Contracts for difference are a small part of the wholesale market now, but soon they will be affecting market behaviour, argues Mark Meyrick
Under the government’s Electricity Market Reform, its replacement for the Renewable Obligation is now upon us. The CfD, or Contract for Difference, now means there is about 6.4GW of power coming onto the grid, mostly between now and 2021. This is likely to have a profound effect on the wholesale market.
Consider how the contracts are designed: owners of a CfD only have to make sure they are dispatched via the day ahead auction to win their CfD price. This will influence their behaviour in two ways.
As merchant plant under the old Renewables Obligation system – if eligible – they would to some extent have sold power opportunistically in the forward market to ensure they captured a worthwhile sale price. Now they no longer have to do that – they simply need to ensure they get their CfD price. This will have a distorting effect on the wholesale market because there is now a whole slab of power with no interest in selling forward. Yet suppliers have to buy forward to hedge their customer positions, even if it’s only chunks of forward base and peak products. Result? We get an asymmetric market with a lot of suppliers looking to buy forward power, but a lot of generators only interested in selling day ahead. With too little demand chasing too little supply along the curve, the forward curve will steepen – go into contango – and more aggressively so as more CfD plant comes on to the system.
In the short-end, by contrast, CfD plant has to get scheduled – so it will come into the auction at zero, or lower, knowing it will get the cleared price anyway, but more pertinently, get its CfD price.
With too little demand chasing too little supply along the curve, the forward curve will go into contango
Given the volumes of CfD plant this raises the distinct possibility of auction prices actually out-turning at zero, something the Department for Business, Energy and Industrial Strategy (BEIS) eventually realised. So it revised the CfD rules for new CfD contracts from April 2016 such that if the day-ahead price out-turns as zero for six hours in succession, the CfD price will not apply, the difference payment will be set at zero and affected plant will get paid per auction clearing price – possibly negative. Existing CfDs will not be amended, so generators with these contracts will receive difference payments, capped at their strike price, during periods of negative prices, regardless of the length of those periods.
This does mean that second round CfD plant will have to be a little more intelligent when offering into the market.
We won’t know how much plant this will affect until the round two out-turn is known in September, but it does mean that round two CfD owners will have to try to forecast day-ahead prices and offer into the auction accordingly – at a positive price. If all participants do this, it should avoid a deluge of zero-priced offers and the possibility of negative pricing – thus preserving the value of the CfD contract. But 8GW of plant – if we include Hinkley Point C, if it ever gets built – is not an insignificant amount of plant, and it is not affected by the new provisions.
The further implication of this is that it’s likely to have a depressing effect on the short-term market, exacerbating the contango, as well as unhinging the merit order effect, particularly in relation to plant that does have a marginal cost, such as biomass. Marginal cost issues that normally underpin the merit order effect will be overpowered by the marginal revenue.
In any case, the merit order effect may be dislocated by the grid having to constrain renewables and run fossil plant for system stability reasons. But that’s another story.
First published in the July 2017 issue of New Power
Further reading:
OpenLink: Get ready to manage fast-response trading (subscriber-only)
Secure and promote: Is the liquidity fix in good shape for the future?
Subscribe to New Power for full analysis, comment, interviews and data in our monthly report, and access to our database, or sign up to our FREE e-newsletter for website updates
Not a subscriber? To see if you qualify for our next FREE TRIAL send your name, job title, and telephone number to [email protected]
I think the basis of your argument is based on a number of statements which I would not consider to be correct and hence the conclusions which I am less knowledgeable on may also be incorrect. Firstly CfD generators are not guaranteed their strike price. They are guaranteed a strike price – Market Reference Price. For Base Load generators the market reference price is based on seasonal forward prices so if a generator wants to ensure their revenue they are motivated to sell ahead so as to match the market reference price. For intermittent generators the market reference price is set day ahead so in this case a generator wanting to match the market reference price will be motivated to sell day ahead. This motivation is not driven by the CfD it is a function of intermittent generation. So rather than suggesting that the CfD is the cause of the problem I suggest you look elsewhere. Intermittent generation is a fact of life so as it is now widely recognised we need to find ways to work with it hence the interest in storage and DSR.
Good points Nic, and I admit that I failed to distinguish between the baseload generators and the intermittent generators market reference price. Nevertheless, all the intermittent generation, often hedged forward at the moment, will be dumped in the DA market – and that will have an effect.
As for the baseload generators MRP, well this is nothing more than cloud cuckoo land to think that a 3.2 GW baseload generator will be able sell that sort of volume forward without moving the index that their CFD is being referenced against – thus creating an even larger CFD payment than would be contemplated in a ‘normal’ market. That is because there simply isn’t enough liquidity to absorb those kind of volumes. Currently large generators carefully feed the volume into the market so as not to move the market. EDF doesn’t have to take such care with the CFD in their back pocket – they can just dump it. And this discussion doesn’t begin to address the point as to which index would be an appropriate MRP – that will be quite a challenge.
Hi Mark – To match the Baseload MRP a generator will sell a slice each day so if this causes the MRP to dip then surely after a few days traders will see an arbitrage opportunity of MRP versus outturn (all be it with price change risk) and buy at the MRP. With DA I see where you are coming from but there could be a solution. The CfD has a mechanism to enable the MRPs to be changed if certain conditions are met and there are change controls in the CfD if a change benefits all. So if the case can be made for a combination of long and short term reference prices for the intermittent technologies then the CfD should go that way. Is there another point here though which is that reference prices are struggling to gain traction in the electricity market?