Professor Sir Dieter Helm gave a barnstorming performance at Policy Exchange recently. He tore, with typical ferocity and fearlessness, into the failures of UK energy policy over the past two decades. It is well worth a watch (or the accompanying paper is here).
I agree with much of his analysis about the shortcomings. It makes uncomfortable viewing for any of us, including me, who have been lucky enough to have had some influence over energy policy over the past decade.
However, there are some important areas where I disagree with Dieter’s diagnosis, and therefore his remedies. I think these differences are important for the future of power market design in the UK, and in particular, the Government’s recently announced and very, very, very welcome, Review of Electricity Market Arrangements (REMA).
Before getting to where I disagree, it is worth stressing the areas where I am in complete agreement with Dieter:
- Need for Government to create limited ‘stakes in the ground’ (to use Greg Clark’s phrase). This means getting serious about a programmatic approach to nuclear. There were very promising signs in the Energy Security Strategy on this which may help tackle the ‘multi-decadal dithering’ which I wrote about here. My long-term hope is you could build a market structure that would allow nuclear to compete with other technologies [see this thread], but I recognise that is a long way off and may never be possible for large-scale nukes. In the short term, we may need a stake for CCUS, including BECCS. But we should try and limit the stakes as much as possible.
- Moving historical costs of renewables into a ‘bad bank’. I think taking the costs of necessary early support for renewables off consumer bills, an incredibly regressive way to pay for them, would help ease horrific pressures on bills [see this excellent Resolution Foundation work] in the short term. It would also help make electricity more attractive as a low carbon heating option (although recent prices spikes may have already made the flip on running costs for good quality heat pump installs vs boilers, see this compelling Jan Rosenow analysis). But they also need to be paid for — in the medium term, once gas prices fall, I would cover some of the revenue shortfall for Government through introducing a carbon tax on gas-use (potentially through including heating in the ETS). The Public First paper is the best I have seen on this.
- Reducing the alphabet soup of Government interventions. Good news for acronym-generating machines and lobbyists, but makes it much harder for many of the new innovators to navigate.
- Need for clear objectives in energy policy. Security of supply (including, now, reducing demand for Russian fossil fuels) and decarbonisation are obviously two. There is another to manage impacts of energy costs on the vulnerable, including industrial consumers. The debate is how much industrial/innovation policy should also be included. A blog for another day, perhaps.
- Independent system operation (again, bravo to the Government for finally taking the plunge on this for the ‘Future System Operator — I should write a blog on how the first attempt to do so was stymied). Still more work to be done on regional system operators. And my favourite bit of the Energy Security Strategy was about the new strategic approach to code governance (why this did not dominate the global headlines that week vs some low-grade scuffle at the Oscars remains an enduring mystery).
- Need for a plan. This is not about central planning, but what we call a ‘Living Roadmap’ to get to decarbonisation (this would be a central part of a systems approach to Net Zero). Net Zero Strategy goes some of the way there, but not enough. By the way, the excellent Net Zero Research and Innovation Framework, which gets nowhere near enough attention, is an important component of that. And I would say the hint towards network ‘investment ahead of need’ in the Energy Security Strategy is very welcome (and mechanisms like Local Area Energy Planning could be a crucial mechanism for delivery of that aim in a way that protects both consumers and investors, and engages local citizens in the transition).
Anyway, enough of the yawn-inducing agreement. Where do I disagree with Dieter and what are the implications, in particular for power market design?
I think the fundamental difference I have with Dieter is about the future of the wholesale market. Dieter’s argument, as I read it, is that the introduction of more and more zero marginal cost technologies (wind, solar, nuclear) mean that the wholesale market diminishes in importance. This logic leads him to one of his key arguments for a ‘firm power auction’, which is central to his power market reform package.
I agree that there is going to be a lot more zero marginal cost power generation on the system. In pretty much any scenario, offshore wind is going to be the workhorse of the future power system. There is uncertainty about nuclear, as other technologies compete (our view is that 10GW of nuclear after Hinkley, a mixture of small and large, is likely low regrets). You can also construct credible scenarios without nuclear or CCUS, but you have to have some pretty heroic assumptions on other technologies. Net Zero is hard.
But just because there is a lot more zero marginal cost on the system does not diminish the importance of the wholesale market. In fact, I would say it could increase its importance. Here is my logic. Challenges welcome.
- All things being equal, if you added more zero marginal cost onto the system, particularly intermittent technologies, you would expect a ‘spikier’ wholesale market. You would have times when there was a lot of wind vs demand, so prices would be low or negative. There would be times of low wind and solar, when prices would be very high as back-up generation recovered its costs.
- And commodity-based technologies are likely to remain important in setting that marginal price, a long way into the future. See some great Aurora analysis that found the even if technologies like hydrogen turbines, CCUS only produced 15% of the energy of the future power system, they would still set the wholesale price 60% of the time.
- Negative pricing is not necessarily a bad thing. It is providing important information to the market (ie now a good time to charge up your EV or pump water to the top of Dinorwig).
- While volatility is important for some consumers, the thing that matters for most consumers for the cost of energy is the area under the curve. While time of use tariffs may see some consumers prepared to manage their energy use on a half-hourly basis, many will prefer the comfort blanket of smoothed out prices (even if that is delivered by them agreeing for their load to be more flexible in the background).
- Smoothing will be an even more important role for suppliers in future markets. Their ability to manage this smoothing through a portfolio of generation assets and demand side response will be a competitive advantage. Vertical integration anyone?
- That lack of consumer demand for volatility does not mean you avoid aligning the market signals with the physics of this future, decentralised system; it makes it more important. This is why nodal pricing could be such a powerful driver of innovation in the energy system (and reduced costs for consumers).
- One of the ways that smart suppliers will manage those risks of volatility is long-term contracts. At the moment, the Government infantalises suppliers to some extent by doing the job for them through CFDs*. This is great for generators, of course, and has been part of the reason for the very low cost of capital and therefore falling cost of renewables. A huge success story. However, it is not clear it is good for consumers in the medium-term, as it stifles the development of technologies with potentially more system value, broadly under the flexibility umbrella*.
- As a result, industry experts tell me there is falling liquidity in forward power markets. This reduces confidence of people who would want to contract many years ahead. We are going in the wrong direction [any data on this much appreciated].
- So what you want to aim for is a wholesale market where participants have enough confidence in the information it contains to build long-term contracts upon it. In simple terms, a supplier looking to manage risk (and perhaps meet an obligation) would contract with a low carbon generator over the long-term at a discount to the forward wholesale curve. They would then top up their energy needs in short-term wholesale markets. The price the consumer would pay would be a combination of these long-term and short-term positions — benefiting from cheap renewables when wind is blowing, and also for the security of the back-up generation when the wind does not blow.
- By the way, this is why we favour a long-term standard/obligation on suppliers to increase the amount of energy they source for zero carbon sources over time. This gives even greater confidence to investors that there is a market for the product of low carbon generation (and yes, I know, it is not as good as a CfD for investors).
So, in my view, the wholesale market becomes both more and less important. It is important for giving a clear signal on the marginal cost of energy at different times, reflecting all the information in a complex market like electricity. This is important for both short-term dispatch decisions and for building long-term contracts (alongside an obligation). So decisions about capacity needs, technology mix, potential of demand-side etc etc are revealed through market processes, not through lobbying.
If you follow the logic — and, as always, I may well have got things wrong — then that leads you away from a solution like the a Firm Power Auction favoured by Dieter, and towards a mix of obligations, wholesale market reform, carbon pricing and enabling conditions. While I agree strongly with the Firm Power Auction’s focus on outcomes, not technologies, I would aim for the same thing through the mix of policy instruments set out above.
One coda to this. Because of the very high gas prices at the moment, there is a lot of chatter about consumers losing out because the marginal price is set by gas and therefore they do not benefit from low cost renewables. There are calls for a bifurcated market, basically one market for renewables and one for commodity-technologies. The very interesting work of Catherine Mitchell and the iGov team at Exeter is probably the most developed exploration of this issue. I have also read some of Michael Grubb’s intriguing work on this. I have not delved into these in detail, so am cautious about passing judgement. But I am profoundly nervous about the idea of moving away from marginal pricing, and to further separating markets.
My starting point is you should try and get as much information into the wholesale markets as possible (including making them more granular), therefore increasing confidence in them and reducing the importance of other markets. The most nail-bite-inducing chart in ESO’s brilliant recent market reform work was the increasing proportion of half-hours that the system operator was having to redispatch more than 50% of the stack. That is partly because information has been sucked out of wholesale markets, which do not yet reflect the physics of the new system.
If you create more confidence in wholesale markets, you should be able to build long term, bilateral contracts off the back of them. This will allow you to do long-term deals between generators and industry (and consumers), of the kind Grubb advocates and appear to have happened in other markets (I would be interested in exactly what ‘facilitation’ by Government of those contracts entails, of course — guarantees?).
The risk of moving toward bifurcated markets or average pricing is, in my mind, that it will lead to a more centrally planned system. As I have said before, that is a perfectly defensible approach and perhaps reasonable with the climate emergency. I think it will make the transition more expensive and stifle innovation, and actually make the whole thing more beholden to political decisions. But if we are going to do it, we should do it properly…
* Rightly, reform of CFDs is going to be central question for the REMA team in BEIS. This is a hugely important issue, which we at ESC will continue to write on, with strong and credible (and occasionally ill-tempered) opinions on all sides. In simple terms it is a cost of capital vs system cost trade off, but there are huge uncertainties and unknowns on both sides. In my view, it is not clear the analytical, market simulation tools to answer the future design question are mature enough yet (although happy to be corrected). The team are thinking at Energy Systems Catapult about how we might develop them…