To say the T-1 Capacity Market (CM) auction clearing at £45/kW per year was a surprise feels like an understatement. It looks like a major windfall for successful participants in the auction – many are asking just what happened?
Until February, it looked like prices would clear at a relatively low price (similar to the £1/kW and £0.77/kW seen in recent years), with 4.2GW of prospective Capacity Providers competing for just 400MW of Capacity Agreements.
However, updated analysis by National Grid issued last month saw the procurement target increased to 2.4GW. This was primarily the result of the capacity expected to be on the system being lower than anticipated, which increased the amount of capacity that needed to be procured to meet system security standards. This included two CCGT power stations (Severn Power and Sutton Bridge, totalling 1,700MW) which unexpectedly entered into a “dormant” state in summer 2020 when its owner Calon Energy went into administration. Furthermore, the 1GW ElecLink interconnector with France also has a CM agreement for next winter but is now not expected to be in operation until mid-2022. Finally, EDF’s nuclear fleet has too faced considerable downtime over the last year.
This increased procurement target meant that the auction was going to be less competitive than previously thought. When we ran the new procurement target in our in-house CM forecast, we noticed that our forecast clearing price became highly dependent on large-scale coal units’ bidding behaviour. (Spoiler alert for later, but we were right).
Dancing in the dark
To understand the pricing dynamics, it’s worth giving a brief overview of how the CM auction process works. The CM is the GB government’s flagship security of supply scheme. It aimed was to encourage investment in new generation and demand-side response capacity by providing Capacity Payments to participants (“Capacity Providers”), which are awarded through auctions held both four years ahead (T-4) and one year ahead (T-1).
Potential Capacity Providers are first required to prequalify for the auction in the summer of the year preceding the auction. After this, they have to compete in the auction itself. The auctions use a descending clock format, which means the price starts high and then falls over time.
The CM is run in a ‘descending clock’ format. This means that the auction starts in round one with a price of £75/kW per year, and then prices fall by £5/kW per year in each round. All participants submit a bid for their Capacity Market Units (CMUs), and if the auction price falls below their bid, the CMU ‘exits’ the auction. This continues until there is only enough capacity left in the auction to meet the procurement target. At this point, the most expensive bid by a CMU needed to meet the capacity procurement target sets the clearing price, and the auction ends. The auction is ‘pay as cleared’, so all participants that are still in the auction at this point then get this price, regardless of the price they actually bid.
Those watching the auction unfold would have seen 200-300MW of capacity exiting in the first round, which is to be expected in most CM auctions as a number of participants make the late decision to not compete and intentionally price themselves out. There was then very little movement until round 6, when a significant amount of capacity exited the auction. This was driven by the bidding behaviour of a small number of large units.
In the past, we often see a noticeable amount of capacity exit when the price falls below £25/kW, which is the “price taker threshold”. This is because most participants that are already in operation (“price takers”) have their bids capped at £25/kW so that prices can cluster around this point.
However, there was only 626MW of new build CMUs in the auction, meaning the auction could meet the procurement target with only existing capacity. As these existing CMUs generally have their bids capped at £25/kW it would therefore be reasonable to assume the price would be at or below the £25/kW mark.
However, in reality, it appears that a number of existing generators likely applied for “price maker” status. By default, all new build generators are given price maker status, so they are able to bid above the £25/kW per year threshold. But existing sites can also apply for price maker status if they can demonstrate that their operational costs are high enough to justify a price above the usual £25/kW per year threshold.
It appears that a number of existing assets likely did this, including the West Burton coal units, and this is why we got such a high clearing price. Despite being existing generator CMUs, the West Burton coal units did not secure agreements at the £45/kW clearing price, and therefore must have bid in higher than this.
While these were not the only units with price maker status, , the West Burton units had a material impact on the auction because of their size. The four unsuccessful West Burton CMUs had a combined capacity of ~1.7GW. This means that after some other CMUs exited in earlier auction rounds, the auction was able to clear when the West Burton units exited. The high clearing price is, therefore, a direct result of these units bidding high.
Land of hope and dreams
So, what have we learnt from this auction? My takeaway is that this result highlights the need for an accurate view of the procurement target at the point of prequalification. As prequalification is held in the summer (mid-July until mid-September) before the auction (usually held in February and March), there will be some variation in capacity procurement targets. The issue, in this case, was that there was a material change in the amount of capacity expected to be on the system, which necessitated a significantly increased procurement target.
We can’t know if the amount of capacity that applied for the T-1 auction back in July and August last year would have been greater if the original procurement target had been higher, but I suspect that the anticipation of a low price due to high competition put potential participants off.
Related to this, this auction also highlights the need for National Grid to have greater notice of generators closing. In this case, we have seen that generator closures can have had a noticeable impact on bidder behaviour. In other jurisdictions (e.g. Ireland), there are greater requirements to give notice of closures.
Hopefully, this blog has given a flavour of how complicated and uncertain the Capacity Market can be. We at Cornwall Insight offer a wide range of services to support participants in navigating these issues, ranging from price forecasts to prequalification support. If you have any questions about the Capacity Market or Cornwall Insight’s services, please get in touch with James on firstname.lastname@example.org.