Chapter 3

Options for the retirement of coal fired power stations

3.1        Evidence to the inquiry highlighted that Australia's coal fired power stations will need to be retired in the medium term in order to make way for lower-emissions sources of power generation. Various options could be utilised to facilitate this process, and are discussed through this chapter.

3.2        Broadly, the options for facilitating the retirement of coal fired power stations include the following:

'Barriers to exit' and need for policy certainty

3.3        Much of the policy discussion in this area focusses on whether there are 'barriers to exit' which impact on the decision-making of coal plant operators when determining if (and when) to close.

3.4        The question is not merely whether any barriers to exit exist, but whether these barriers are significant enough to prevent an 'efficient' or 'orderly' restructuring of the market to occur (with older, high-emissions plant capacity retiring first). As explained by the Australian Energy Market Commission:

A barrier to exit is any cost or foregone profit that a firm must bear if it leaves an industry. While these costs therefore represent barriers to exit for individual generators they are only a problem if they are a barrier to efficient exit decisions.

For example, based on this definition, it will not always be efficient for generators with the highest variable cost to exit the market first. Where generators with high variable costs have high shut down costs, it can be an optimal outcome for them to exit the market after generators with low variable costs but low shut down costs.[1]

3.5        Several barriers to exit for coal fired power stations have been identified in the Australian context, which can be summarised broadly as follows:

3.6        This final factor, policy uncertainty, was identified by numerous stakeholders to the inquiry as a key issue creating instability in industry decisions­—along with the corollary observation that introducing more policy stability in this area would promote better outcomes for investors and market participants. For example, Associate Professor Frank Jotzo argued:

Australia's energy sector has been exposed to significant investment uncertainty due to pervasive policy uncertainty and climate policy reversals for over a decade. Such uncertainty has detrimental effects on the investment climate and potentially on the cost effectiveness of investment... For an effective and efficient low-carbon transition, stable and predictable policy settings are needed.[6]

3.7        The Australian Energy Council argued similarly:

A benefit of the market is that it can discover what the real economic life of a power station is and when it is worthwhile to invest in refurbishing a plant to extend its operating life. Stable carbon policy is needed to inform this investment decision making, and potentially signal that coal-fired power station emissions intensity may lead them to close earlier than without a carbon policy.[7]

3.8        AGL Energy submitted:

The transition to a decarbonised and modernised generation sector requires large scale investment, recent AGL analysis estimates this at $23 billion in renewables alone to achieve an emission reduction consistent with a 27% reduction in [greenhouse gas] emissions by 2030.

Such investment will be supported by policy that provides macro level certainty as to the timeframe and operating life of incumbent plant.

Such certainty has the potential to benefit a range of factors contributing to the efficient transition including new investments, management of existing capital stock, policy development, community transition and energy market development.[8]

Leaving retirement decisions solely to industry and market forces

3.9        The status quo approach would leave any retirement decisions on the closure of coal fired power stations up to the plant owners themselves, with no external changes in government policy settings to assist this process. This approach was endorsed by the COAG Energy Council in December 2014, which stated:

The Council considers it is for the market to provide signals for investment and de-investment for generation, and opposes the transferral of the costs of retiring assets onto consumers or taxpayers.[9]

3.10      Advocates for this position argue that plant operators will choose to cease operations as necessary, in accordance with existing market conditions, and that there are no barriers to exit that are significant enough to warrant government intervention. The Australian Energy Market Commission (AEMC) undertook work in 2015 to identify barriers to generators exiting the NEM, and found that 'there is nothing in the National Electricity Law or Rules which would constitute a barrier to efficient exit decisions by generators'.[10]

3.11      The AEMC stated that recent experience shows that generators are not being prevented from leaving the market under current policy settings:

While it is possible the uncertainty around exit costs is creating a barrier to efficient exit, a number of generators have announced exit decisions in recent years. The evidence suggests that any barriers to exit have not deterred generators from commencing various stages of exit or the full retirement of plant. This would support leaving it to the market to determine which plant should exit.[11]

3.12      In particular, the AEMC pointed to the closure in May 2016 of the Northern and Playford B coal power stations in South Australia and the announced closure of the Hazelwood plant as examples of generator exit without further policy intervention.[12]

3.13      The AEMC stated further in its submission to the inquiry:

The decision of a generator to retire should be a commercial decision.

