scheme in a legislative framework could not be readily adjusted. There was
no ability to alter the Scheme parameters in response to changes in demand without
introducing amending legislation. There was no way of stopping the Scheme, other than by
asking Parliament to amend the legislation. This differs greatly from other schemes, such as
the Commonwealth Government’s insulation scheme, where controls can be put in place fairly
quickly as the scheme is not legislated.
This legislative amendment is what happened following the 2010 Review which recommended
a cap on total Scheme capacity as a mechanism for limiting Scheme costs.
But, even then, the Scheme was not completely ‘capped’. Rather, provision was made that
the Minister could declare that no further tariff payments be made–after he was satisfied the
total installed generating capacity had reached 300 MW.
Even though total connections and applications exceeded 300 MW during November 2010 the
DNSP could continue to accept applications. At 30 June, the total capacity of applications to
connect to the electricity grid was 371.6 MW.
3.7 Was information relevant and timely?
I was concerned to establish if relevant and timely information was provided to Government
decision-makers, potential applicants and other stakeholders to take informed action at each
stage in the development and operation of the Scheme.
I found that there were significant shortcomings in the provision of information to
Government decision-makers.
The taskforce provided the Government with relevant and timely information on scheme
design at the outset, although it was given relatively little time to do so. Subsequently, the
agencies provided their views to the Government and made sure it was aware of the issues in
relation to the Scheme, as their knowledge improved.
Scheme lacked
the most
elementary
operational
controls
Significant
shortcomings in
information
34
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
However, by the time the Government was considering the switch from a net tariff to a gross
tariff, the taskforce’s report was no longer as relevant. DTIRIS had commissioned a more
thorough expert report to assess the Scheme costs so that the network businesses could
develop ‘cost pass through’ applications to submit to the Australian Energy Regulator (AER).
While this report was available, I have seen no evidence that it was considered. Copies were
not provided to agencies to allow them to consider its findings in providing advice to the
Government. Treasury advised me that the report was not sighted by them until after the
Government’s decision to switch to a 60 cents gross scheme.
With the advantage of hindsight, there were insufficient review points based on applications
so that progress could be assessed and so that the Scheme could be amended or even
stopped, if necessary.
The taskforce identified the need for regular reviews, pointing to likely reductions in the price
of solar PV systems and changes in the level of Commonwealth Government subsidy. I found
that one review had been planned-under the Electricity Supply Act 1995 the Minister for
Energy was required to review the Scheme when capacity reaches 50 MW or in 2012,
whichever occurred first. This was similar to reporting requirements built into equivalent
schemes in other Australian jurisdictions. The capacity was reached by mid 2010, the installed
capacity had doubled to 100 MW.
DTIRIS asked the electricity distributors to report six monthly on the take-up rate in
accordance with the requirements of the legislation. In the light of the rapid growth of the
Scheme and the lag between applications and installations, this information was subsequently
realised to be far from timely.
The distributors were asked early in 2010 to report more frequently as DTIRIS took action to
address the growing seriousness of the problem. Reports were requested monthly, and then
fortnightly. There were concerns about data quality. Despite rapid take-up, accurate
expenditure data on tariff payments was not available. Reliable expenditure data lagged
connection data reflecting the variability in solar system generation rates due to factors such
as weather patterns, location of panels and three-monthly billing cycles.
No forecasting model was being used. There were no systematic and regularly updated
forecasts of take-up and expenditure against expectations.
DTIRIS commissioned the statutory review, but four months elapsed after reaching the
50 MW review trigger before the results and recommendations to cap the Scheme were
brought to the Government. During this time the connected capacity doubled (from 50 MW to
101 MW). This appears to have been the first time the Government was provided with a
financial assessment of the Scheme’s high take-up rate.
A wide range of information was provided to applicants. But some potential applicants were
left confused by the sudden changes to the Scheme due to the complexity of eligibility
requirements, with a long transitional period in place for those still wishing to apply for the
60 cent tariff.
In January 2011, DTIRIS provided information to Scheme participants (including energy
retailers, industry bodies and consumers) about the application of the 300 MW cap and
implications for future participants. Information was provided on its website that due to the
cap, some applicants may not be eligible to receive the tariff if the cap is reached. It provided
further advice that consumers should seek to ensure that all deposits paid to solar panel
providers/installers were refundable.
On 13 January 2011, DTIRIS issued a media release advising that more than 300 MW in
applications to join the Scheme had been received by the businesses, and therefore there
was no guarantee that any future applications would be eligible to participate in the Scheme.
This media release was followed up by advertisements placed in industry publications.
Insufficient
review points
No regular
forecasts of
take up and
expenditure
35
NSW Auditor-General's Report
NSW Solar Bonus Scheme
KEY FINDINGS
3.8 Was there an effective audit process to provide program assurance?
I found that at the outset, there was no audit process to provide program assurance.
Potential areas of concern could have included regulatory compliance (by applicants,
installers and distributors), fraud, safety, accuracy of reporting against take-up, accuracy of
tariff payments and accuracy of reporting against the Scheme’s objectives.
The 2010 statutory review provided the only relatively independent assessment of the
effectiveness of the Scheme. The review suggested developing a suitable compliance and
safety regime that takes into account the nature and level of risk.
A regulatory compliance program has now begun, but it is initially only looking at applications
made in the transition period from 27 October 2010 to 18 November 2010, as this is
considered to be a period of significant risk.
The DNSPs and retailers had existing programs to meet the regulatory compliance
requirements of their operating licences.
A safety review is being conducted by NSW Fair Trading to assess the quality of installations.
