Skip to content

The Real Story Behind Wind Farm Constraint Payments

August 18, 2019

By Paul Homewood


One common defence of the obscene amounts paid to wind farms to cut output is that these constraint payments are made to all types of generators.

This is grossly misleading. As the REF show, constraint payments for wind farms have risen from virtually nothing a decade ago to £124 million last year, purely because of the intermittency of wind and solar power.


Of last year’s £124 million, £115 million alone went to Scottish wind farms, according to the REF, because of the lack of enough transmission capacity to take surplus wind power into England, where the demand is.

With offshore wind capacity now just beginning to rapidly expand in England, we can expect these payments to drastically rise. According to the Telegraph:

Ben Guest, a specialist in renewable energy companies and markets at asset management firm Gresham House, said compensation could even rise to £1 billion a year – nearly six times last year’s payout – in the foreseeable future.

“The amount of power delivered by renewables in the coming years is going to result in far larger amounts of over-supply than you see today,” he said.

It is anticipated that power generated by renewable energy could increase by up to 50 per cent over the next five to ten years as more wind farms are established, many set up to take advantage of generous government subsidies.

To make matters worse, these constraint payments are not just for the value of electricity that would have been produced, but include the subsidies foregone as well. This year, for instance, the payments average £70/MWh.

The National Grid has an array of tools for ensuring that demand and supply is balanced. These are known as Balancing Services Use of System (BSUoS), and cost in the region of £1bn a year.

A large proportion of this cost is designed to ensure there is plenty of reserve capacity available. However constraint payments are now growing rapidly as a share.

The reason is simple. Before intermittent wind and solar power came on board, the National Grid merely had to estimate what demand would be for a certain time on a given day, then arrange that there would be sufficient capacity available.

Now, as well as estimating demand, they have to forecast how much generation there will be from renewables. Inevitably there will be times when there is too much output, and constraint payments are wheeled out. The alternative is of course even worse, that they overestimate.

An interesting case study that shines a light on this occurred last weekend, Aug 10th and 11th. This was when the Telegraph reported that Hornsea wind farm was paid £100,000 to switch off, just a day after they triggered the black out.

The National Grid publish their BSUoS charges for each half hour here. So far this financial year, these have averaged £1.9 million a day. As I say, these include costs for reserves as well as constraints.

On Aug 10th and 11th, these increased to £5.5 million and £4.1 million respectively. Moreover, the bulk of these payments were made between midnight and 7.00 am , when demand was lowest.

Both days were extremely windy, and certainly demand was low, so it is reasonable to assume that most of these costs related to constraint payments for wind farms.

BM Reports for those two days show (circled) the drop in demand in the early hours. They also show the volatility of wind power over the two days.



Wind power was running at well above 8GW for much of that period, around 40% and more of total generation. It is pretty clear that the National Grid rightly believed that such a high reliance on wind power was far too risky.

Meanwhile nuclear was generating flat out at 6GW, with gas and biomass at about 5GW.

If they had not constrained a dollop of wind power, the grid would have been far too reliant on the wholly unreliable wind power, thus risking another major blackout.

  1. Trevor Shurmer permalink
    August 18, 2019 10:05 pm

    8gw, not mw?

  2. GeoffB permalink
    August 18, 2019 10:32 pm

    i had always expected the electricity to fail in the middle of winter, when demand is high, due to not having sufficient capacity . However it seems more likely that failure is going to occur when the demand is moderate and there is not enough inertia (spinning mass) to damp down frequency variation particularly when wind is a high percentage. of generation I do not see how the grid will be able to take the proposed wind generation on the new Scottish wind farms , Moray and Shetland, particularly at 3am in the summer. Even more capacity constraint payments…..the whole situation is a costly farce.

    • buxtond permalink
      August 19, 2019 8:49 am

      It was designed so to be! the powers to be are deliberatly stealing from us all for their virtue signalling, because that is all it is. Natually the ace is the part that Deben is playing, not an Honourable man just a thief in a powerful Committee which shoulds not even exist!

  3. roger permalink
    August 18, 2019 10:44 pm

    Not content with taxing electricity to the extent that those poor people at the margin have at times to choose between warmth or food, according to today’s Sunday Telegraph the idiotic Tory govt has put out a consultative document to consider ways of taxing farm produce and especially meat that involves high CO2 emissions in production.
    The idiot Gove is fingered as the probable main proponent of this nonsense.
    If this takes off then The Brexit Party will be the the party of choice for almost everyone that I know irrespective of the outcome on 31st October

  4. markl permalink
    August 19, 2019 3:36 am

    It will get worse before it gets better. Here in California I bought some wood to fix a gate and was charged a “wood tax” as stated on the receipt. When I inquired about it the answer was “I don’t know, we also have a “paint tax” that I can’t explain”. Bottom line being you will pay for whatever they want to indulge in their largess.

  5. Graeme No.3 permalink
    August 19, 2019 5:36 am

    I don’t know why they have to pay wind turbines to shut down, they do it often enough off their own bat.

  6. August 19, 2019 7:04 am

    The thing is that it has been known for years that there will be two main problems arising from all this asynchronous, intermittent and uncontrollable generation from wind and solar:
    1 In summer when wind and solar can approach and even exceed demand.
    2 On weekdays in winter when during calm, cold mornings and evenings there is not enough firm capacity to meet the peak demand.
    The cost to consumers to overcome these problems (such as constraint payments and use of “smart” meters) will become increasingly severe.

    Heads should roll in many organisations for the way those in charge have ignored the warning signs and have not spoken out. Organsations such as REF and the GWPF that hve spoken out have been totally ignored because of the influence of the Greenblob in the media and in influential NGOs and charities.

    It won’t be pretty when blackouts, together with civil unrest, become a feature of llfe.

  7. August 19, 2019 7:47 am


    FAMILIES fled in terror as an electricity pylon exploded into flames during the Didcot power station demolition – leaving 40,000 homes without power.

    • dennisambler permalink
      August 19, 2019 12:07 pm

      State vandalism, blowing up coal plants so they can’t be brought back in.

    • It doesn't add up... permalink
      August 19, 2019 1:35 pm

      The report on the cause of this will be interesting. If it was just a drone providing a short circuit route between different phases on the transmission line, it suggests a novel and dangerous way to bring the grid to its knees via acts of terrorism. Knocking out key links can overload others, and lead to a cascading blackout.

  8. August 19, 2019 8:32 am

    Reblogged this on ajmarciniak.

  9. August 19, 2019 9:09 am

    Energy storage e.g. batteries would provide a buffer between wind power supply and demand. How much of it could be installed and operated for a lower cost than the existing constraint payments is anyone’s guess, but buffering looks a more economic option than shutdowns – in theory at least.

    • August 19, 2019 9:14 am

      Hence the virtual power plant.

    • Ivan permalink
      August 19, 2019 11:28 am

      Reported in today’s Times is that National Grid is also constraining off interconnectors, when inertia is low, typically from 1GW down to about 0.7GW (the French one is treated as 2 x 1GW). They reckon it is cheaper to do that and pay everyone compensation than to build extra battery storage to be able to cope with a 1GW trip at low inertia. This is not going to work very well in a high renewables future when we rely on many more interconnectors for reliability.

      The 2.2GW Western HVDC undersea link (W of Glasgow to W of Liverpool) is supposed to substantialy remove the transmission constraint from Scotland, although there can still be material local constraints within Scotland when it is windy.

      They consider cost in deciding who to constrain off – more expensive subsidy means more expensive to constrain off. This is part of why off-shore wind farms are relatively rarely constrained off.

      • It doesn't add up... permalink
        August 19, 2019 11:48 am

        That rather makes a nonsense of the plans to build out to 8.8GW of interconnectors to France alone. Connections to Norway, Denmark and Germany are all predicated on helping wind to dump surplus production – but I don’t think they did their sums properly.

        Wind is far too highly correlated across Europe.

  10. Corrupt de Lion permalink
    August 19, 2019 9:35 am

    I mistrust the smart meter programme. Haven’t seen a terminal yet but assume it sits in the kitchen and when demand is high and supply low the price is hiked, it shows on the meter, people say ‘blimey’ and switch off the life saving 2KW electric fire. It’s called Demand Management.

  11. Athelstan. permalink
    August 19, 2019 10:44 am

    Paul said:

    “If they had not constrained a dollop of wind power, the grid would have been far too reliant on the wholly unreliable wind power, thus risking another major blackout.”

    So true but that is not how the wimin aka UK gretas see it, now is it?

    UK greta cabbage patches, they are thick enough to believe that whirlygig generated juice will be able to solve all if not lecky ‘ex la Belgique et la belle france n’cest pas? , including carpeting the sea and land with birdmincers, plugin electric car points and all the bloomin rest, moonbeam tech for the avatar generation snowflake and at this rate we will be all back in the green and blue fooking jungle, soon enough.

    Have we all gone stark staring bonkers?

    • A C Osborn permalink
      August 19, 2019 12:48 pm

      No, not “we”, but “they” have.

  12. DevonCamel permalink
    August 19, 2019 11:08 am

    When most of mainstream politics and media have unwavering belief in something, money usually follows. The obsession with CO2 will continue unless or until the lights go out more regularly.

