Skip to content

Cost Of CfD Subsidies – 2021

January 25, 2022

By Paul Homewood

We looked at how the CfD system was operating during December yesterday. But we really need to take a step back and look at last year as a whole to get the full picture.



During the year, there was as we know a steady rise in prices, with a big jump in the autumn.

CfD Generation 21.7 TWh

Av Strike Price

Av Market Price £97.75
CfD Subsidy £996.6 million

The CfD system generated a refund to bills of £83.1m in December,as I explained yesterday, but overall we are still £996.6m worse off.

Meanwhile generators signed up to Renewables Obligation, the system which preceded CfDs, were paid subsidies of £68.56/MWh in the year to Sep 2021. This is, of course, on top sky high revenue for sales of electricity they are now raking in.

  1. It doesn't add up... permalink
    January 25, 2022 3:40 pm

    The breakdown of the CFD payments for 2021 by technology was

    £402,294,339 Biomass Conversion
    £4,162,640 Energy from Waste with CHP
    £612,269,129 Offshore Wind
    -£21,748,364 Onshore Wind
    -£205,803 Solar PV
    £996,771,941 Total

    The average payment per MWh was worth

    £62.30 Biomass Conversion
    £29.62 Energy from Waste with CHP
    £44.85 Offshore Wind
    -£15.88 Onshore Wind
    -£7.56 Solar PV

    • January 26, 2022 10:09 am

      So is it correct to say the recent “reduction” in electricity price due to wind power was solely due to ONSHORE wind? It looks to me that offshore wind still got a massive subsidy of £612.3 million.

      • It doesn't add up... permalink
        January 26, 2022 12:39 pm

        No, that’s not correct. Market prices are dictated by whatever is the marginal source of generation at any particular time. See my reply below to Phoenix44. These numbers reflect the difference in guaranteed CFD prices between onshore and offshore wind. Offshore units are much more expensive to build, so they have a higher guaranteed price to makethem viable. The wind generators that set market prices during periods of surplus are the cheapest to curtail, which turn out to be onshore wind getting ROCs. That picture is unlikely to change until there are wind farms operating under CFDs that guarantee less than the cost of an ROC, or when there are wind farms from the next CFD round currently being conducted, but that is several years away at best.

  2. Phoenix44 permalink
    January 25, 2022 3:41 pm

    If I understand this correctly, then the CfDs work to produce a low volatility in prices only when there’s plenty of wind, because only then does the strike price have a significant influence on the average price. As wind production falls, the influence of market prices in the average price increases.But there will be a reasonable correlation between low wind and high market prices because one causes the other and very little correlation between high wind and low prices because we will be taking wind generation at the strike price.

    Thus renewables do reduce volatility somewhat but but unlike a proper fixed price contract, we are still exposed to both high and low market prices but with a much greater exposure at the high end and with the risk that our exposure increases our exposure!

    This seems to be the opposite of what MPs like Skidmore claim.

    • It doesn't add up... permalink
      January 25, 2022 7:06 pm

      When there is plenty of wind then market prices are driven down by the surplus: since those who have CFDs do not care what the market price is until it falls below zero they will be prepared to sell at any positive price or even any negative price that is less negative than their strike price so they get a positive revenue (CFDs do not offer protection against any negative element of price, so e.g. if the CFD strike price is £150/MWh, they get £150/MWh at any market price between zero and plus infinity, but only £150/MWh of compensation when prices are negative, so if the price is minus £60/MWh they end up with £90/MWh net. There is provision that if prices stay negative for six or more contiguous hours then there is no CFD payment at all for the duration: in these circumstances they will curtail output. Mostly they manage to bid the sixth hour up to a mildly positive price to maximise on subsidy some of which then goes to the benefit of export customers as we export at negative prices.)

      Wind farms earning ROCs will carry on producing until the value of their ROCs are extinguished by negative pricing, or they will demand at least equivalent compensation to curtail. This is normally the factor that sets negative prices in times of surplus.

      In more normal conditions, CCGT is the marginal source of generation that flexes as wind output and demand change, and its costs therefore set the market prices. There is no single cost for CCGT as actual costs vary according to how intensively a station gets run, the skill of the traders in purchasing its gas supplies, its own technical performance characteristics, its location etc. However there is a broad band that reflects the range of costs. In the current very volatile gas market the range of costs is much wider than normal. Even a CCGT station that hedged ahead might decide in the first instance simply to resell its gas to help meet other gas demand and bank the profit on that before being called on to operate to meet a demand shortfall, then having to pay top dollar for supply to cover that peak. All CCGT stations face the alternative of not operating and reselling any gas they bought ahead of time.

      At the other extreme, market prices are set by the demands of plant used for peak provision. We now have the absurdity that these are now often coal, which has added costs for ramping up and down to be available – yet they can charge almost whatever they like to keep the lights on.

      We will see progressively more surplus pricing as wind capacity is increased. At the moment, it only tends to produce surpluses at times of low demand in windy conditions. As capacity increases, those surpluses will get bigger, and there will be more hours when wind output exceeds demand (less must run generation required to stabilise the grid). The consequence is more curtailed output and higher prices needed to provide for the extra costs of providing backup when the wind doesn’t blow. Curtailed output is a deadweight cost that has to be paid for either through curtailment payments or higher prices for useful output – which is much the same thing. In other words it gets more expensive from here on.

      • Mikehig permalink
        January 27, 2022 11:10 am

        Idau: thanks for that excellent explanation of this fiendishly complicated system. I almost feel that I understand! (Won’t last!)

        There’s another aspects of the CfD set-up which puzzles me: why are the strike prices 100% escalatable?
        The majority of wind farms’ cost is repayment of the capital and the financing cost thereof. Both are fixed for the contract life: the capital has been spent and the financing set up at a fixed rate (opting for variable rates seems very unlikely).
        I used to work in a business which installed and ran plants for customers. The contracts usually comprised a fixed facility charge to pay for the investment, etc together with an operating charge which was adjustable in line with the costs of labour, materials, etc..
        I do not understand why the whole strike price is escalatable given that a large part of the underlying costs are fixed.

      • Jordan permalink
        January 27, 2022 2:35 pm

        We can do it either way Mike.
        The CfD could be a 15 year fixed nominal price, where the producer has no inflationary increases. The producer will then factor-in a view of inflation (including their view of the risk of inflation taking off), to arrive at the price to be held for the 15 years. As it has inflationary adjustments included, it will be higher than today’s prices (over-recovering costs in early years and under-recovering in later years).
        The public will not grasp the subtlety of the fixed nominal price and will only look at the value of the price. So an alternative is to state a price at a date (e.g. April 2012) and include an inflationary adjustment for each of the 15 years into an annual nominal price. This gets the lower “headline” price that people will talk about. Some will even claim it is “cheap”.

      • Mikehig permalink
        January 27, 2022 3:58 pm

        Jordan; Thanks. After posting I did wonder whether this was a bit of smoke and mirrors to make things look cheaper than they really are.
        Paul put up figures for the Achlachan wind farm which show that the strike price has increased by nearly 12% in a bit over 3 years – call it 4% pa, roughly.
        However Opex accounts for only 30 – 40% of their electricity cost, from what I have read, so an increase of only 1.2 – 1.5% on the whole price would be enough to cover it.
        This does make it look as if the headline price has been artificially lowered by a bit of financial jiggery-pokery. Even if this was done on an NPV-neutral basis it would cost more overall due to “back-loading” the capital recovery. I very much doubt any such altruism on the part of the promoters.
        My suspicion is that this format is more lucrative compared to a more logical approach of a fixed charge to cover the investment etc and an escalatable charge for Opex costs.

  3. Tim Pateman permalink
    January 25, 2022 5:53 pm

    Paul, is there a record of how much we paid, either for energy we could not use and had to sell on at a loss, or had to discharge, or for restraint payments for them to switch off when their wind energy could not be used for any reason. Plus all the associated Grid balancing costs for 2021.

Comments are closed.

%d bloggers like this: