Climate Change Authority

You are here

Chapter 5: The small-scale Renewable Energy Scheme

Table of contents


This chapter considers the architecture of the Small-scale Renewable Energy Scheme (SRES) and in particular its uncapped nature and lack of a legislated end date. It considers the costs and benefits of different options for addressing these issues, including combining it with the Large-scale Renewable Energy Target (LRET), establishing a scheme end-date and capping the SRES. Finally, it considers potential enhancements to the clearing house and the utility of data currently collected by the Regulator.

5.1. History of the Small-scale Renewable Energy Scheme

Small-scale systems, in the form of solar water heaters and small generation units (small-scale solar photovoltaic (PV), wind and hydro), have been included in the Renewable Energy Target (RET) since its inception in 2001. Historically, the uptake of these systems was relatively low (see Figure 26). This changed in 2009 when the Commonwealth Government introduced Solar Credits to replace the Solar Homes and Communities Plan. Solar Credits was intended to provide an upfront capital cost subsidy worth around $7 500 by applying a multiplier – initially set at five – to certificates generated from small generation units (Swan et al, 2008). At the same time, system costs began falling rapidly, the value of the Australian dollar increased, and states and territories put in place generous feed-in tariffs (see Box 6). These factors contributed to a large increase in the installation of small generation units and, consequently, renewable energy certificates (exacerbated by the Solar Credits multiplier).

The rapid increase of certificates caused certificate prices to fall dramatically, creating uncertainty for both the large-scale and small-scale renewable energy sectors. The certificate price was not sufficient to support large-scale energy projects and investment stalled. Similarly, businesses selling small-scale systems were unable to provide accurate information to customers regarding the price they were likely to receive for certificates. As explained in Chapter 1, the Commonwealth Government responded by legislating to separate the RET into two schemes: the LRET – for large-scale projects – and the SRES, to assist households and businesses with the upfront costs of small-scale systems.

Figure 26 Number of installations of small-scale systems
This figure shows the number of annual installations of small-scale systems by technology type over the period 2001 to 2011. Since the Mandatory Renewable Energy Target commenced in 2001, almost all small-scale renewable technologies installed have been solar water heater or solar PV systems. Between 2001 and 2008 there was a modest uptake of small-scale systems, however when the solar credits multiplier was introduced in 2009, there was a rapid increase in the number of installations of solar PV in particular. The number of solar PV panels increased almost 6 fold from approximately 60 000 installations in 2009 to over 350 000 installations in 2011.

Source: Clean Energy Regulator, 2012.
Notes: Installation numbers in 2011 are likely to be higher as owners/agents have one year to register the instalment of small-scale systems. Installations of small-scale wind and hydro systems are very low and are not visible – hydro ranges from zero to five installations per year from 2001-2011 and wind ranges from one to 136 installations per year over the same period.

Box 6 State and territory feed-in tariffs

Until recently, most feed-in tariffs offered by state and territory governments have been considerably above the wholesale price of electricity. The Productivity Commission’s 2011 review Carbon Emissions Policies in Key Economies found that 'subsidy rates for solar PV often have been set at excessive levels, essentially providing windfall benefits to households that install solar PV.' (p.80). The NSW Independent Pricing and Regulatory Tribunal’s (IPART) 2012 report on solar feed-in tariffs stated 'the generous subsidies offered by governments contributed to a much higher than anticipated uptake of PV in NSW, and led to higher than anticipated costs' (p.8).

Most states and territories have now revised their feed-in tariffs for new applicants to a level reflecting expected wholesale electricity prices. The Victorian Competition and Efficiency Commission’s (VCEC) 2012 inquiry into distributed generation noted that 'advice to the Commission suggests the efficient and fair market price for 2013 to be, at a minimum, in the range of 6 to 8 cents per kilowatt hour (kWh) (compared with 25 cents currently for the Transitional Feed-in Tariff). This minimum range is consistent with rates announced in New South Wales, Queensland, South Australia and Western Australia over the past year.' (p.xxi)

Return to top

5.2. Small-scale Renewable Energy Scheme design

There are a number of issues with the current design of the SRES:

  • its uncapped nature means that the number of installations – and therefore total scheme costs – can be unpredictable and difficult to control;
  • the subsidy provided to small-scale systems does not automatically reduce as technology costs fall and wholesale electricity prices rise (unlike in the LRET, where certificate prices would be expected to fall in such circumstances);
  • there is no legislated end-date for the scheme; and
  • the 15 year deeming periods are unlikely to be justifiable for larger solar PV systems that are currently below the 'small-scale' threshold (100 kilowatts (kW)).

To address these issues, the Authority has considered a number of potential changes to the SRES, including:

  • recombining the SRES and the LRET into one scheme;
  • lowering the solar PV threshold;
  • capping the scheme by setting a gigawatt hour target, capping the small-scale technology percentage (STP), discounting the number of certificates that can be created, or lowering the existing price cap; and
  • setting an explicit end-date for the SRES (in line with the LRET).

Some of these options address several of the issues raised with the scheme's design, others are focused on a particular aspect. The costs and benefits of each are assessed in turn.

5.2.1. Recombining the two schemes

Recombining the SRES with the LRET would address several of the issues raised with the SRES design: it would cap the scheme so that increased installations are matched with a price signal, and establish an end-date. Many of the distorting factors that led to separation of the RET into two schemes are no longer in play. The reduction of the Solar Credits multiplier has been brought forward and will end on 31 December 2012 (see Table 3) and state and territory feed-in tariffs are now generally comparable to the wholesale electricity price. In light of this, there could be a case for recombining the schemes.

Table 3 Solar credits multiplier
Time frame Multiplier Notes
9 June 2009 – 30 June 2010 5x  
1 July 2010 – 30 June 2011 5x  
1 July 2011 – 30 June 2012 3x The Commonwealth Government reduced the multiplier from four to three times from 1 July 2011 to 30 June 2012 (Combet, Dreyfus 2011)
1 July 2012 – 31 December 2012 2x  
1 January 2013 onwards 1x (no multiplier) The Commonwealth Government announced on 16 November 2012 that the Solar Credits multiplier would be phased out on 1 January 2013, six months earlier than planned (Combet 2012)

Source: Renewable Energy (Electricity) Regulations 2001 (Cth) (REE Regulations 2001).

Cost and administrative requirements

Operating two separate schemes is likely to impose a greater cost on society than a single scheme. It effectively creates separate incentives – or ‘bands’ – for large-scale and small-scale renewable energy generation (see Chapter 8). This potentially increases the RET's overall costs as more expensive technologies may be deployed than if large and small-scale generation competed to meet the same target. Furthermore, it imposes greater administrative requirements, and therefore costs, for the renewable energy industry, liable entities, and the Clean Energy Regulator.

Most stakeholders that supported merging the schemes did so on the grounds that it was likely to lower costs. For example, Australian Paper submitted:

We would recommend … a wholesale review of the SRES and RET schemes as this aspect of the [RET] has created significant problems and expense for business. The uncapped nature of the [SRES], along with an inappropriate [feed-in tariffs] and deemed multiples resulted in unforeseen and uncontrolled cost imposts. (Australian Paper, sub.53, p.4)

Ergon Energy also noted the administrative burden of complying with two schemes:

The separate scheme has posed an additional administrative burden on liable entities. Ergon Energy has been required to establish and maintain separate models to administer, track and settle both large and small certificates in two separate markets. (Ergon Energy, sub.88, p.8)

The Authority's modelling estimates that the resource cost of maintaining separate schemes is higher than combining the schemes, costing almost $1 billion (in June 2012 dollars) more from 2012-13 to 2020-21 and around $2.4 billion more from 2012-2013 to 2030-31. This is because operating separate schemes is estimated to result in around 5 300 gigawatt hours (GWh) of additional renewable energy generation in 2020-21, resulting in around 16 million tonnes fewer greenhouse gas emissions compared to a combined scheme. However, combining the schemes is estimated to increase wholesale prices on average by around $1 per megawatt hour (MWh) over the period 2012-2013 to 2020-21, and $1.70 per MWh over the period 2012-2013 to 2030-31. It is projected that combining the schemes does not result in a reduction in retail electricity prices, as the increased certificate costs of separate schemes is offset by the decreased wholesale rates achieved through greater renewable energy generation.

Level and mix of renewable energy generation

The majority of submissions – including from environmental and business groups, liable entities, and the renewable energy industry – supported retaining two separate schemes. The main reasons put forward were regulatory certainty and concern that further changes might jeopardise the prospect of meeting the 41 000 gigawatt hour target. Some stakeholders also argued that merging the schemes would reduce investment in more efficient large-scale renewable energy projects.

Merging the schemes could affect the likelihood of meeting the Renewable Energy Target if it results in regulatory uncertainty and reduced investment in renewable energy projects. In the last three years, the RET has undergone several significant amendments – the expansion and inclusion of multipliers in 2009, and separation of the scheme in 2011. The Commonwealth Government has also twice brought forward the reduction of the Solar Credits multiplier. A constantly shifting regulatory framework (or the perception of one) may reduce investors’ willingness to invest in further renewable energy, and increased perceptions of risk may increase the cost of making such investments.

Many stakeholders considered that a further change to merge the schemes could undermine policy certainty and investment confidence. For example, AGL noted:

The separation of the RET scheme was vital to … creating conditions conducive to investment in large scale renewable generation. If this separation was removed, the market for large scale renewable certificates could again face distortion, jeopardising the 20% target and stymieing large scale renewable electricity generation in Australia (particularly if any new State-based policies emerged)… There have been no fundamental changes to the market dynamics which made necessary the division of the RET scheme in 2010. Accordingly there is no rationale upon which to remove this separation now. (AGL, sub.38, p.4)

Many stakeholders expressed concern that merging the schemes would affect the mix of renewable energy generation and potentially disadvantage large-scale projects, which require a greater degree of investment certainty due to the high capital investment and lack of deeming provisions. Hydro Tasmania submitted:

Any re-introduction of small-scale technologies into the LRET will almost certainly immediately stall investment in large-scale projects due to the recent experiences of certificate supply volatility and the increased market risk this would bring. (Hydro Tasmania, sub.40, p.8)

5.2.2. Lowering the solar photovoltaic threshold

Even if the schemes remain separate, there may be a case for moving some small-scale systems into the LRET by reducing the threshold of small generation units in the SRES. The Renewable Energy (Electricity) Act 2000 (Cth) (REE Act) sets the capacity limits for eligible small generation units under the SRES. While the capacity limits for small-scale wind turbine systems and small-scale hydro systems are relatively low (10 kW and 6.4 kW respectively), solar PV systems that have a capacity of 100 kW are still included in the SRES. This is considerably larger than the average size of solar PV systems installed in 2011 and 2012, which was approximately 2.6 kW (sourced from the Clean Energy Regulator, 30 September 2012).

To date, the vast majority of solar PV installations in Australia have been installed on residential dwellings (see Table 4). Over 99 per cent of small-scale PV systems installed are below 10kW. This is unusual compared to other countries where there are significant amounts of solar PV on commercial buildings (see Table 4). A number of stakeholders have highlighted the as yet untapped potential for commercial deployment of larger solar PV units on shopping centres, storage facilities, office blocks or farms. Potentially, these installations could generate a relatively high number of certificates compared to residential systems, increasing the overall cost of the scheme.

2011 installations by country Installed capacity MW Residential proportion % Residential capacity MW
Italy 9 301 8 744
Germany 7 500 9 675
China 2 200 27 600
US 1 867 37 698
France 1 634 16 261
Japan 1 296 90 1 166
Belgium 958 68 651
UK 899 56 503
Australia 865 95 822
Spain 345 5 17

Source: REC Agents Association, sub.47, p.11.

One option to guard against this potential cost increase, while still providing an incentive for commercial deployment, would be to lower the capacity threshold of solar PV so that larger installations are captured in the LRET – a capped scheme. Many of the disadvantages identified above with merging the schemes would not apply to lowering the solar PV threshold. In particular, business models for operating in this market are only now being developed – there is no existing, established industry that would be disrupted by the change.

A number of LRET participants raised concerns that including systems of greater than 10 kW in the LRET could crowd out investments in large-scale projects. However, the potential for disruption by two key 'artificial' sources – multipliers and generous state and territory feed-in tariffs – is now low.

The Authority considers that it is important to retain some level of deeming for all systems under 100 kW, regardless of whether they are included in the small-scale or large-scale scheme. Deeming provides an efficient method for allocating a meaningful number of certificates to smaller sized systems without the administrative burden of metering each individual system's output. Furthermore – unlike large-scale PV systems – there is generally no third party data to verify a systems owner's claim of generation.

Some LRET participants argued that deeming arrangements for these systems would give them an unfair advantage. Analysis by the Authority suggests that while deeming does bestow some benefit – by removing the risk associated with future regulatory change – there is no inherent financial advantage to receiving certificates up-front rather than having them issued periodically in line with generation. In a scheme with banking, the market can be expected to take account of the value of certificates in the future.

That said, there is a clear case for reducing the deeming period for larger solar PV units. The larger the system, the less justification there can be for long deeming periods, since the scope for inaccuracies is greater and the additional compliance costs as a proportion of total certificate revenue created by the system is lower. Participants who install larger systems are likely to have more capacity to respond to the greater administrative requirements of more regular deeming.

5.2.3. Capping the Small-scale Renewable Energy Scheme

As noted above, one issue with the SRES design is that, unlike the LRET, it is 'uncapped' with liability tracking certificate creation. Annual liability (set through the STP) is based on the number of certificates expected to be created that year, adding or subtracting any surplus or deficit of certificates from the previous year. This means that as installations increase, so does the STP and, consequently, the cost of the scheme to liable entities and, through them, electricity consumers.

The uncapped nature of the SRES has become particularly relevant because the number of installed small-scale systems, and therefore scheme costs, has been so much higher than expected. When the SRES was legislated in 2010, it set an ‘implicit target’ of 4 000 GWh of generation in 2020. It has already exceeded this target with current estimates of approximately 5 000 GWh per annum and the Authority estimates it will reach around 11 000 GWh in 2020-21.

Many review participants – particularly large energy users and liable entities – expressed concern regarding the cost of the SRES. IPART stated:

The design of the SRES, combined with generous State and Territory Government financial incentives, has put the annual costs of complying with the SRES at almost twice that of the LRET. The costs of complying with the SRES were a driver of retail electricity price increases, particularly on 1 July 2011. (IPART, sub.81, p.14)

The modelling undertaken for the Authority estimates that the total SRES certificate cost was around $1.3 billion in 2011-12 and may fall to around $300 million in 2020-21.

The high uptake of small-scale systems might also suggest that the level of subsidy provided by the SRES is too high and that small-scale systems are becoming affordable in their own right. Origin Energy noted:

…the twin effects of falling solar panel costs and rising retail tariffs will create a situation where the subsidy required to support distributed solar PV will continue to reduce over time. There may be a point in the latter part of this decade when such subsidies are no longer required. (Origin Energy, sub.69, p.11)

On the other hand, a number of review participants considered that, while the level of support may have been too high in the past, the cost of the SRES was likely to stabilise in the near future because the factors that drove the rapid growth in uptake are no longer at play. For example, in information provided to the Authority by Warwick Johnston from SunWiz, he commented that system costs were likely to stabilise:

While PV prices are hard to predict, clearly the massive price reductions in recent years cannot continue... the global market for PV grew extraordinarily in recent years, and recognising that scale was required to reduce manufacturing costs, a massive supply of PV manufacturing capacity was built... As a result we have an international oversupply of PV, which has created a price war that now sees panels being produced below-cost. The largest PV manufacturers are all struggling to turn a profit, a situation that cannot be sustainable… Hence, such rapid reductions in PV prices cannot be expected to continue. Instead, I expect that PV prices will stabilise within a year as wise companies turn their focus towards profit-based survival instead of attempting to capture market share through a price-war. Reductions in manufacturing costs will continue, but for the medium term prices should remain steady as major manufacturers pay down their debts. (SunWiz, email correspondence, November 2012)

The Australian PV Association supported this assessment.

Others, such as the Australian Aluminium Council pointed to the history of underestimating SRES generation, noting:

The cost burden on electricity users of the SRES component of the scheme has been many times greater than the modelling that was used by the Government to undertake the separation and the nominal 'assigned' target of 4 000 GWh for the SRES. Any statement or modelling about future SRES permit generation levels will therefore be treated with a healthy amount of scepticism. (Australian Aluminium Council, sub. 73, p.5)

A number of participants suggested capping the SRES to provide certainty about the number of future installations – and therefore cost – of the scheme. This section considers the costs and benefits of four potential ways of controlling the compliance costs under the SRES – a gigawatt hour target, an STP cap, a discounting mechanism and lowering the existing price cap.

A gigawatt hour target

A gigawatt hour target for the SRES would cap the quantity of certificates that were required to be surrendered each year – in the same way the annual LRET targets currently work. Accordingly, if there were an oversupply of certificates, the price would be expected to fall. This option has the benefit of being simple and familiar to many RET participants.

A number of review participants expressed concern that a gigawatt hour target for the SRES would create a 'boom-bust' cycle because small-scale systems are relatively inexpensive and households are able to respond promptly to changes in incentives. For example, the Clean Energy Council stated:

If the scheme were to be capped you would see installations of small scale systems pulled forward (to avoid being outside the cap) which would create a cycle of boom and then bust, as once the cap was reached demand would plummet until the cap reset the following year. (Clean Energy Council, sub.12, p.17)

The Ai Group expressed similar concerns in its submission, noting:

…in the context of the market for small-scale systems, a cap is likely to cause considerable problems and dislocation. The experience with other capped benefits, such as the former rebates for solar PV or state government grants and tariffs, is that demand spikes when the public believes that time is running out; governments often find it hard to enforce a cap; and neither government nor industry may have a clear picture of total activity or the pipeline for certificates. The risk is that the cap does not hold, and that the cap drives annual boom-and-bust cycles that damage the industry. (Ai Group, sub.46, p.16)

Investing in small-scale systems is quite different from investing in large-scale renewable energy projects. Large-scale projects are generally planned and announced years in advance – investors are generally well-informed about the progress of projects that will contribute to the LRET target. By contrast, quantities of installed small-scale systems can change very rapidly, and accurate information is only available after the fact. For this reason, boom-bust cycles are more likely in a capped SRES scheme than a capped LRET.

An SRES boom-bust cycle would create difficulties for both households and businesses. In particular, in terms of equity, it would be unfair if non-expert participants, such as households, invested on the basis of a certain set of circumstances and then discover they have missed out on the expected subsidy because the cap has already been reached and the price of certificates has plummeted. This was demonstrated with the off-grid multiplier that gave an additional incentive to off-grid systems. The incentive had an annual cap and in its first year of operation it led to a rush to install systems, which resulted in the scheme being oversubscribed. This resulted in many applicants missing out, leaving them significantly out of pocket.

In summary, while introducing a gigawatt hour target would likely limit the overall costs of the SRES, it would require a major structural overhaul, creating significant regulatory uncertainty. This would adversely affect the small-scale industry, as well as households and, ultimately, may not be sustainable.

A cap on the small-scale technology percentage

Annual SRES liability is determined by the STP, which is set each year based on the expected number of certificates that will be created that year, plus or minus any carry-over from the year before.

An STP cap could be used to set a maximum level for the STP and thereby limit the amount liable entities (and electricity consumers) had to pay on an annual basis. It would also – over time – limit the incentive to install more systems as the price of certificates would be likely to fall if the STP cap was reached.

The main disadvantage of an STP cap is that its effectiveness depends on setting it appropriately. If it were set too high, it would not bind and therefore would not limit liability. If set too low, it could cause the price of certificates to fall dramatically, potentially disadvantaging those that had already invested on the basis of a higher expected certificate price. This might threaten the viability of some small businesses and lead to arguments to increase the cap. Setting an appropriate cap would depend, to some extent, on being able to predict future STPs. This has been notoriously difficult in the past: the Clean Energy Regulator's non-binding estimates have tended to significantly underestimate the STP. For example, early in the year the non-binding estimates for the 2012 STP was 16.75 per cent (31 March 2011) and by the end of the year it had almost doubled to 23.96 per cent (16 December 2011).

If the STP cap was not set appropriately it could either be ineffective, or cause significant disruption to the SRES market.

Discounting certificates

A discounting mechanism would apply a multiplier of less than one to each certificate, effectively discounting the number of certificates created for each megawatt hour. For example, a multiplier of 0.5 would mean that each certificate represents two MWh of renewable energy generation. A discounting mechanism could control uptake of small-scale systems under the SRES, as demonstrated in reverse by the Solar Credits multiplier.

A discounting mechanism would be unlikely to create the potential boom-bust cycles of a gigawatt hour target or an STP cap because fractional reductions below one would translate into small changes in the level of support. A discounting mechanism also has the advantage of not affecting existing investors.

The main disadvantages of a discounting mechanism relate to its implementation. Applying it on a discretionary basis would allow it to respond to changing circumstances, but would be unpredictable and potentially disruptive for industry. On the other hand, a 'set-and-forget' approach has the advantage of providing certainty but is essentially arbitrary and difficult to justify on the basis of principled analysis. The Clean Energy Council made this point, noting:

Implementation of the proposal to utilise average payback period as a criterion for lowering the SRES multiplier would be highly complex and problematic. If the approach were simplified, it would inevitably be perceived as unfair. (Clean Energy Council, sub.191, p.6)

The Authority's preliminary view, as set out in its discussion paper, proposed the use of a discount mechanism to be applied at the Minister's discretion on the basis of:

  • the payback period falling below ten years;
  • changes in net system costs; and
  • electricity prices and whether the SRES constituted more than 1.5 per cent of an average bill.

Many review participants expressed concern with this proposal. Some – such as the Ai Group – strongly supported the concept of discounting but were concerned that the proposed method of application could be too uncertain, depending on how the proposal was further developed. IPART expressed concern with the discretionary application, stating:

In our view, … providing the Minister with discretion to discount the number of certificates created by small-scale investors, [is] likely to significantly reduce investment certainty to small-scale investors and create further uncertainty for retailers and regulators in determining the costs of complying with the scheme. (IPART, sub.206, p.2)

A number of stakeholders, including the REC Agents Association, opposed the application of a discount mechanism at all, considering it too extreme. These participants also argued that a discounting mechanism would put small-scale systems at a disadvantage vis-à-vis large-scale projects in the LRET.

Lowering the existing price cap

The SRES has a price cap of $40 set through the fixed clearing house price. The REE Act allows the Minister to lower the price cap, taking into account:

  • whether the total number of small-scale technology certificates created in 2015 exceeded or is expected to exceed the equivalent of 6 000 GWh;
  • any changes to the costs of small generation units and solar water heaters;
  • the extent to which owners of small generation units and solar water heaters contribute to the costs of small generation units and solar water heaters;
  • the impact of the clearing house price, and the number of small generation units and solar water heaters installed on the electricity market, including on electricity prices; and
  • any other matters the Minister considers relevant.

To date, the Minister has not exercised this power, preferring instead to accelerate the reduction of the Solar Credits multiplier (Combet, Dreyfus 2011).

The price cap does not directly limit the number of installations; however, it does reduce the maximum price paid for each certificate and in this way may lower the overall cost faced by liable entities. If the certificate price falls, it might also affect uptake, as the return would be smaller.

There are a number of complexities regarding lowering the price cap. First, it affects investments that have already been made on the assumption that the price cap is $40. This was raised by the Clean Energy Council in its submission:

…if the $40 price were to be adjusted, the impact on the small scale technology market would be highly detrimental. Firstly, the value of STCs in the spot market would likely fall dramatically, as the expectation that the Clearing House will eventually come into play in a significant way over the next 12 to 24 months would be removed and this would lower estimates of the longer term value of STCs. Many investors from major banks to solar PV businesses and dedicated certificate trading businesses are holding substantial quantities of STCs. Material changes to the SRES or the Clearing House could devalue those assets and undermine the viability of those businesses. As these certificates trading businesses help to provide cash flow to PV businesses anything that harms these businesses or discourages new entrants in the STC market will harm the PV sector more broadly. At the very least it would reduce the value of their asset which is unfair to them. (Clean Energy Council, sub.12, p.23)

On the other hand, the power to change the clearing house price cap is included clearly in the legislation and prudent investors in small-scale certificates could reasonably be expected to understand that the asset they hold is subject to this discretionary power.

Lowering the clearing house price will create transitional issues for the certificates already on the clearing house transfer list. If the decision were taken to lower the clearing house price, a decision would need to be taken regarding how to treat these certificates.

5.2.4. Phasing out the Small-scale Renewable Energy Scheme

Unlike the LRET, which ends in 2030, there is no legislated end-date for the SRES. Setting an end-date ahead of time – and establishing a clear, graduated path to it – would provide industry and investors with certainty regarding the future of support. It is also in keeping with the overall transitional nature of the RET (see Chapter 3) – as a temporary measure to provide industry support and encourage additional renewable energy generation ahead of an established, credible carbon price consistent with delivering on Australia’s long-term environmental goals.

An argument can be made that the policy intent was for the SRES to end in 2020; at the time the SRES was split from the LRET, all discussion of the SRES contribution to the target was in terms of 2020 gigawatt hour generation (Wong 2010, Commonwealth Government 2010). Another possible option would be 2030, which would align the SRES with the LRET.

There are a number of possible ways to set an end-date to the SRES, including: reducing deeming periods so there is no subsidy past 2030, and prescribing an end-date for the scheme in the REE Act.

Reduced deeming

Under a reduced deeming approach, small-scale systems would only be provided with certificates for generation up to 2030. A solar PV unit currently receives 15 years' worth of certificates upfront. Reduced deeming would mean that a solar PV unit installed in 2019 would only receive 12 years' worth of certificates – rewarding generation up to and including 2030, but not beyond.

Reduced deeming would provide a predictable path to the end of the scheme. It would slowly phase out support over time, providing industry with both certainty regarding the future of the SRES and a graduated step-down that could be managed. Figure 27 illustrates how reduced deeming would affect certificate creation for an average solar PV unit and an average solar water heater.

Figure 27 Example of phase-out of deeming years on small-scale technology certificates
The figure provides an example of how reduced deeming would affect certificate creation for an average solar PV unit and an average solar water heater. If a solar PV system was installed prior to 2016, certificates could be created for the full deeming period of 15 years, however after 2016, the number of certificates would be reduce to reflect a shorter deeming period of one year. Each year after 2016 would see the number of certificates reduced further as the deeming period is further shortened. For solar water heaters, where the deeming period is ten years, the number of certificates that could be created would be reduces each year after 2021.
Source: Climate Change Authority, 2012.

A potential disadvantage of this option is that the administrative costs of providing certificates for very short periods (for example, less than five years) in the latter years of the scheme could outweigh the benefits. This could resolve itself naturally, as participants may not consider it worthwhile to apply for a small number of certificates.

No future small-scale technology percentage

The LRET ends in 2030 because the REE Act does not provide for any gigawatt hour targets beyond this date. The renewable power percentage (RPP) will therefore fall to zero and no further liabilities will be created. Although the SRES does not have a gigawatt hour target, a similar approach could be taken by stipulating that there will be no further small-scale technology percentage (STP) after a certain date.

The advantages of this option are that it is simple and would work effectively for either a 2020 or 2030 end date. The disadvantages are that it may provide a sudden drop in support as the STP may not fall smoothly up until the end date.

This option could be combined with reduced deeming to provide for a graduated phase-out over time to 2030, after which no further STPs would be set.

5.2.5. Conclusion

All of the proposed options have advantages and disadvantages. On balance, the Authority has assessed that the following combination is likely to provide the highest net benefit, while minimising disruption to the schemes and their participants:

  • retaining separate schemes to maintain regulatory stability;
  • lowering the SRES threshold for solar PV units to guard against a future boom in larger-scale installations (this essentially recombines a component of the SRES with the LRET);
  • reducing deeming periods to provide a graduated and predictable phase-out of the scheme by 2030, after which no further STPs would be set; and
  • retaining the ministerial power to lower the price cap in the event that there is another boom in installations of small-scale systems and a lower level of subsidy would be appropriate.

The Authority considers that while recombining the schemes would have addressed most of the issues associated with the SRES, it would also require significant regulatory change, which would affect both the small- and large-scale schemes. In light of this, the Authority considers there are less disruptive ways of addressing the issues associated with the SRES.

Similarly, while a gigawatt hour target, an STP cap or a discounting mechanism could all effectively contain the cost of the SRES, they also require significant changes and would likely result in considerable uncertainty for scheme participants. A gigawatt hour target or an STP cap could also create certificate price volatility and 'boom-bust' cycles.

Such measures would be justified if uptake of small-scale systems under SRES was likely to continue to grow rapidly. However, the factors that drove this boom (falling system costs, generous feed-in tariffs, and multipliers) are no longer at play and installations of small-scale systems are expected to stabilise and the cost of the SRES to fall. The Authority's modelling estimates that with the current policy setting in place, the contribution of SRES certificate costs to the average household bill will fall from 2.4 per cent in 2012-13 to 0.6 per cent in 2020-21 (see Table 5).

Table 5 Contribution of Small-scale Renewable Energy Scheme certificate costs to average household bill
Year 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 2019-20 2020-21
SRES 2.4% 1.4% 1.1% 1.0% 0.8% 0.7% 0.6% 0.6% 0.6%

Source: SKM MMA, 2012

In these circumstances, the Authority considers that the benefits of introducing a gigawatt hour target, STP cap, or a discounting mechanism are unlikely to exceed the costs of the significant disruption these options are likely to cause. The Authority’s recommendations therefore focus on mechanisms to guard against a possible future boom in installations (and consequent scheme cost), rather than on mechanisms that actively limit the number of installations.

Solar PV on commercial buildings currently constitutes a very small proportion of total installations; future growth is possible but hard to predict. Should they remain in the SRES, a boom in such installations would lead to rapid increase in compliance costs given that larger systems create more certificates. The Authority therefore recommends lowering the SRES threshold of solar PV units from 100 kW, PV systems above this threshold would be eligible under the LRET. This would still provide support to commercially-installed solar PV – but in the context of a capped scheme.

Consultation with stakeholders suggests a threshold of somewhere between 10 kW to 30 kW would be appropriate – the Authority recommends the Commonwealth Government conduct further consultations with stakeholders to determine a precise threshold. In the Authority's view, a 10 kW threshold would be an appropriate starting point for these consultations.

In addition, the Authority considers that larger systems should be subject to five year deeming similar to the current five year deeming option available to small generation units in the SRES. This will encourage better accuracy around deeming arrangements – as is appropriate for larger systems – and also would not confer an unfair advantage on PV compared with other LRET participants.

The Authority recommends that the ministerial power to lower the price cap be retained. While not ideal, it could act as an ‘emergency' cost containment measure if unexpectedly high levels of installations of small-scale systems continued, driven, for example, by further falls in technology costs or the continued rise of the Australian dollar. Lowering the price cap has the advantage of being known to scheme participants as it is part of the existing scheme design. Some of the disadvantages associated with lowering the price cap – such as transitional arrangements for certificates on the transfer list – may be reduced should the Commonwealth Government adopt the Authority’s recommendations regarding the clearing house – discussed in Section 5.3.1.

Finally, the Authority notes that state and territory feed-in tariffs have a significant effect on uptake of small-scale systems under the SRES. Consultation with state and territory governments indicates that feed-in tariffs are unlikely to be increased in the future. Nevertheless, greater coordination of policies would be useful. The Council of Australian Governments’ Standing Council on Energy and Resources is considering the merits and options for developing guidelines for a consistent national approach to fair and reasonable feed-in tariffs for small-scale renewable generation. Any such guidelines would aim to encourage competition, provide clear rights and obligations around the terms of connection and what constitutes a fair and reasonable return for a small-scale system (Standing Council on Energy and Resources 2012). The Authority considers the Council’s work based on these principles will help promote the stabilisation of the SRES.

Recommendation

R.5. The Small-scale Renewable Energy Scheme should remain separate to the Large-scale Renewable Energy Target.

R.6. The threshold for solar photovoltaic units in the Small-scale Renewable Energy Scheme should be reduced from 100 kW to, say, 10 kW. The Authority recommends the Government conduct further consultation with stakeholders to determine an appropriate threshold. Units over the small-scale threshold would be included in the Large-scale Renewable Energy Target with five year deeming.

R.7. The ministerial power to lower the price cap should be retained to provide an immediate cost containment mechanism should installations of small-scale systems boom.

R.8. The Small-scale Renewable Energy Scheme should be phased out by reducing deeming so that renewable energy generation from small-scale systems is not rewarded after 2030.

5.3. Small-scale Renewable Energy Scheme administration

There are a number of administrative issues associated with the SRES that could also be amended to improve the operation of the scheme. This section considers:

  • the clearing house;
  • generation returns; and
  • the collection of data regarding out of pocket expenses.

5.3.1. The clearing house

The ‘STC clearing house’ is a voluntary mechanism designed to facilitate the exchange of STCs between buyers and sellers (owners or agents) at a fixed price of $40. Sellers may enter their STCs on the ‘transfer list’. The list clears as buyers purchase STCs from the clearing house. If there are no STCs listed, the Clean Energy Regulator will create an STC, which will be replaced with the next certificate entered on the transfer list. While the clearing house provides a set price per certificate, there is no guarantee how long it will take to sell.

The clearing house was designed to fulfil two purposes: to cap the price of certificates for liable entities (Commonwealth Government 2010) and to deliver a set subsidy of $40 per STC for households, small businesses and community groups (Department of Climate Change and Energy Efficiency 2010).

The clearing house operates as a price cap by allowing liable entities to acquire a limitless number of certificates from the clearing house for the set price of $40. The price cap role played by the clearing house is an important cost containment mechanism given the SRES has no quantitative cap.

While the clearing house has provided an effective price cap, it has failed to deliver a set subsidy of $40 per STC to owners of small-scale systems (or their agents). The clearing house is a voluntary mechanism and liable entities have chosen to acquire certificates outside the clearing house where STC spot prices are around $25 to $32 (see Figure 28). This has meant the clearing house transfer list has not cleared for 18 months (the last sale was on 25 February 2011).

Figure 28 Small-scale technology certificate spot price
This figure shows the monthly spot price of small-scale technology certificates over the period January 2011 to August 2012. It shows that over the period the spot price has never reached the clearing house price of $40, but has ranged between $19.75 and $39.60. The unweighted average spot price of small-scale certificates over the period has been $29.55.
Source: The Green Room – Next Generation Energy Solutions (NGES), 2012.
Note: In nominal prices.

Although the Commonwealth Government has never guaranteed a timeframe for clearance (and this is made clear on the Clean Energy Regulator’s website), the existence of the clearing house has created an expectation among some non-expert participants that a $40 set price per STC is obtainable. The Authority received several submissions from individuals who had installed small-scale systems anticipating $40 per STC, but whose certificates had not yet cleared – in some cases, the individuals had waited a considerable period of time.

The Authority has considered a number of possible amendments to the clearing house that could address this issue and improve the operation of the SRES, including:

  • amending the clearing house so that it is compulsory and therefore delivers the set $40;
  • abolish the clearing house and use the shortfall charge as a price cap; and
  • amend the clearing house to be a ‘deficit sales facility’.

Amend the clearing house to be compulsory

A number of stakeholders proposed making the clearing house compulsory to ‘[stop] the market under cutting [the clearing house price] and restore the price to a set $40 per STC’ (Robin Morgan, sub.1, p.1).

When designing the scheme, a compulsory clearing house was considered (Commonwealth Government 2010). It was decided it should be voluntary on the grounds that it provided greater flexibility to liable entities – this remains a valid consideration. Importantly, making the clearing house compulsory would most likely increase the cost of the SRES and, given that cost is one of the primary concerns with the scheme, the Authority does not recommend this course of action.

If the objective is to establish a set subsidy for small-scale systems, other policy mechanisms – such as feed-in tariffs or rebates – would be more suitable than a certificate trading scheme. However, it is not clear that a set subsidy is needed to drive the installation of small-scale systems. Installation of small scale systems has successfully continued, despite fluctuating certificate prices of $20 to $30 since the inception of the SRES.

Abolish the clearing house

Another option is to abolish the clearing house. This would effectively remove the ‘promise’ of a $40 set price, pushing all activity onto the secondary market. This would have the advantage of removing the unrealistic expectation regarding the attainment of $40 per STC in a timely manner.

Abolishing the clearing house may also have some benefits in terms of administrative savings. However, these are likely to be small. The largest cost associated with the clearing house was its establishment; its ongoing operating costs are relatively low.

At face value, abolishing the clearing house may not impact adversely on the market: there is a functioning secondary market and many householders now interact with the scheme through agents, (particularly given the clearing house does not deliver prompt payment for certificates). However, the clearing house may play a more active role should the SRES stabilise. The Clean Energy Council made this point in its submission, stating:

While the Clearing House has not necessarily played the role it was intended to play as part of the SRES, it is nevertheless now an important part of the scheme. The challenges in forecasting the uptake of small scale systems has limited the extent to which the clearing house has played an active role in the market to date. [The] Clearing House should remain in place and will over time play an increasing active part in the functioning of the SRES. If the Clearing House were to be changed or abolished … the impact on the small scale technology market would be highly detrimental. (Clean Energy Council, sub.12, p.23)

Return to top