  13. It doesn't add up... permalink
    August 19, 2019 11:32 am

    In November 2017, Roger Andrews made an excellent post at Euan Mearns’ site on the topic of accommodating wind generation into the UK grid. It can be found here:

    It rapidly debunks the idea that storage could be used to solve the problems of wind intermittency at anything approaching an economic cost, and concludes that flexible CCGT generation offers the best way to balance variable wind output, but even that would be costly and entail a lot of curtailment. I made a contribution in comments inspired by his post:

    If we imagine a grid comprised of just wind and CCGT, then if the wind capacity is constrained to minimum demand (i.e. 19GW or what we normally think of as baseload), it can always run without being curtailed (transmission constraints being conveniently assumed away, with grid stability assumed provided by batteries – we can always amend the assumptions later to incorporate e.g. nuclear baseload or a requirement for a minimum mechanical inertia relative to demand). CCGT would then gain the balance of the market, while having to provide essentially the full maximum capacity (i.e.about 53GW on present demand patterns), since it is very likely that wind would be insignificant at that time, as this example demonstrates:.

    To this basic grid design we can always add more wind. As we do so, it will become necessary to store or curtail wind, but we will be unable to retire CCGT capacity at all. Adding wind will thus at best save fuel cost of CCGT, less added maintenance, while also adding in transmission costs, since wind resource is usually remote from centres of demand.

    I looked at 2016 on the basis of all CCGT, and then progressively adding in chunks of Danish wind. At zero wind, average CCGT utilisation is 61% assuming 53GW of available capacity (discount if you will for maintenance and breakdown by assuming say 60GW in toto). If the wind is allowed to meet anything up to each hour’s demand, there is no curtailment until 23.3GW producing 86.7TWh over the year, but CCGT utilisation has fallen to 42%. Beyond that, curtailment increases quadratically, assuming no storage, and CCGT utilisation falls slightly more slowly. Storage would of course see CCGT utilisation continue to fall at faster rates, while eating into curtailment.

    These results can be summarised in the following chart:

    From this it should be clear that diminishing returns to adding wind capacity because of increasing curtailment at the margin should limit the build. Likewise there is a clear relationship between wind capacity and CCGT utilisation. If CCGT can earn a sufficient margin on the basis of 60% utilisation, then a fall to 30% would require payment of half the annual capital cost, plus additional compensation to cover wear and tear and extra maintenance and fuel from the thrashing regime of operation as a first approximation. Complications arise when you start factoring actual revenues from operations, since CCGT would be likely to secure all the best revenue hours. (It is a coincidence that the 3 lines appear to meet at a point in the chart, arising from the axes used)

    A sample summary of daily generation and curtailment is in this chart for 35GW of nominal wind:

    Note that these are daily totals, so curtailment mainly occurs overnight when demand is low. The calculations are based on hourly data. Note too that the occasions when curtailment occurs are fairly sporadic, making it hard for storage to generate sufficient stock turns per year to earn a margin from this business. There are 129 days when curtailment occurs, but only 72 where it exceeds 15GWh, which would have to suffice to justify incremental capacity.

    I didn’t re-do the sums with nuclear baseload or minimum inertia factored in, but the effect is to lower the point at which curtailment is necessary even with no transmission constraints, and for a much more rapid rise in the total curtailed energy. A slight offset might be had via the use of interconnectors, but because there is a high degree of correlation across much of Europe with wind speeds, it is far less than many would assume.

    • Ivan permalink
      August 19, 2019 2:39 pm

      I think this is well understood by the energy industry. National Grid does not think large scale storage is how to deal with the intermittency of wind in a high renewables low carbon future. See their Future Energy Scenarios document, where you will find gas generation and carbon capture and storage (CCS) are a large part of their scenarios.

      Also see this on creating “digital inertia” with batteries, using only 10% the thermal capacity required. National Grid is currently saying it is cheaper to curtail than use batteries for inertia. Maybe that changes as wind penetration increases.

      • It doesn't add up... permalink
        August 19, 2019 3:23 pm

        The only way that batteries become the more economic way of doing things is if their costs fall dramatically. If you have to have generation capacity anyway because wind can fall to nothing you may as well use it in providing inertia (especially since it is so much cheaper than these renewables on a full cost basis before green taxes and subsidies). National Grid has some 200MW of batteries that they contracted for under their experimental EFR programme. I think that practical experience has shown that this is not as useful as they had hoped: the programme now seems to be firmly on the back burner. Indeed, I’ve not noticed them even mention the role of batteries in the power cut, although I have drawn attention to them several times.

        I did note that this outfit

        were quick to promote the potential for more pumped storage in Scotland at the expense of churning up Loch Ness and other sites.

Comments are closed.

%d bloggers like this: