It is a well understood fact that the UK will have to replace a large volume of its generation capacity this decade. The security of supply issues associated with plant retirements have been a key driver behind the recent launch of the government’s Electricity Market Reform (EMR) policy measures. Indeed the focus of EMR as it has been presented, is the replacement of old and dirty plant with new low carbon technology. But what volumes of plant are likely to retire and how much of a role can low carbon capacity play in replacing it?
UK generation capacity closures by 2016 and 2020
In order to outline the scale of generation capacity closures, it is helpful to break the problem down into closures that have been scheduled for regulatory reasons (predominantly coal, oil and nuclear plant) and potential closures due to plant economics (predominantly CCGT gas plant) as set out in Chart 1.
There are 11.5 GW of scheduled coal and oil plant closures by 2016 as a result of generators opting out of the Large Combustion Plant Directive (LCPD). In the absence of a major policy backflip (never to be ruled out) these closures will happen. In addition about 6.5 GW of nuclear capacity is scheduled to close by 2020, about 2.8 GW of this by 2016. If there is a capacity crunch it is possible some of these plants may receive life extensions, but the regulatory process around extending nuclear plant lives is not a simple one. Scheduled LCPD and nuclear closures form the black and the grey bars in Chart 1.
Greater uncertainty exists around the volume of retirements of older CCGT plant (built in the ‘dash for gas’ in the 1990s). The purpose of this article is to define a reasonable range of potential plant closure volumes rather than try and predict the exact outcome. In order to do this we focus our analysis of gas plant retirement on the plant owner’s decision to close the plant or invest in major capital expenditure to extend plant life. The timing of this decision is driven by the number of plant starts and operating hours (occurring at around 100,000 equivalent operating hours). However it is not unreasonable to use a plant age of 20 years as a proxy for the time by which major capital spend will be required to keep the plant operational.
There are 9.5 GW of gas plant that will reach 20 years of age by 2016 and an additional 9GW which will reach 20 years by 2020. In addition there is at least 5GW of older coal plant that may close by the end of the decade in response to the carbon price floor and weak dark spread environment. The potential volumes of gas and coal plant retirements are represented by the red shaded areas in Chart 1.
To summarise the capacity retirement picture, it is likely that at least 15 GW of coal, oil and nuclear plant will close by 2016, with 20GW closing by the end of the decade. In addition there is the potential for a further 20-25GW of mostly CCGT capacity to retire by 2020. If, for the sake of argument, half of these potential closures eventuate, it would increase total plant retirements to around 20GW by 2016 and 30GW by 2020.
We have not attempted to overlay predictions of demand growth or capacity margin on top of this picture. However the numbers above give a sense of the capacity replacement problem at hand and the importance of gas plant new build and life extensions in determining the eventual capacity outcome.
Low carbon contribution to capacity replacement
The government has set out a range of policy measures designed to support low carbon technologies. But to what extent can new low carbon capacity counteract thermal and nuclear plant retirements?
The delays in EDF and Areva’s prototype Flamanville plant in France means it is realistic to assume new nuclear will play no role in the UK until after 2020. The delivery of CCS projects in the UK is inhibited by a lack of clarity on policy support and a wariness of project risks from developers. At best an optimist might assume that the UK’s CCS competition and EU funding together could deliver 1GW of new capacity by 2016 and a total of 3GW by 2020.
That leaves renewable capacity, with onshore wind, offshore wind, biomass (in the broadest sense) and solar being the only technologies that can realistically be delivered in meaningful volumes by 2020. Again we are not in the business of predicting outcomes but we have constructed two simple scenarios to reflect the potential impact of new renewable capacity based on build rates to date. Renewable capacity volumes are then overlaid on the scheduled and potential capacity closure volumes from Chart 1 to illustrate renewable contribution to capacity replacement. It should be noted that wind and solar capacity volumes are de-rated for average historical load factors between 2007-10.
Growth in renewable capacity is projected based on a multiple of the average historical build rate by capacity type from 2007-2010. In this scenario it is assumed that build of biomass and onshore wind continues at the historical average rate, which may still be a substantial challenge given planning and resource constraints. Offshore wind and solar are assumed to grow at a faster than historical rate given technology cost reductions and more targeted support. Assumptions on build rate multiples and implied annual MW of capacity build are shown in the table to the right of the chart.
To replace the volume of scheduled UK plant closures under this scenario (the black bar in the chart), around 8GW of new gas capacity would need to be commissioned on top of low carbon build. Additional new gas capacity would be needed to replace retirements of existing thermal plant. For example if half the potential retirements of gas capacity (the red bar in the chart) eventuate at the 20 year stage, an additional 12 GW of new gas capacity would be required to counter plant closures (on top of the 8GW to meet scheduled closures). The market would also require additional thermal plant backup to cover the 5GW of de-rated intermittent capacity operating by 2020.
Again, this simple analysis does not explore the evolution of demand and the power market capacity margin. New capacity build may fall short of plant retirements without the lights going out, but the importance of thermal capacity to maintaining security of supply is very clear.
In this scenario we stretch the boundaries of optimism. It is assumed that build rates for biomass and onshore wind immediately accelerate to 150% of the historical average. Build rates for offshore wind are assumed to triple and solar to increase fivefold. The government may argue that even greater build rates can be delivered if appropriate financial support is provided under the new EMR policy incentives. However we have specifically defined our build rate assumptions in the context of historical build rates because there are substantial non-financial constraints to new build, specifically in relation to planning and transmission access. Assumptions are again shown in the table to the right of the chart.
Despite the optimism, the incremental difference to the amount of new gas capacity required to fill the scheduled closure gap is remarkably small (2-3 GW less). In addition the market now needs thermal backup to cope with 7GW of (de-rated) intermittent wind and solar capacity as opposed to 5GW in the first scenario.
Thermal capacity is the key to security of supply
The conclusion from the simple analysis above is clear: it is gas capacity not low carbon capacity that is critical to maintaining security of supply in the UK. This is not an argument against supporting the development of low carbon capacity, but it raises big question marks over the government’s EMR policies. As we set out previously, the carbon price floor directly increases thermal generation costs and other EMR policy interventions weaken the market price signal and increase thermal generator risk. Add in a backdrop of weak market spark spreads and capital constraints and there is a good case for the UK facing a capacity crunch and significant increase in market volatility by the end of this decade.