Investment and divestment decisions are based on a range of factors. A decision to retire a generator can take a number of years and requires intimate knowledge of the commercial and operating structures of that generator as well as clear expectations about future revenues and costs. Generators are best placed to manage the risk of their own investment or divestment decisions. The added benefit of this approach is that the risks of poor investment decisions are borne by generators rather than taxpayers or electricity consumers (as would be the case if a government were to intervene).[13]

3.14      Other stakeholders have maintained that existing barriers to exit do risk distorting the process of market transition, arguing that additional policy intervention may be required in order to facilitate the phased closures of older, higher-emissions generators. The imperative to reduce Australia's carbon emissions in line with our international commitments is also cited as a reason for implementing policies that would have the effect of curbing emissions in the electricity sector, even if a consequent result of such policies is to force coal powered generators to close sooner than they otherwise would have.[14]

3.15      In its submission, AGL Energy stated:

There is a role for governments to establish policy that facilitates 'orderly' rather than 'disorderly' exit of emissions intensive aged power stations. Such policy could be based upon age (e.g. Canadian rule which requires power stations to be closed or retrofitted with carbon capture and storage when they turn 50), emissions intensity or a market mechanism (as proposed by Jotzo and Mazouz). Ultimately, policy makers should view such a closure policy as not only an important means of securing energy supplies from modern generation equipment; but also an effective way of systemically reducing greenhouse gas emissions and providing communities the certainty they deserve to plan for such a transition.[15]

3.16      Mr Andrew Stock of the Climate Council told the committee that without a coordinated closures policy, it is difficult for generators to properly plan and announce plant retirement decisions:

Planning for closure is actually quite problematic at an individual operator level for some quite difficult commercial reasons—that is, the electricity market operates much like another financial market would in that people selling electricity not only trade in the physical product on a day-to-day basis where they dispatch but they also trade financially in the futures market to support their physical retail contract positions. So when a decision for closure is made, it is very hard to telegraph that because if you are doing that you are trading with inside information potentially. This is one of the reasons why closure announcements come in the current market in the way they do. If the owners of power stations make a final decision before they announce that decision to the market, they are potentially trading with inside information, and that has quite serious consequences.[16]

3.17      Various policy mechanisms have been discussed as potentially aiding the transition away from coal fired power generation and towards lower emissions generation. These approaches are discussed further below. Several of these proposed mechanisms have been investigated by the Climate Change Authority (CCA) as part of its Special Review of Australia's climate action, initiated in 2014 and completed in August 2016.[17] As part of this special review, the CCA commissioned two sets of modelling on the effects of different carbon pricing policy options on the electricity sector.

Policy mechanisms based on direct regulation

3.18      Policy options based on direct regulatory responses by government (as opposed to market-based mechanisms implemented by government) considered by stakeholders to the inquiry included payment-for-closure schemes and several other models for regulating the closure or ongoing operations of coal power stations.

Payment-for-closure schemes

3.19      Under this model, governments agree to pay certain power station owners to close, encouraging an orderly exit of older and high-emission coal power stations from the market. The Australian Government previously announced a 'contracts for closure' scheme in 2011, as part of its clean energy package that also included the introduction of a carbon price.[18] Dr Jenny Riesz summarised the outcome of the proposed scheme as follows:

This scheme aimed to permanently close around 2000 MW of highly emissions intensive generation capacity by 2020 via payments to particular plant owners from the Federal Government. The amount paid was to be determined by negotiation...

Closure proposals were received by the Government from all eligible generators in early 2012. Negotiations ceased on 5 September 2012 with the announcement that no agreement had been reached. Again, there were differing views on the reason for this outcome. However, the expectation of a low carbon price, high gas price and high black coal price appear to have pushed up the asking price of brown coal generators beyond that which the Government was prepared to pay.[19]

3.20      A variant of this kind of scheme to retire brown coal power stations is due to be implemented in Germany: starting from October 2016, a capacity of 2.7GW of power from three brown coal plant operators will be taken out of production, with payments of 230 million euros per year made to the operators over a seven year period. The cost of these payments is borne by electricity consumers (increasing costs to consumers by 0.05 euro cents per kilowatt hour).[20]

3.21      Direct payment-for-closure schemes have been criticised for a number of reasons in the Australian context. Professor Frank Jotzo and Mr Salim Mazouz argued in their 2015 paper on the retirement of coal fired power stations:

...payments-for-closure schemes can lead to unhealthy expectations of future industry subsidies from government and therefore a deferral of plant closure decisions with associated emissions.