Preliminary results of an audit by NSW Fair Trading inspectors of 658 solar panel systems in
Sydney’s north-west found that:
• 122 (18.5 per cent) had major defects related primarily to incorrect wiring
• 418 (63.5 per cent) had minor defects related to cabling, and marking and signage.
Recommendations
In relation to section 194(1)(c) of the Act, I recommend:
• Schemes that involve significant recurrent expenditure and economic costs to consumers
should be assessed in a similar fashion to major Government infrastructure expenditure.
Prior to approval, major programs need:
− specific objectives that are measurable
− a cost-benefit analysis
− an economic analysis
− a business case including options, costs, time frames and risks
− a risk assessment and risk management plan
− a budget
− an implementation plan
− a performance monitoring framework
− an exit plan
• in the interests of promoting transparency, and ensuring that Government decision-
makers have access to adequate information on such schemes, such assessments
should be made public
• the performance of such schemes needs to be monitored more closely. Agencies need to
be prepared to step in and take corrective action more quickly if important implementation
risks begin to materialise
• there needs to be an ability and a willingness to alter the parameters of such schemes
more quickly, particularly if there is a risk that they will not meet their objectives.
No audit
program to
provide
assurance early
on
36
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 1: Chronology of the Key Scheme Events
Date
Event
24 November 2008
Former Minister for Energy and the Minister for Climate Change and the
Environment announced Scheme, with the expectation to commence in
mid-2009.
Late 2008
NSW Feed-in Tariff taskforce was established to advise the NSW
Government on the design of a feed-in tariff scheme.
February 2009
NSW Solar Feed-in Tari
ff Report to Ministers. It contained financial analysis
of ten different scheme scenarios. ‘The taskforce undertook an analysis of
simple financial calculations based on assumed uptake rates’.
23 June 2009
NSW Government publicly announced (but did not le
gislate) details of a new
Scheme: a net feed-in tariff paying eligible households 60 cents per kWh for
renewable energy that is fed back into the grid that will operate for 20 years
,
starting 1 January 2010.
5 November 2009
Industry & Investment NSW (now known as the Department of Trade &
Investment, Regional Infrastructure and Services) received final modelling
which estimated that a 20 year, 60 cents net tariff scheme would cost
$318 million at nominal value (net present value $155 million) and the costs
of a seven year, 60 cents gross tariff scheme would cost $362 million at
nominal value (net present value $289 million). (External consultant, NSW
Solar Bonus Scheme Advice Forecast Scheme Capacity and Costs,
prepared for Industry and Investment NSW, 5 November 2009 - not publicly
available). Treasury advised us that the report was not sighted by them until
after the Government’s decision to switch to a 60 cents gross scheme.
10 November 2009
NSW Government announced the final scheme design whereby the
Scheme runs for seven years, paying a gross feed-in tariff of 60 cents per
kWh with a Statutory Review required when 50 MW capacity is reached or
2012, whichever came first. Commencement date of 1 January 2010
remained unchanged.
1 January 2010
Scheme commenced.
Mid-2010
Scheme reached the 50 MW installed capacity.
24 August 2010
The then Minister for Energy announced the statutory review and invited
public submissions until 30 September 2010.
October 2010
Reaches largest amount of installed capacity of any state at 101 MW.
27 October 2010
The then Premier announced outcomes of the statutory review of the
Scheme. Legislation is passed to introduce a reduction in the tariff rate to
20 cents per kWh from midnight and 300 MW cap. The estimated total
nominal cost of tariff payments of up to $2.0 billion (net present value of
$1.4 billion) (external consultant, Solar Bonus Scheme, Forecast NSW PV
Capacity and Tariff Payments, prepared for Industry and Investment NSW,
25 October 2010).
18 November 2010
Last day for customers that had purchased or entered into a binding
contract to purchase a PV system by 27 October 2010 to have lodged an
application for connection to the grid in order to be eligible for the 60 cents
gross scheme.
December 2010
The estimated total nominal cost of tariff payments of up to $1.8 billion.
(Industry and Investment NSW, NSW Electricity Network and Prices Inquiry,
Final Report, December 2010).
Appendices
37
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Date
Event
13 January 2011
Industry and Investment NSW press release advising customers that
applications may not be eligible if not connected prior to Scheme cap being
reached and the Scheme closed as the applications had reached 300 MW.
Customers advised to seek refundable deposits.
1 February 2011
The then Premier announced the New South Wales Government will ‘off-
set
the full impact of the Scheme on household electricity bills by redirecting
uncommitted funds from the NSW CCF, as well as drawing $55.0 million in
efficiencies from the electricity network business’.
4 April 2011
The Premier committed to a Solar Summit within the 100 Day Action Plan.
27 April 2011
The Government
released an independent assessment of the integrity of the
NSW Budget financial estimates which included some commentary on the
Scheme costs.
29 April 2011
Minister for Resources and Energy announced a hold to new applicants
effective midnight on 28 April 2011.
6 May 2011
Solar Summit Stage One held.
13 May 2011
The New South Wales Government announced new changes to the
scheme, including:
- it remaining closed to new applicants effective midnight 28 April 2011
- all applications received before 29 April 2011 will be assessed and if
eligible, will receive tariff payments even if this breaks the 300 MW cap.
- households already receiving, or who applied (but are not yet connected)
for, the 60 cents tariff will receive a 40 cents tariff rate from 1 July 2011.
(However, this did not eventuate).
7 June 2011
The Government announced there would be no change to tariff payments
for customers already receiving, or about to receive, tariff payments under
the Scheme.
29 June 2011
The Premier announced
preliminary results of an audit by NSW Fair Trading
inspectors of 658 solar panel systems in Sydney’s north-west.
