Skip to content

UK Capacity Market “Back In Play”

November 11, 2019

By Paul Homewood

Timera bring news that the UK’s Capacity Market Auction has been deemed to be valid by the EU, following a spurious legal challenge:


The UK capacity market was re-instated two weeks ago. This means the immediate resumption of payments for both on-going and historical capacity provided.  Importantly, the European Commission found no evidence of discrimination against Demand Side Response (DSR), the claim that underpinned market suspension.

During the period of suspension, the UK government has run a separate consultation process on reforms to the capacity market. This should result in some relatively minor adjustments e.g. wind & solar are set to participate from Jan 2020 provided they receive no other support. But in summary, the capacity market is now back in play and essentially unchanged.

Reinstatement means that three auctions will now be held across the space of five weeks in Q1 2020.  These will play a very important role in shaping the UK power market capacity mix into the mid-2020s.

In today’s article we look at the current UK capacity balance ‘state of play’. We also set out why we think capacity prices may surprise vs a pessimistic market consensus.

UK capacity balance: a 15 year perspective

There is a large & growing pipeline of UK wind and solar projects under development. Our scenarios are for between 25 to 30GW of aggregate nominal wind & solar capacity build across the next 10 years.

In Chart 1 we show a scenario for capacity growth from Renewable Energy Sources (RES) in the blue bars above the horizontal axis. This is broadly in line with National Grid’s Future Energy Scenarios, recognising some differences in timing and breakdown of capacity types.

But this large nominal volume shrinks to a relatively small volume once capacity is derated for intermittency.  For example, 23GW of nominal RES growth by 2030 in the scenario in Chart 1 shrinks to 2GW when derated (using the latest wind & solar de-rating factors announced for the Q1 2020 auctions).

Chart 1: The net capacity deficit – RES build vs capacity closures

The grey lines below the axis in Chart 1 show a scenario for aggregate thermal retirements.  This assumes:

  1. Coal asset closures as announced, with Drax and Ratcliffe closing by end of 2024 in line with the government’s 2025 deadline
  2. Nuclear plant closures as per current regulatory schedule
  3. CCGT closures based on major capex timing and economics of individual units.

Derating has much less impact on thermal units (e.g. CCGTs have a 90% factor vs 7% for onshore wind and 3% for solar).  In the chart scenario, 23GW of derated capacity is set to come off the system by 2030.

The blue line shows the net derated capacity gap between thermal closures and RES build.  These are primarily replacement volumes to be delivered via the capacity market. In other words 23GW of net closures and 1GW of RES build leaves a capacity deficit of 21GW by 2030.

That is a lot of capacity.  And as things stand, it is likely to be primarily delivered in the form of gas engines, batteries, DSR and CCGTs.  It is also possible that merchant RES starts to play a more important role from the mid-2020s, although still with more penal derating for intermittent sources.

Why two of the Q1 auctions are important

The T-3 auction will be held on 30th Jan 2020 for capacity delivery in the 2022-23 year (Oct – Sep). Then a T-4 auction will be held for 2023-24 delivery on 5th Mar 2020.  These two auctions cover a key window of coal and CCGT closures.  A ‘top up’ T-1 auction for 2020-21 will be held in between these, although today we just focus on the more important T-3 and T-4 auctions.

There is 9GW of coal capacity remaining in the UK. In 2021 this will likely fall to 3GW (i.e. Ratcliffe and 2 remaining Drax units).  If the current weakness in coal generation margins (CDS) continues, this 3GW may close earlier than assumed in Chart 1 (e.g. by 2023).  That could accelerate the requirement for new capacity across the Q1 auctions.

In parallel, 3-5GW of older and less flexible CCGT units are likely to close by the mid-2020s. These are mostly 1990s commissioned units that were not designed for lower load factor flexible operation and are reaching major life extension capex milestones.  The level of capacity prices will influence closure timing.

These closure volumes of coal and CCGT units are creating a structural capacity deficit across the auction horizon.  This means new build capacity will be required and it will likely be more expensive than in the last T-4 auction.

Capacity price pessimism vs bidding reality

One of the classic human traits that can drive investor behaviour is ‘recency bias’. We tend to place too much emphasis on recent events when making investment decisions.

In this context, the last T-1 auction clearing price (0.77 £/kW) and the last T-4 price (8.40 £/kW), shown in Chart 2, have set a tone of heightened pessimism around future capacity prices.

Chart 2: UK Capacity Auction volumes and contract price


The take home message is that we are likely to see the closure of 30GW of conventional capacity within the next ten years, including all the remaining coal capacity, much CCGT (particularly the older plants not designed for flexible working, and older nuclear.

A similar amount of renewable will be built, but crucially the derated capacity will be much less. (Arguably, of course, the “derated capacity should be zero”).

Even within five years, we could lose nearly 20GW.

The reinstated Capacity Market is designed to fill this gap. But the question is how much will we have to pay.

Timera reckon a price of around £20/KW/Yr may be enough, as it was in the earlier auctions, costing around £1bn a year. I believe that figure is underestimated. Previous auctions were awarded largely to existing capacity, coal, gas and nuclear. Consequently prices were kept low.

With much of that capacity due to close soon, the Auction will need to attract a tranche of new capacity. With an assumed lead time of five years, new CCGT plant will need to be successful at next year’s auctions, if sufficient capacity is to be ready by 2025.

Based on BEIS levelised costs assumptions, I calculate that the fixed/capital costs for a 1GW CCGT plant are about £90m a year. Given the uncertainties going forward, concerning carbon pricing, subsidised renewables and political threats, there is no guarantee that new gas plants will be able to operate for their full technical life (certainly without investing in CCS technology). Indeed, there is little confidence that they can even make money from selling electricity in the short term up to 2030, as subsidies to renewables prevent them from running at full capacity.

They will consequently need to cover most, if not all, of their fixed costs from the Capacity Market, via 15 year contracts. At £20/KW/Yr, a 1GW plant would only receive £20m a year.

It is also worth recalling that the proposed new Trafford CCGT plant, which won a Capacity Market subsidy of £30m a year in 2014, never got off the ground, as it could not attract finance.

My guess is that new plants will need at least double that subsidy to be worth investing in.

The Capacity Market auction runs on a cleared basis, meaning that all successful bidders receive the highest price at which the auction clears. At £40/KW, therefore, the cost to consumers would be around £2bn a year.

  1. November 11, 2019 6:59 pm

    Reblogged this on Climate-

  2. Joe Public permalink
    November 11, 2019 7:03 pm

    So the Green Party’s Fabricate-a-Figure-then-four-fold-it 10-year plan is already outdated.

    What a pity.

  3. Mike Skuse permalink
    November 11, 2019 7:20 pm

    I’m afraid I can’t understand much of this. But if you would like a simple down to earth resumé of why all this stuff is so much nonsense, find on youtube “is carbon dioxide making earth greener” which is Prof William Happer, who is the chief physicist at Princeton Uni.

    • November 11, 2019 8:18 pm

      Basically Mike, because we are mandating and subsidising huge amounts of unreliable wind and solar power, we have to pay for reliable generators , such as coal and gas, to standby doing nothing in case the wind stops blowing or sun stops shining.

      The cost of this ends up being paid by electricity bill payers

      • Mike Skuse permalink
        November 11, 2019 10:30 pm

        Thanks. I certainly understand that!

      • Phoenix44 permalink
        November 12, 2019 8:48 am

        But not “doing nothing”. Are they not burning fossil fuels but not dispatching power? Stand by generation dies not start from cold.

  4. The Man at the Back permalink
    November 11, 2019 8:21 pm

    I guess I will be pushing up the daisies before this all comes tumbling down – I say pushing up daisies because surely burning me to a cinder will have too large a “carbon footprint” (whatever that is).

    I was pondering all this stuff earlier today. The lunacy gets more stupid every day with political parties competing to ruin the UK in fewer and fewer years (perhaps I will be around to see it).

    As every day goes by it becomes more and more evident that the CO2 GHG effect is so small no one will ever be able to determine its contribution to the whole. The whole scientific fraud is a political edifice. Increases in CO2 are such a small part of the GHG effect with water-vapour being the dominant constituent, it cannot have a measurable effect.

    More than a decade ago (and I can’t find the links now) a retired Dutch scientific engineer worked out that the whole temp rise of the 2nd half of the 20th century was due to reduced cloud cover and therefore reduced albedo, and increased isolation. Sadly he died suddenly only a few months later. Now recently we have –

    And also

    But the usual unscientific gatekeepers of climate science – the one club golfers still hold sway.

  5. Bloke down the pub permalink
    November 11, 2019 9:00 pm

    Panorama prog on tonight about airline industry emissions. Seems relatively balanced for the BBC, taking into account of course their total submission to the climate cult.

  6. November 11, 2019 9:49 pm

    If generation does not equal demand then system frequency will quickly go outside limits leading to collapse.

    Load shedding is one form of (emergency) demand-side response (DSR) when there is insufficient generation.

    Another example is when scheduling for a winter tea-time peak, should one add (real) generating capacity, or call on consumers to reduce demand through prearranged agreements.

    Obviously, there are many scenarios but the idea of DSR is quite sound – simple.

    However, I’m struggling to figure out what it is that is being said here –

    The EU must (still) have a say in the matter especially if the ‘subsidy’ comes from it?

    • It doesn't add up... permalink
      November 12, 2019 3:35 pm

      We all could be paid to disconnect from the grid provided the payment were sufficient to fund our own private generators and diesel supply – which is of course how much of the existing DSR and Triad avoidance operates already. That might make economic sense for handling peak demand, but it makes no sense at all when the purpose is to provide rather more sustained backup for missing wind or solar generation. Then DSR becomes failure of the grid to provide adequate capacity, and is in effect blackouts with compensation – or even simply load shedding with no compensation.

  7. November 11, 2019 10:16 pm

    To make renewables work reliably and efficiently I believe that the idea of a ‘national grid’ would have to be partially abandoned. Instead, the system is divided up into ‘micro’ networks which hark back to the days of ‘town gas’. The gasometers would be replaced with tesla battery packs and only industrial consumers would be interconnected to major sources of generation. Just kidding, but, we are heading into unknown territory!

  8. Steve permalink
    November 12, 2019 8:13 am

    The Libdums have joined Labour in calling for zero carbon by 2030. The derated capacity should definitely be zero with storage incapable of supplying during a midwinter lull. 20GW to find to replace gas and nuclear by 2030 and no viable carbon capture for CCGT or reformed hydrogen. Someone please explain the figures to dim Jo and Magic.

  9. Ian Cook permalink
    November 12, 2019 6:18 pm

    With all these climate change scientists who are, rather strangely, habituated to get together and give us their latest update on the Climate Emergency, lying around, could they answer a question I have. We get that they don’t like being challenged on their beliefs, so let’s take them at their word that they know everything about climate and ask for the answer to this conundrum; what caused the Little Ice Age that ended in the 19th Century? And, before you go into standard climate change mode, I do mean that you show your working-out.

  10. It doesn't add up... permalink
    November 12, 2019 10:02 pm

    I think it’s difficult to assess what the income stream might look like by the time you include some extra ancillary services on top of the capacity payments: batteries are earning £7/MW of capacity per hour (which would be £61.32m p.a. if scaled up to an always on 1 GW plant – but that’s not going to be the case for a CCGT plant that fills in for missing wind or handles the diurnal variation in demand) and more for providing Enhanced Frequency Response for instance. But if I look at the levelised cost estimates you refer to, we see that CCGT totals £12/MWh for fixed costs against an unstated utilisation, but if we assume 85%, that’s a cost of £89.35m for a 1GW station. Looking at the OCGT costing based on 500 hours of operation, the total fixed element is £85/MWh produced, or £42.5m p.a. for a 1 GW station. It is perhaps no surprise that we see planning applications for OCGT stations, despite their much lower fuel efficiency, reflected in the £80/MWh fuel and carbon tax cost for OCGT vs £54/MWh for CCGT (the fuel cost assumes some steep increases in gas prices). Of course, at higher utilisations OCGT is not at all economic, since the extra fuel cost completely outweighs the lower capital cost.

    Batteries are of course quite useless for providing capacity for anything other than the shortest of peaks – few are capable of delivering at capacity for more time than it takes to fire up an OCGT plant. They are certainly no replacement for major capacity shortfalls in periods of low wind and sun.

    Power cuts, or back pedalling about plant closures probably loom. Meantime, expect a lot of pressure to complete interconnectors to the Continent and Norway. I don’t fancy being reliant on them to keep the lights on. I noted that National Grid essentially commented in their WInter Outlook report that Carbon Floor Tax is helping to ensure that UK prices remain at a premium to those on the Continent, thus encouraging export of electricity to the UK. It also ensures our electricity costs are uncompetitive.

  11. The Man at the Back permalink
    November 13, 2019 8:57 am

    O/T but I recently glanced at an article on one of the many blog/websites that I read each day. It’s title was something like “How many skeptics at NASA”. I forgot to bookmark.
    It mentioned a retired Climate Scientist who is a sceptic and has written a book. I intended to go back to it, but can’t now find it.
    Did I just dream it? Can anyone help?

Comments are closed.

%d bloggers like this: