It’s been nearly six months since the close of the second REMA consultation and more than two years since the first one opened. Even by the standards of market reform, and with an unenviable churn in governments, that’s a long time to wait without any real clarity on the direction that reform is going to take. Here, Regen director Johnny Gowdy gives his view on where REMA may have got to and what to expect next.
Looking back and looking forward
It’s a bit too early to talk about REMA in the past tense. We’re still waiting for a positive outcome and policy direction, but after more than two years of sometimes-intensive engagement, it’s inevitable that people will begin to ask questions about what should happen next.
Over the summer, we were expecting a response to the second REMA consultation and maybe even a-minded-to decision on some of the main REMA options. Of course, that was optimistic, with an election in the middle and a new government to brief and settle in. Instead, the DESNZ REMA team has been working hard to develop a coherent set of policy recommendations. There have also been bilateral discussions with organisations across the sector, several expert panel meetings and a lot of solicited and unsolicited advice flowing in REMA’s direction.
There has also been a lot of public lobbying and cajoling, some of which has been unhelpful. A good time not to be on X!
This makes policymaking very difficult, and I don’t envy the REMA team for trying to disentangle the evidence from the passionately held views and opinions that have been expressed. From Regen’s perspective, we’ve been happy with the level of engagement in the bilateral meetings and do not doubt the level of work and commitment from officials. Still, there is a sense that the precepts of good policymaking have suffered somewhere along the REMA journey. We still have no consensus on a future market design, a fractious debate and significant scepticism around the process, and hence a poor basis for delivering change.
With a new government and a new sense of direction, the hope is that REMA’s many positive elements can be plucked out and moulded into an effective reform programme. Regen has a reasonably clear view of where the priorities should lie. We have set these out in our Progressive Market Reform paper, which inevitably has been described as too long, but (I’m told) gets better from p45 onwards. However, we’d be the first to say that we don’t have all the answers, so the main thing is to return to a constructive reform process and open industry engagement.
Zonal pricing: the gorilla in the jacuzzi
Of course, the big REMA question: what to do about zonal pricing.
We don’t know the answer. Officials could still recommend a zonal pricing solution or, more likely, recommend that the government continues exploring zonal options with a view to a future reform, maybe post-2030. They may already have made a recommendation. Either way, the government will likely sit on this decision for a while longer.
It’s a difficult political decision because completely ruling out zonal pricing will incur the ire of its vocal supporters and, it has to be said, run against the strong support it’s had from the newly launched NESO. On the other hand, making a firm commitment to zonal in the face of widespread industry scepticism and the need to focus on investment and the Clean Power Mission would seem a very odd decision.
Meanwhile, the debate about zonal pricing and locational signals, in general, has been overtaken by a complete shift in how we plan for a future energy system, intend to build network infrastructure and manage network connections. The Clean Power Plan 2030 and longer-term Strategic Spatial Energy Plan (SSEP), plus connection reform that will align the award of grid connection to the overall plan, will largely address the question of locational signals for generation and interconnectors, removing one of the main arguments for a zonal market. We will come back to the SSEP in a future blog.
Regen has made its position on zonal pricing clear. Although it does attempt to address some real market and system issues, we don’t think zonal is the way forward because:
- The benefits claimed are highly challenged, not just by Regen but by a number of independent academics, such as Michael Pollitt, who have reviewed previous studies (the recent LCP Delta study for DESNZ has not been peer-reviewed).
- They are based on hypothetical modelling scenarios that assume an ongoing misalignment between generation and interconnector capacity and grid infrastructure and the very unlikely transfer of value from producers to consumers.
- If we build the network capacity and deliver the projects identified in the Pathway to 2030 and the Beyond 2030 grid investment plan, we should reduce the volume of constraints. The forthcoming Clean Power 2030 Plan and the SSEP are expected to confirm this.
- The cost, complexity and inefficiency of splitting the GB market into zones and then dealing with inter-zonal transfers, PPAs, balancing, transactional cost, risk and hedging have largely been unquantified.
- The fundamental point is that smaller markets could lead to inefficiency, market power and a loss of competition unless heavily centralised and regulated.
- Zonal pricing would have significant distributional impacts between zones, between consumers that are more or less able to respond to price signals and between consumers who are paying to subsidise generation and those that may benefit from lower energy prices in high-generation areas. This will probably kill it.
- A change to the market of this nature is likely to divert resources and increase uncertainty at a time when we need to focus on delivery and investment.
- The system issues that zonal pricing might address can, as discussed below, largely be addressed through other progressive reforms.
The concern that the introduction of zonal pricing would directly impact the delivery of clean power by making projects un-investable has, to some extent, been mitigated as we understand that existing generators would have their connection rights grandfathered and new generators would be protected under the CfD, RAV models or Cap and Floor schemes. There will still, however, be a residual policy and implementation risk, and a concern about how future governments may change or renege on their commitments. So we would still expect to see some increase in the cost of capital and potentially a delay to investment commitments or capital withdrawal. The mitigation measures would, however, reduce the claimed consumer benefit case and create a significant hostage to political fortune in the longer term.
A further very important point is that, even after two years, we don’t actually have an agreed design of how zonal pricing might work in practice. Depending on the discussion, two very different versions of zonal (and perhaps more) are being proposed:
- A Centralised Zonal Locational Marginal Pricing (LMP) design, which sits behind the current modelled benefit case, that would feature a centralised market with a largely centralised dispatch. This design would put NESO (or another body) in the place of the market operator and would take us back to a mandated pool arrangement, albeit driven by a more sophisticated LMP-type dispatch algorithm. Congestion and infra-marginal rents would accrue to the market operator to be used and distributed as policy determined. A lot of value would probably be stuck with traders, forecasters, consultants and hedge providers. This model could then, in theory, be a stepping stone to a form of nodal LMP.
- A Decentralised Zonal Pricing model which would seek to retain the bilateral trading features of the current market, although the extent to which PPAs and other forms of bilateral trading could operate would be limited by the zones, with interzonal trading handled by a form of coupled transmission capacity trading, something like the way EU markets currently trade over interconnectors.
How and whether demand consumers would be exposed to zonal pricing, and the impact this would have on the thousands of distribution-connected assets, has not been clarified. The argument that some Scottish consumers would enjoy a bonanza of cheap energy under zonal pricing would depend on the willingness of consumers in the rest of the country to continue to cross-subsidise generators in Scotland and on the timing of grid investment. In general, the distributional impacts of zonal have not been fully explored.
Most of the proponents of zonal pricing and those who are still neutral, including storage and flexibility providers, would very much prefer a Decentralised Zonal Pricing model. So would we. The case for a return to a central market is extremely weak. It is getting weaker with IT and automation aiding the control room and as we shift away from using large pre-scheduled CCGT plants to manage day-to-day balancing and constraint management.
The challenge, however, is that there is very little detail on how a decentralised zonal market design would work in practise. This raises many questions: If there were 12 or more zones, would they be sustainable/liquid as decentralised trading markets? How would interzonal trading work? Would we be able to support inter-zonal PPAs? What is the design for transmission rights and new hedging instruments? Would market participants have to balance at a zonal level? Would a balancing mechanism still be needed with zonal physical notifications, system prices and balancing charges? At what point and through what mechanism would demand flexibility and storage be able to participate in the market? Where would constraint risk now sit, with generators or the system operator? What about all the smaller distributed generators, as well as flex and storage assets – how would these be impacted? Just pointing to Norway or Italy as examples of zonal markets does not address the question for GB – not if we want a market that is fit for our purpose.
It’s widely agreed that, if we do explore a zonal option, a lot more work needs to be done to establish how the design would work, and this needs to be subject to further industry and stakeholder consultation.
Progressive Market Reform
In part, the difficulty that policymakers and industry now face with taking REMA forward is that the case for change was not properly and critically examined at the outset.
The 2020 Energy White Paper posed the question: How would the market operate with high renewables, price cannibalisation and periods of excess energy? When the marginal price no longer represented the ‘true’ cost of energy. Given this brief, REMA should have focused on how to ensure that the market(s) is agile and competitive to pass the value of low-cost energy to the consumer and optimise the use of renewables, balanced with a need to increase investment. But REMA started in 2022, mid-energy crisis. Politics shifted and it was easier to just say the market was ‘broken’. We also had a growing problem with constraint cost (due to delayed grid upgrades and a rise in balancing prices from CCGT plants), which grabbed the agenda.
Analysis was stultified and, before understanding the problem or opportunity, we jumped to poorly defined options and a poorly informed debate about market structures. Regen highlighted this in our first REMA consultation response with a recommendation that more work was needed to understand how the current market actually works, its strengths and weaknesses, and to build the case for reform.
In Regen’s Progressive Market Reform paper, we propose the hypothesis that the current market is not fundamentally broken. In fact, our decentralised and liberalised energy markets have helped the UK achieve an impressive level of renewable energy, especially the build-out of many decentralised and small-scale generation projects. We also have the second most active PPA market in Europe.
Rising constraint costs, which have been mischaracterised as a problem of building generation in the wrong place, are more accurately described as a problem of failing to build network capacity and then relying on expensive and unresponsive CCGT plants to provide balancing services.
We think there are a range of market reforms that could be (and in some cases are already being) introduced to address the areas where the market is sub-optimal. For example:
- Constraints volumes should fall to an economically efficient level with better spatial planning and grid infrastructure delivery. Or at least the decision around an acceptable level of constraint is made by policy informed by the system plan.
- Constraint costs should then fall further with a far more open and competitive balancing mechanism, and with the addition of local constraint markets and flexibility contracts to mobilise flexibility providers. By 2030, in a clean power system, we should not be using CCGT plants for day-to-day balancing and constraint management.
- Interconnector flows and inefficiencies can be better managed by, in the first instance, aligning interconnector build within the overall strategic plan, harmonising cross-border
- Trading and then giving the system operator better tools and processes to manage flows when the GB market is constrained.
For more ideas about how we can reform our existing market arrangements, please see our paper Progressive Market Reform for a Clean Power System.
An area that still needs to be explored is whether our current trading market plus balancing mechanism is sufficiently agile and liquid to ensure that we can optimise the use of low-carbon energy when it’s in abundance. That was the original brief of the 2020 Energy White Paper and is the question that REMA still needs to address.