1 July 2011
The Minister for Resources and Energy announced an Independent Pricing
and Regulatory Tribunal review to determine a fair price for small-scale
generated solar assuming:
- no increase in electricity price
- no additional funding from the NSW State Government Budget.
Gazette notice published to formally close the Scheme and a regulation
made to allow customers who had submitted an application by midnight of
28 April 2011 to join the Scheme.
38
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 2: A Generous Scheme
One reason for the high cost of the Scheme is that it was a generous scheme.
The Scheme was generous and one of only two that paid a tariff on all electricity generated
(the other being in the ACT, which pays 45.7 cents a kilowatt hour), known as a ‘gross’ tariff.
All other states offer ‘net’ tariffs, which pay only for the excess power householders produce
above what they use.
Features of feed-in tariff schemes in Australia present at or near the time of the introduction of
the NSW Scheme.
Jurisdiction Length Type Rate per KWh Cap
NSW
End 2016 Gross 60 cents (until 27 Oct 10)
20 cents (thereafter)
300 MW cap
(introduced 27 Nov 10
and closed 1 July 2011)
ACT
20 years from
connection
Gross Initially 50.5 cents
Reduced to 45.7 cents
Rate for new systems
reviewed each financial
year.
240 MW (proposed 15
MW micro, 15 MW
medium, 210 MW large)
VIC
End 2024 Net 60 cents (equivalent to 39
cents gross)
100 MW or $10/year on
bills
SA
End 2028 Net Initially 44 cents (equivalent
to 28.6 cents gross)
Increased to 54 cents. Rate
is capacity-determined with
reduced rates for larger
increments.
60 MW (proposed)
QLD
End 2028 Net 44 cents (equivalent to
28.6 cents gross). Rate is
capacity-determined with
reduced rates for larger
capacity increments.
N/A
WA
10 years from
connection
Net 40 cents (equivalent to
26 cents gross)
N/A
Source: Industry and Investment NSW, NSW Solar Bonus Scheme Statutory Review Report, October 2010
39
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 3: Uncertainty within Projections
Uncertainty Potential impacts on current Scheme estimates Options to reduce
uncertainty
Highly Significant Variables
Panel
Performance -
Difference in
average
generation per
kW
Typical estimates of PV performance are for well
installed systems with reasonable orientation, tilts and
no shading. There are no available estimates for
overall NSW residential PV systems that the UNSW
consultants are aware of at present. Some limited
assessment on Sydney PV systems in DNSP#2 and
DNSP#3 regions suggests average performance may
be significantly less than for ‘good’ systems. UNSW
consultants suspect that average performance may be
up to 20 per cent less than for the ‘good’ systems used
in modelling, including our reference scenario. This
would reduce Scheme costs by some 18 per cent
against current estimates. There is also some limited
possibility that systems might perform better than this
typical ‘good’ system (for example, from improved
efficiency of inverters over the last few years).
As systems are
connected and a year of
metered performance
data is collected, actual
tariff payments made by
DNSPs could provide
useful guidance on actua
l
average PV performance.
This could also be ‘tuned’
with respect to actual
solar insolation for that
year to provide better
estimates looking
forward.
Weather -
Variation in solar
insolation
Research suggests an approximate 20 per cent annual
range of solar insolation in Australia (Trocolli and Davy,
2010). UNSW Consultancy modelling using SAM for
PV system generation over 1998-2010, similarly, found
a +/–10 per cent range around approximately
1,400 kWh/kW/year. Uncertainty hence with respect to
both increased and decreased the Scheme tariff
payments of perhaps +/–10 per cent for each year of
the scheme. Likely less an issue over full the Scheme
life unless annual variability has significant longer
(multi-year) patterns.
Long-term climatic
forecasting may be able
to assist (El Nino and La
Nina patterns may be
relevant).
Less Significant Variables
Churn from
60 cents net to
60 cents gross
The 60 cent gross tariff is far more financially attractive
than a 60 cent net tariff for all scheme participants with
significant loads (that reduce exports under a net tariff).
Nevertheless, as of June 2011, three to four per cent of
60 cents systems are still on net metering. It is unclear
how many and at what rate these might transition to
gross tariff, or the impact of this (those on net may
have virtually no load at that metering point hence little
change in the Scheme tariff payments received).
Based on DNSP#1 advice, I have assumed 10 per cent
per year. Current models generally assume that all
eligible generation is, or will transition to, gross so the
impact would be to reduce Scheme costs.
Actual data will become
available as the Scheme
progresses.
Churn from
60 cents to
20 cents
PV systems on houses that are sold over the period of
the Scheme transition from 60 cents to 20 cents tariff.
Typical estimates of three to five per cent annual
household churn suggest impact might be significant,
although the Scheme participants may not represent
typical households because they made a long-term
investment decision installing PV. Note that churn rates
of two to five per cent/year might see some nine to
21 per cent of scheme participants transition to 20
cents by 2016 reducing scheme costs in that final year
by some seven to15 per cent.
Actual data will become
available as the Scheme
progresses.
Percentage of
generation
exported
The per cent of generation exported will vary between
households and is influenced by both system size and
the nature of the load profile. This is only relevant to
net tariffs.
Actual data will become
available as the Scheme
progresses.
40
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Uncertainty Potential impacts on current Scheme estimates Options to reduce
uncertainty
Annual churn
between 20
cents
net and 20 cents
gross
The financial advantage of the 20 cents gross tariff
over the 20 cents net tariff is far less clear depending
on any premium payments on exported generation
from retailers, the proportion of generation
self-consumed and future electricity tariffs.
Actual data will become
available as the Scheme
progresses.
Participant
transition to net
Feed-in tariff
Some potential that will see gross 20 cents participants
transition to a net tariff. This would reduce Scheme
tariff payments but note that the impact is only
significant at very high churn rates since 20 cents tariff
payments only represent a small component of the
overall tariff payments.
PV system
degradation
As discussed above, PV system output reduces over
time. I have assumed a rate of 0.8 per cent per year.
This has little impact on the overall scheme costs.
This may become more
apparent over time but
could be difficult to
extract from other
reasons for reduced
output.
Registration/data
/metering errors
The very rapid uptake of the scheme has created major
data issues for the DNSPs including double
registrations (which aren’t necessarily resolved until
connection) but also some potential registration errors
(such as installed PV capacity, or gross versus net).
There may also be incentives for PV system owners to
under-report PV panel capacity, or increase it over time
(although note that this is illegal under the Scheme).
Impact on scheme costs could be in both directions.
Simple checks on
metered data of PV
systems by DNSPs
(similar to those
conducted on metered
consumption data)
should be able to identify
many potential
discrepancies.
Approval queue
At 30 June 2011, some 19 per cent of approved
applications for eligible systems still hadn’t been
connected. Likely that most will be connected
reasonably soon, and current modelling of scheme
costs assumes this. However, there are some reasons
why some approved systems might not transition to
actual deployment. Impact would be to reduce scheme
tariff payments.
Actual data will become
available as the Scheme
progresses.
41
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 4: Sensitivity Analysis for Uncertainty within Projections
Uncertainty UNSW
Consultants
reference
scenario
Sensitivity range assessed (+%/-%) Impact on ‘reference’ scenario costs
(%) ($)
Panel
Performance -
Difference in
average
generation per
kW
0%
(assumed
average
NSW PV
performs at
level of a
‘good
system’
+ 5/–20% around 1382 kWh/kW/year
+4.5/–18% + $65m to-
$259m
Weather -
Variation in
solar
insolation
0% (typical
year)
+/-5% around ’Typical Mean Year’
values (note that modeling
for overall
scheme tariff
payments representing
six years – year to year variability of
+/–10% previously noted likely to
overestimate variability over six
years)
+/–4.5% (of
reference
scenario cost
+/–$65m
Churn from
60 cents net to
60 cents gross
10%/per
year
0% to 75% of those on 60 cents net
transition to 60 cents gross each
year
+6.6% to –3% +$94.4 to–
$42.9m
Churn from
60 cents to
20 cents
5% 1% to 7%/year –0.7% to +2% –$10.5m
to+$28.5m
Percentage of
generation
exported
50% 10% to 60% of PV generation
exported and hence eligible for net
Feed-in tariff.
–2.6% to
+0.6%
–$36.8m
to+$9.2m
Annual churn
from 20 cents
net to 20 cents
gross
0%/per year 0% to 25% of current net 20 cents
tariff participants transition to gross
20 cents tariff each year
0% to –0.4% $0m to
+$5.9m
PV system
degradation
0.8%/per
year
0% to 1.6% per year (reflecting PV
panel degradation but also potential
other reasons for system
performance to degrade such as
dirty panels, growth of trees that
shade system)
–0.1% to
+0.1%
+/– $0.9m
Annual churn
from 20 cents
gross to
20 cents net
0%/per year 0% to 25% of current gross 20 cents
tariff participants transition to net
20 cents tariff each year
0% to–1% $0m to -
$22.6m
Approval
queue
conversion to
connections
100%
UNSW Consultant model assumes
that all currently approved but
unconnected systems are connected
by November 2011
Not considered
42
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 5(a): Review of Models
Overall PV uptake: For forward estimates the achieved rate and level of deployment is
clearly critical and, as noted above, highly uncertain. The use of historical trends or
fixed scenarios is of limited value when evaluating possible budget impacts of different
scheme designs (except if these include caps). A range of modified NPV and payback
approaches have been used to correlated financial outcomes to system owners with
levels of deployment. Despite their limitations, they may be the best available option at
present. For such forward looking estimates the UNSW consultants’ view is that such
approaches should, ideally and given sufficient time and resources, be carefully
undertaken with detailed and formal exploration of the potential uncertainties involved.
If no more systems can be installed under the Scheme then the maximum cost of the
Scheme is far easier to calculate, although there can be continuing uncertainties for
some time as the Scheme settles out.
PV performance and eligible generation: This depends on estimates of PV system
performance (kWh/kW) over time (typically annual), and what proportion of this is
eligible for the tariff. Engineering-oriented models are more likely to provide realistic PV
generation than those applying average insolation with simple system assumptions. For
overall scheme costs the key issue is aggregate performance of all eligible systems –
generally less than the expected performance of appropriately located and oriented
high quality systems. In the UNSW consultants’ view, it should be possible to develop
reasonable models of this although I am not aware of publicly available work in the
Australian context that has yet done this in any systematic manner.
43
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
The Scheme models reviewed by Consultants Two key aspects of the modelling were assessed – Uptake and PV performance
Report Models Overall Comments Uptake model Comment on
appropriateness of
uptake assumptions
PV performance
modelling
Assumed performance
(kWh/kW/yr)
Comments on
appropriateness of
performance
assumptions
UNSW best
estimate for typical
year (basis for
comparison).
Average of ORER,
PSyst and SAM for
Sydney in a ‘typical’
climate year
1,382 Using Sydney
modelled data with
estimated
performance from
two packages
averaged to give
final figure. ORER
estimate of 1,382
applies for all of
NSW other than the
Western region.
Note, this may still
underestimate
performance of
systems in some
areas of regional
NSW (See
discussion in
AECOM (2010).
44
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
The Scheme models reviewed by Consultants Two key aspects of the modelling were assessed – Uptake and PV performance
Report Models Overall Comments Uptake model Comment on appropriateness
of uptake assumptions
PV performance
modelling
Assumed
performance
(kWh/kW/yr)
Comments on
appropriateness of
performance
assumptions
NSW Feed-in
taskforce (2009)
Estimated the
Scheme costs for
different tariff
settings given two
assumed scenarios
of deployment as
part of scheme
design.
Estimated Scheme
costs of $91m to
$354.5m (60 cents
gross, 7 years), and
$193m to $856.9m
(60 cents net,
20 years).
A particularly
challenging
modelli
ng task at the
early stages of
Scheme design. PV
deployment not
modelled as such,
instead assumed.
PV generation
appears to have
been estimated in
house.
Estimated the
Scheme costs for
different tariff
settings given two
assumed scenarios
of deployment with
linear uptake. Low
growth scenario of
2,000 additional
systems per year,
high scenario with
10,000.
A simple spreadsheet
model. Forward looking
estimates of deployment
are extremely difficult,
hence not attempted.
Note that the two
deployment scenarios did
not factor in impact of
different tariff settings
despite their major
impact on financial
attractiveness of
deployment. This makes
interpretation of the
taskforce modelling
results and tariff costs
potentially challenging.
Uncertainty also not
managed in any formal
way beyond the two
deployment scenarios.
Modelling therefore
needs considerable
caution in interpretation.
PV generation
model appears to
be based on
in-house
calculations.
1662 (+20%)
Appears high.
ORER estimates of
system performance
used in deeming
RECs referenced in
report but
discrepancy
between these and
their own estimate
not addressed.
45
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
The Scheme models reviewed by Consultants Two key aspects of the modelling were assessed – Uptake and PV performance
Report Models Overall Comments Uptake model Comment on appropriateness
of uptake assumptions
PV performance
modelling
Assumed
performance
(kWh/kW/yr)
Comments on
appropriateness of
performance
assumptions
AECOM Advice
(2009)
Estimated future
deployment and the
Scheme costs for
two tariff options as
part of scheme
design.
Review of this
modelling limited
and based on
discussions with
stakeholders that
were intended to
better understand
the AECOM (2010)
modelling. PV
uptake model was
an earlier version of
AECOM (2010)
model and
incorporated 2009
PV system price
estimates. PV
performance model
apparently the same
as AECOM (2010),
as was eligible
generation estimate.
Model fits an
underlying
polynomial of NPV
against historical
data regarding
number of systems
deployed to estimate
potential future
uptake under
different Scheme
scenarios.
As noted earlier, UNSW
consultants have based
their assessment on
discussions regarding
this model (the model
itself was not available).
Approach taken relatively
common and, in our
view, appropriate.
However it does involve
significant assumptions
that require careful
explanation. Our
understanding is that the
model was used to
undertake a number of
sensitivity studies which
represents good
practice.
See AECOM
(2010).
1640
(+18.5%)for
most of DNSPs’
customer aprt
from
region calc.
for one DNSP
See AECOM (2010).
46
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
The Scheme models reviewed by Consultants Two key aspects of the modelling were assessed – Uptake and PV performance
Report Models Overall Comments Uptake model Comment on
appropriateness of
uptake assumptions
PV performance
modelling
Assumed
performance
(kWh/kW/yr)
Comments on
appropriateness of
performance assumptions
AECOM (2010)
Estimated future
deployment and
the Scheme costs
given early
experience with
Scheme and 50MW
review trigger.
Scenarios of
possible feed-in
tariff changes
investigated.
Estimated scheme
costs of $2.7 billion
(60 cents gross,
7 years).
PV deployment
model based on
modified NPV
function using
statistical tariff
from
historical Australian
PV uptake.
Incorporated much
lower installed PV
system costs than
earlier models
given 2010 data.
PV performance
model provided by
AECOM in-house
research group.
Revised model to
above using a
different polynomial
given early
experience with the
Scheme.
Additional
information from
early experience
with the Scheme,
and significant falls
in PV system costs
led to changes in
the PV deployment
model. Chosen
approach seems
appropriate.
Did some
sensitivity testing
on issues such as
PV system prices
and existence, or
absence, of a
carbon price which
represents good
practice.
PV generation
model from
inhouse research
group
1640 (+18.5%) PV generation model
from in-house research
group using a model that
does not appear
specifically designed for
PV performance
simulation. ORER
estimates of system
performance used in
deeming RECs
referenced in report, but
discrepancy between
estimates not
addressed. Resulting
performance estimate
appears high, although
consistent with original
taskforce work. AECOM
notes that they were
asked, and therefore
sought to provide an
‘upper bound’ estimate
of potential future
deployment and hence
the Scheme costs.
47
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
The Scheme models reviewed by Consultants Two key aspects of the modelling were assessed – Uptake and PV performance
Report Models Overall Comments
Uptake model Comment on
appropriateness of
uptake assumptions
PV performance
modelling
Assumed performance
(kWh/kW/yr)
Comments on
appropriateness of
performance assumptions
DNSP#1 (ongoing)
Two separate
budget impacts
models: one used
for input to Parry
Duffy and one used
for
ongoing reporting
to DTIRIS.
In-house budgeting
models based on
system application
and connection
data. PV
performance based
on DNSP’s
consultant’s advice.
Ongoing model
development and
tuning given the
Scheme
closure and
system application
and connection
data.
No uptake modelling
based on financial
impacts, only
‘assumed’ levels of
future uptake.
In-house budgeting
model based on
assumed levels of
uptake. Basis of
some key
assumptions unclear
and significantly
impact results.
Some limited
ongoing model
development and
tuning since the
Scheme closure.
Based on
DNSP#1’s
consultant’s
advice
1,767 (+28%) and
1,752 (+27%) (two
separate models)
Appears high.
DNSP
#
1advises that
figure was provided by
DNSP#1’s
consultants. They
have the particular
challenge of widely
varying climatic
conditions over their
franchise area
including regions with
the best insolation in
NSW. ORER has
1536 kWh/kW
estimate for western
NSW, however,
unclear where most
PV deployment is
occurring and used
value still seems high.
Modelling apparently
now being ‘tested’
according to actual
measured
performance of
systems.
48
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
The Scheme models reviewed by Consultants Two key aspects of the modelling were assessed – Uptake and PV performance
Report Models Overall Comments
Uptake model Comment on
appropriateness of
uptake assumptions
PV performance
modelling
Assumed performance
(kWh/kW/yr)
Comments on
appropriateness of
performance
assumptions
DNSP#2 (ongoing)
Three separate
budget impacts
models: one used
for input to Parry
Duffy, one used for
ongoing reporting to
DTIRIS and one
provided as
including the latest
data.
In-house budgeting
models based on
system application
and connection data.
PV generation
based on Clean
Energy Council data.
Ongoing model
development and
tuning given the
Scheme cl
osure and
system application
and connection data.
No uptake modelling
based on financial
impacts, only on
connections and
applications to date.
In-house
deployment model
based on system
application and
connection data.
Model reviewed by
DNSP#2’s
consultant. Some
ongoing model
development and
tuning since the
Scheme closure.
Modelling seems
appropriate given
DNSP context.
Based on data from
the CleanEnergy
Council.
1,423 (+3%)
Appears reasonable.
49
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
The Scheme models reviewed by Consultants Two key aspects of the modelling were assessed – Uptake and PV performance
Report Models Overall Comments
Uptake model Comment on
appropriateness of
uptake assumptions
PV performance
modelling
Assumed performance
(kWh/kW/yr)
Comments on
appropriateness of
performance
assumptions
DNSP#3 (ongoing)
A budget impacts
model provided by
Oakley Greenwood,
that includes some
scope for modelling
deployment.
Model used for
projecting
deployment and
internal budgeting
and AER pass
through request. PV
generation based on
ORER performance
estimates. Ongoing
model development
and tuning given the
Scheme
closure and
system installation
and connection data.
Limited uptake
modelling based on
industry capacity to
install and financial
returns STCs.
Oakley Greenwood
model used for
internal budgeting
and AER pass-
through request.
Some ongoing
model development
and tuning since the
Scheme closure.
Modelling seems
appropriate given
DNSP context.
Based on ORER
estimates
1,382 (+0%) Appears reasonable.
50
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
The Scheme models reviewed by Consultants Two key aspects of the modelling were assessed – Uptake and PV performance
Report Models Overall Comments
Uptake model Comment on appropriateness
of uptake assumptions
PV performance
modelling
Assumed performance
(kWh/kW/yr)
Comments on
appropriateness of
performance
assumptions
Parry Duffy
No uptake modelling
within the report as
such – instead used
budget estimates
that were requested
from DNSPs.
No guidance or direction
was given to DNSPs
when requested to
provide their current
cost estimates based on
take up rates and
applications received.
At
the time of Parry Duffy
the tariff had been
reduced from 60 cents
to 20 cents and it might
have been assumed
that additional tariff
applications would end,
or at least greatly
reduce in number. Does
not appear to have been
a formal process for
reconciling the
significantly different
estimates provided by
DNSPs that appear to
reflect different
assumptions and
context.
Not directly
applicable but a
value can be
backed out from
DNSP data
1,677 (+21%) See details on the
DNSP models used
to provide budget
estimates for Parry
Duffy. Backed out
estimates of
assumed PV
performance
(calculated from
300 MW cap and
$302m/yr) is
1,662 kWh/kW/yr
which seems high.
51
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
The Scheme models reviewed by Consultants Two key aspects of the modelling were assessed – Uptake and PV performance
Report Models Overall Comments
Uptake model Comment on
appropriateness of
uptake assumptions
PV performance
modelling
Assumed performance
(kWh/kW/yr)
Comments on
appropriateness of
performance
assumptions
Combined DNSPs
(ongoing)
Estimated tariff
payments
spreadsheet that
consists of three
main components.
June 2011
spreadsheet
suggests total the
Scheme tariff costs
of $1.75 billion.
Component 1 sums
ongoing DNSP
reporting of
estimated tariff
costs.
Component 2
summarises the
Scheme applications
and connections with
respect to net and
gross 60 cents and
20 cents the
Scheme participants.
Component 3
presents various
scenarios based on
former Scheme
options. Reviewed by
DNSPs’ consultant.
Model sections that
were reviewed did
not include any
uptake modelling as
such. Does includes
some scenarios
based on
predetermined
levels of
deployment
Cost estimates
based on ongoing
DNSP reporting of
their estimated the
Scheme costs
although actual
DNSP connection
and approval MW
for
60 cents and
20 cents the
Scheme participants
also collected, and
used to ‘back out’
PV performance
check for each
DNSP. Approach
seems reasonable
given the effective
closure of the
scheme. Modelling
spreadsheet
implementation
reviewed by
DNSPs’
consultant.
Not directly
applicable but does
‘backed out’
estimates
1,362.5 (-1.5%)
From network data
and appears
reasonable.
52
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 5(b): Comparison to DNSP#1 Scheme tariff payments
It can be seen that the values calculated by DNSP#1 are significantly greater than those using
the UNSW Consultant’s model for its ‘reference scenario’ – with the scheme total for the
former being $193.7 million higher.
The main reason for the difference between the UNSW and DNSP#1’s figures is the assumed
high level of annual generation (1752 kWh/kW/year). Increasing the annual generation to
1752 kWh/kW/year in the UNSW model increases the cost by $141 million.
The DNSP#1 model assumes that all the connections are gross metered. Incorporating this
assumption into the UNSW model increases the cost by $27.3 million.
Although the DNSP#1 model allows for churn from 60 cents to 20 cents over time, its use of
the 2011–12 figure for all years of the Scheme out to 2016 (pro rata for the last six months),
increases the estimated costs. Incorporating this assumption into the UNSW model increases
the cost by $21.9 million.
Removing the allowance for system degradation over time increases the cost by only
$300,000.
The DNSP#1 model assumes that all systems are 2.5 kW, which is slightly below the actual
average, and so incorporating a 2.5 kW size into the UNSW model decreases the cost by
$16.3 million.
The DNSP#1 model also does not allow for churn from net 60 cents to gross 60 cents,
however this is immaterial as they assume all systems are gross metered.
The aggregate effect of all changes is to increase the cost according to the UNSW model by
$186 million, bringing the total to $769 million, only $7 million less than according to
DNSP#1’s figures. This final difference simply reflects the slightly different rates of installation
in each month.
0
20
40
60
80
100
120
140
Jun 2011 Jun 2012 Jun 2013 Jun 2014 Jun 2015 Jun 2016 Dec 2016
$m
DNSP#1 - Solar Bonus Scheme Tariff Costs Estimate
UNSW ‘reference’ estimate DNSP#1 estimate
53
NSW Auditor-General's
Report
NSW Solar Bonus Scheme
APPENDICES
53
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
53
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 5(c): Comparison to DNSP#2 Scheme tariff payments
The DNSP#2’s values appearing in the graph and table below are similar to those produced
using the UNSW model for its ‘reference scenario’ – with the Scheme total for the former
being $48.1 million higher.
The main reason for the difference between the UNSW and DNSP#2’s figures is the
DNSP
#
2’s use of the 2011-12 figure for all years of the Scheme to 2016 (pro rata for the last
six months). Incorporating this assumption into the UNSW model increases the cost by
$27.8 million. The most recent version of the DNSP
#
2’s model assumes that systems on a
60 cents tariff average 2.35 kW and those on a 20 cents tariff average 2.07 kW. Incorporating
these sizes into the UNSW model increases the cost by $14.2 million.
Increasing the annual generation in the UNSW model to the 1,423 kWh/kW/year in the
DNSP#2 model increases the cost by $11.3 million. Removing the allowance for system
degradation over time increases the cost by only $0.2 million. The DNSP#2 model also
doesn’t allow for churn from net 60 cents to gross 60 cents, however this is immaterial as they
assume almost all systems are gross metered.
The aggregate effect of all changes is to increase the cost according to the UNSW model by
$56.9 million, bringing the total to $485 million, only $8.8 million more than according to
DNSP
#
2’s figures. This final difference simply reflects the slightly higher rates of installation
during 2011.
0
10
20
30
40
50
60
70
80
90
Jun 2011 Jun 2012 Jun 2013 Jun 2014 Jun 2015 Jun 2016 Dec 2016
$m
DNSP#2 - Solar Bonus Scheme Tariff Costs Estimate
UNSW ‘reference’ estimate DNSP#2 estimate
54
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 5(d): Comparison to DNSP#3 Scheme tariff payments
The values used by DNSP
#
3 are similar to those produced using the UNSW model – with the
scheme total for the former being $72.1 million higher.
The main reason for the difference between the UNSW for its ‘reference scenario’ and
DNSP
#
3’s estimates, is the assumption that all DNSP#3’s systems were gross metered.
Incorporating this assumption into the UNSW model increased the scheme cost by
$18.6 million.
The next greatest impact was caused by the use of the 2011–12 figure for all years of the
Scheme out to 2016 (pro rata for the last six months). This assumption increased the cost by
$10.6 million. According to DNSP#3’s consultant, an annual 60 cents to 20 cents churn rate of
3.5 per cent was assumed. When this is applied to the UNSW model (instead of 5 per cent),
the cost increases by $7.1 million. Removing the allowance for system degradation over time
increases the cost by only $230,000.
The aggregate effect of all changes is to increase the cost according to the UNSW model by
$35.5 million, bringing the total to $465 million, which is $36.6 million less than according to
DNSP
#
3’s figures.
0
10
20
30
40
50
60
70
80
90
Jun 2011 Jun 2012 Jun 2013 Jun 2014 Jun 2015 Jun 2016 Dec 2016
$m
DNSP#3 - Solar Bonus Scheme Tariff Costs Estimate
UNSW ‘reference’ estimate DNSP#3 estimate
55
NSW Auditor-General's
Report
NSW Solar Bonus Scheme
APPENDICES
55
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
55
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 6: Glossary
AECOM AECOM Pty Ltd
AER Australian Energy Regulator
CPRS (former) Carbon Pollution Reduction Scheme
DEWHA Department of Environment, Water, Heritage and the Arts
DNSP (NSW) Distribution Network Service Provider
DTIRIS NSW Department of Trade and Investment, Regional Infrastructure and
Services (Trade and Investment
eRET expanded Renewable Energy Target
FiT Feed-in-Tariff
IPART NSW Independent Pricing and Regulatory Tribunal
NPV Net Present Value
OG Oakley Greenwood Consulting
ORER Office of the Renewable Energy Regulator
PV Photovoltaic system
PVSyst PV studies, sizing and simulations software package
REC Renewable Energy Certificate
SAM Solar Advisor Model
SBS NSW Solar Bonus Scheme
STC Small scale Technology Certificates (within the eRET)
wrt with respect to
56
NSW Auditor-General's Report
NSW Solar Bonus Scheme
APPENDICES
Appendix 7: About the Audit
Background to Report
This Report stems from section 194 of Electricity Supply Act 1995 (the Act) requiring the
Auditor-General to commence a review after January 2011 and report on certain aspects of
the Scheme as soon as practicable after 1 July 2011.
Section 194 of the Act reads:
194 Review of solar bonus scheme by Auditor-General
(1) The Auditor-General is to review and report to Parliament on the following
aspects of the solar bonus scheme (being the scheme for the payment of
electricity supplied to the network by small retail customers using complying
generators):
(a) the number of small retail customers that have installed and connected
complying generators,
(b) the costs of the scheme including the total amount credited to small retail
customers under the scheme,
(c) any other matter that the Auditor-General considers to be relevant.
(2) The review is to be undertaken as soon as practicable after the period of 1
year from the commencement of section 15A.
(3) The Auditor-General is to report to each House of Parliament on the
results of the review conducted by the Auditor-General under this section as
soon as practicable after 1 July 2011.
(4) If a House of Parliament is not sitting when the Auditor-General seeks to
present a report under this Part, the Auditor-General is to present the report to
the Clerk of the House concerned.
(5) The provisions of section 63C (Documents presented to Clerk of House of
Parliament) of the Public Finance and Audit Act 1983 apply in relation to a
report presented to a Clerk of a House of Parliament under this section in the
same way as they apply to documents presented to a Clerk under that Act.
Acknowledgements
I gratefully acknowledge the co-operation and assistance provided by the Department Trade &
Investment, Regional Infrastructure & Services, AECOM, The Treasury, AusGrid, Endeavour
Energy, Essential Energy, Department of Premier and Cabinet, Office of Environment and
Heritage whose staff provided valuable explanation and information to us. Solar Energy
consultants, Dr Iain MacGill and Dr Robert Passey from the University of New South Wales,
provided expert advice and assistance throughout the audit. Their report to us is considered
our work paper and thereby not available for distribution.
Reliance on this Report
This Report has been prepared for the benefit of Parliament and my views and findings have
been formed in this light. While the New South Wales Parliament can rely on the findings in
this Report, they are not intended to provide any comfort to other parties. Parties are expected
to undertake their own examinations, and in any event, should not place any reliance that this
Report addresses their interests or concerns.
Professional people with purpose
audit.nsw.gov.au
The role of the Auditor-General
The roles and responsibilities of the Auditor-
General, and hence the Audit Office, are set
out in the Public Finance and Audit Act 1983.
Our major responsibility is to conduct
financial or ‘attest’ audits of State public
sector agencies’ financial statements.
We also audit the Total State Sector Accounts,
a consolidation of all agencies’ accounts.
Financial audits are designed to add credibility
to financial statements, enhancing their value
to end-users. Also, the existence of such
audits provides a constant stimulus to agencies
to ensure sound financial management.
Following a financial audit the Office issues
a variety of reports to agencies and reports
periodically to parliament. In combination
these reports give opinions on the truth
and fairness of financial statements,
and comment on agency compliance with
certain laws, regulations and Government
directives. They may comment on financial
prudence, probity and waste, and recommend
operational improvements.
We also conduct performance audits. These
examine whether an agency is carrying out its
activities effectively and doing so economically
and efficiently and in compliance with relevant
laws. Audits may cover all or parts of an
agency’s operations, or consider particular
issues across a number of agencies.
Performance audits are reported separately,
with all other audits included in one of the
regular volumes of the Auditor-General’s
Reports to Parliament – Financial Audits.
audit.nsw.gov.au
GPO Box 12
Sydney NSW 2001
The Legislative Assembly
Parliament House
Sydney NSW 2000
Pursuant to section 194(1) of the Electricity Supply Act 1995,
I present my report entitled Solar Bonus Scheme.
Peter Achterstraat
Auditor-General
7 November 2011
© Copyright reserved by the Audit Office of New South
Wales. All rights reserved. No part of this publication may
be reproduced without prior consent of the Audit Office of
New South Wales.
The Audit Office does not accept responsibility for loss or
damage suffered by any person acting on or refraining from
action as a result of any of this material.
Our vision
To make the people of New South Wales
proud of the work we do.
Our mission
To perform high quality independent audits
of government in New South Wales.
Our values
Purpose – we have an impact, are
accountable, and work as a team.
People – we trust and respect others
and have a balanced approach to work.
Professionalism – we are recognised
for our independence and integrity
and the value we deliver.
The Legislative Council
Parliament House
Sydney NSW 2000
Professional people with purpose
Making the people of New South Wales
proud of the work we do.
Level 15, 1 Margaret Street
Sydney NSW 2000 Australia
t +61 2 9275 7100
f +61 2 9275 7200
e mail@audit.nsw.gov.au
office hours 8.30 am–5.00 pm
audit.nsw.gov.au
New South Wales Auditor-General’s Report
Special Report
Solar Bonus Scheme
New South Wales Auditor-General’s Report | Special Report | Solar Bonus Scheme