Secondly, the politics of paying significant sums of taxpayers' money to the owners of old, highly emissions intensive power stations would be highly problematic. It also does not fit the narrative of the present Emissions Reduction Fund (ERF) mechanisms, which is one of subsidising businesses taking positive actions to move to cleaner production processes, not of compensation payments to sunset industries.[21]

3.22      The COAG Energy Council expressed the view in December 2014 that it does not support assistance to generators to exit the market.[22]

3.23      Alinta Energy, which closed its Flinders coal mine and power station in South Australia in May 2016, submitted that no government payments or incentives to close are required. It argued that the market 'understand[s] and price[s] the cost of closure into the long term planning', and ultimately the public purse should not pay for private closure.[23]

Direct regulation of power station closures and operations

3.24      Another set of options available to government would be to introduce regulatory measures that directly police the emissions performance of power stations, or mandate the retirement of coal fired power stations based on specified criteria. Direct regulatory responses could include:

3.25      The Australian Energy Council commented on regulatory closure options in its submission:

Regulatory closure, or even the requirement to give an extended closure notice, may prejudice both financing arrangements and supply contracts of power plants. This may then precipitate a disorderly closure if loans are called in early or suppliers terminate contracts. However, all of this depends on the type of regulatory closure.[25]

Emissions standards for power generators

3.26      Mandating emissions performance standards for any new power generators would prevent any new high-emitting coal fired stations from being built.

3.27      Canada has implemented an emissions standard for new and existing coal fired generators, meaning that no new coal fired power stations can be built without carbon capture and storage (CCS) technology.[26]

3.28      Similar to Canada, the United States has adopted emissions standards for new coal generators, which effectively require CCS to be implemented in any new projects.[27]

3.29      Emissions standards of this type have been considered by the Australian and state governments in the past, and have been implemented only to be subsequently withdrawn in some Australian jurisdictions.[28]

Absolute emissions baselines for generators

3.30      This model would set a baseline constraint on emissions output of each incumbent generating facility, without any market-based certificate trading between generators.[29] The emissions baselines for each plant can be decreased over time to steadily increase the level of emissions reductions required and force generators to adopt low emissions technology (e.g. implementing CCS retrofit for coal plants) or exit the market.

3.31      The potential impact of a version of this policy in Australia was modelled by Jacobs Group (Jacobs) in 2016 for the Climate Change Authority, which found that its introduction would cause significant electricity price increases in the 2020s, more so than other policy options.[30]

Regulated closures of coal fired power stations over time

3.32      This policy option involves the regulated closures of coal stations over time, either on the basis of age or on the basis of emissions intensity. As explained by Jacobs:

[These schemes] would close existing coal capacity in roughly linear fashion starting with the oldest or most emissions intensive, with the order of plant closure publicly announced at the time the policy is introduced. Each plant identified for closure would be legally required to either close or CCS retrofit by its closure date.[31]

3.33      Modelling conducted for the CCA by Jacobs in 2016 investigated the option of government mandating the regulated closures of all remaining coal fired power stations operating in Australia by 2030 on the basis of age. Under this scenario, coal generators that do not undergo a retrofit to incorporate CCS technology would be closed on the basis of age, and no new coal capacity could be built without CCS technology.[32]

3.34      This scenario modelling found that pursuing this policy would lead to less overall emissions reductions by 2050 than other policies modelled (which are discussed further below).[33] The CCA also found that regulated closures would be a more expensive means of reducing carbon emissions than market-based mechanisms:

[The CCA's] analysis of regulated closure indicates that using it to achieve a large post-2020 emissions reduction goal in the absence of other measures in the electricity sector would entail higher costs than other policies and would not offer a direct incentive for new low-emissions plant to be built.[34]

3.35      Choosing plant age as the basis for progressive power station closure under this model may also not produce the most efficient outcomes. Jotzo and Mazouz argue that the information asymmetry between governments and plant owners is a significant drawback to the directly regulated closures model:

Direct regulation suffers from government not having sufficient information about business cost structures, and therefore it would be difficult for the regulator to identify which plant would be the most cost-effective to close and how much to offer in compensation if such compensation was offered.[35]

3.36      Further, they argue that in Australia's current political context 'it appears unlikely that a government would choose a pure regulatory approach that singles out power stations and imposes the full cost of early closure on the owners of that station'.[36]

3.37      Associate Professor Jotzo commented further at a public hearing:

The regulated approach, according to a timetable, age or emissions intensity would obviously give great predictability of the schedule of exit. In my view, it has the disadvantage of not being the least-cost pathway. Almost by definition, the least-cost pathway of exit will deviate from 45 years out or whatever it may be. If a government wanted to go down the regulatory closure pathway, you would want to combine that with flexibility instruments such as tradeable operation rights.[37]

3.38      Doctors for the Environment Australia recommended that the degree of pollution and its danger to local communities should be a major factor in deciding priority for closure and in advising community and workers of the need for closure. It noted that several states in Australia already impose pollution licensing fees on power plant operators that could in theory drive the closure of heavily-polluting plants, but argued that these schemes 'have been ineffective due to the inadequate scale of fees imposed'.[38]

Market-based mechanisms

3.39      The CCA concluded in a research paper in August 2016 as part of its Special Review that a market‑based mechanism to reduce carbon emissions should be implemented in the Australian electricity supply sector:

A market mechanism in the sector would allow Australia to meet its targets at a lower cost to the community than would be possible without such a policy in the toolkit. The sector's characteristics (measurable emissions, relatively small number of large emissions sources, sophisticated profit-seeking investors operating in generally competitive generation markets) suggest market mechanisms will be feasible and more cost-effective than the alternatives. In addition, market mechanisms can be scaled to achieve deep emission cuts, and are flexible to changing market and technology conditions.[39]

3.40      The market-based policies considered as part of the CCA's review included: a cap and trade scheme; an emissions intensity scheme; a carbon tax; and a baseline and credit scheme.

3.41      Under all the policy scenarios modelled as part of the CCA's review (including the direct-regulation models discussed above), coal fired generation would decline significantly in Australia in the medium term. As Jacobs stated in its final modelling report, all the policy scenarios modelled involve the entire brown coal fleet and two-thirds of the black coal fleet being decommissioned by 2030.[40]

Emissions intensity schemes

3.42      The CCA ultimately recommended the introduction of an emissions intensity scheme for the electricity supply sector in Australia.[41] Jacobs gives an overview of how such a scheme would operate in its modelling report undertaken for the CCA:

An emission intensity baseline is set for the electricity supply sector as a whole (based on tonnes of carbon dioxide equivalent per megawatt hour sent out). All generators are allocated permits (representing one tonne of carbon dioxide equivalent) equal to their own generation multiplied by the baseline. At the end of the compliance period all generators surrender permits for each tonne of carbon dioxide equivalent emitted. This effectively means that generators with intensity below the baseline have surplus permits to sell (so receive a subsidy) and generators with intensity above the baseline need to buy additional permits (so incur an extra cost). Emissions permits can also be banked indefinitely for future use or borrowed in limited quantities.

Demand for permits available in each year creates an explicit carbon price, and the relative price of electricity made from more emissions-intensive sources increases. In contrast to a conventional cap and trade scheme, there is no absolute emissions cap, so in practice overall sectoral emissions will vary depending on electricity demand.[42]

3.43      Jacobs' modelling on this policy scenario predicts that during the first decade of implementation (that is, 2020–2030) all coal fired power stations are shut-down as a result of the imposed policy, with mostly wind generators and combined cycle gas turbines replacing the retired capacity.[43] The generation mix for electricity supply in Australia to 2050 under this scenario is shown in Figure 3.1.

3.44      Origin Energy stated its support for a mechanism like this to manage the transition to a low-carbon electricity sector:

Origin supports the progressive decarbonisation of the electricity sector in Australia and an eventual goal of net zero emissions by 2050 or earlier. We believe the introduction of a well-designed cost of carbon abatement for the electricity sector, such as an emissions intensity scheme, is the key to managing this transition.[44]

Jotzo model for regulated closure of brown coal power stations

3.45      Jotzo and Mazouz advocate for a different type of market-based mechanism to drive the closure of the most emissions-intensive brown coal station(s) in Australia.[45] They argue that in the absence of any policy intervention, the economics of Australia's fleet of coal fired power stations is such that black coal stations may close operations first, before the more emissions-intensive brown coal fired stations.[46] This would lead to poorer environmental outcomes in terms of overall carbon emissions and air pollutants than if brown coal capacity was closed earlier and black coal generation capacity remained online.

3.46      Their suggested model is in effect a hybrid market-based regulated closures model. It is summarised as follows:

The principle of the proposed mechanism is that government offer power plants the opportunity to bid for the closure of some amount of capacity, leaving it to the bidding process to determine which plant(s) will close and what the magnitude of the payment to the closing plant is. The remaining plants are then mandated by government to make financial transfers to the plant that exits the market, in line with their emissions.[47]

3.47      Jotzo and Mazouz argue that such a mechanism would: provide emissions savings from plant closure at least cost; rely on a market mechanism to identify which plant should close and what magnitude payment is required; avoid budgetary costs by sourcing the payments for closure from the power stations remaining in production; and provide some incentives to adjust the power mix to reduce emissions.[48]

Competitive bidding process to identify which stations to close

3.48      Under the Jotzo model, relevant plants (most likely Victoria's brown coal fired power stations) would be invited to submit a bid for the amount of money they would be willing to accept in return for ceasing operations by a predetermined date, remediating their plant site and funding an assistance package to their workers and local communities. A government regulator would then assess the bids, alongside the likely emissions savings resulting from each possible closure, and choose the most cost-effective bid.[49]

3.49      The generator chosen for closure would then receive the full amount specified in their bid, in pre-determined instalments, paid for by the other generators remaining in the market. Under Jotzo's preferred model, the share of payments each remaining generator would need to contribute would be determined on the basis of their carbon dioxide emissions during the year following the closure of the chosen plant, creating further incentives for high-emitting plants to submit low bids in the bidding process.[50]

3.50      Jotzo and Mazouz consider that the cost of such plant closure (and its capacity exiting the market) would be reflected in some rises to electricity prices. They estimate an increase of five to 14 per cent in wholesale prices over the course of one year (and dropping again afterwards), with a corresponding increase in retail prices in the order of one to two per cent, over one year.[51]

3.51      Associate Professor Frank Jozto discussed his model with the committee at a public hearing:

Our proposal, in a nutshell, is for a market mechanism whereby existing power stations submit bids as to financial compensation required to shut down according to a pre-agreed time line. A government or regulator would choose the most attractive bid, which may well be the bid that delivers the greatest expected emission savings per dollar of compensation required. This is a competitive process—best bid wins—and the money is then levied on the remaining power generators. The logic behind that is that these are the power generators that will benefit through increased capacity utilisation of their plants and, to some extent, through increased prices in the wholesale market. This would enable for exit according to a timetable. It would create a source of funding for structural adjustment, and possibly also for improved site rehabilitation above and beyond the level that is required by law of the exiting companies.[52]

Criticism of Jotzo model

3.52      The Jotzo model has been criticised, most notably by Frontier Economics in a May 2016 paper.[53] This paper argues that the predicted electricity impacts of a closure of one of Victoria's brown coal power stations, as advocated for in the Jotzo model, would be much more significant than Jotzo and Mazouz allow for. Frontier's estimates are that retail prices would rise by up to 25 per cent in Victoria in the year immediately following closure, with sustained price rises of 9 per cent in following years, as well as less severe price rises in New South Wales and South Australia.[54]

Options for implementation of policy combinations and need for further research

3.53      Stakeholders highlighted the fact that a combination of policies may be required to effect an orderly exit from the market of coal generators and concurrent increase in generation capacity from renewable sources.[55] In particular, some argued that the continuation of a large scale renewable energy target beyond 2020, when coupled with other policy mechanisms to constrain emissions from coal generators or regulate their closure, would be the most effective means of managing this transition.[56]

3.54      Associate Professor Jotzo made the point that currently, research on options to facilitate closure of coal fired power stations in Australia has been relatively limited. He argued that additional work is required to fully understand the options and provide input to policy, including by further investigating:

Navigation: Previous Page | Contents | Next Page