The Gridlock Problem: The Economics of Britain's Connection Queue

Britain is undergoing a rapid expansion of its electricity system, with growing numbers of generation, storage and industrial projects seeking access to the grid. But connecting them has become increasingly difficult as available capacity comes under pressure.

On 17th September, Ofgem proposed a new commitment fee for battery projects seeking a place on Britain's electricity grid. Around 90GW of battery capacity is either operating or in the reformed connections queue, compared with roughly 29GW of projected need by 2035.¹

Battery capacity operating or queued is around three times projected 2035 need

The figures raise a question that goes beyond batteries. Why are so many projects seeking access to the same scarce network capacity?

Under the old connections system, a place in the queue was relatively cheap to hold. Developers could apply for capacity without having every part of a project ready, including having secured land rights or planning permission.² Holding capacity was therefore an attractive, low-risk speculative option, even when the probability of actually building the project remained highly uncertain. By February 2025, the grid connection queue had exceeded 750GW, and new applicants faced an average wait of almost six years to connect.³ Even after reform, half of the battery projects in the queue had posted less than £800 per MW in securities.⁴ Ofgem is now trying to change that incentive. The battery proposal follows a commitment fee proposed for large data centres and the earlier Progression Commitment Fee for projects that have passed the Gate 2 readiness requirements. Their designs differ, but the principle is similar: developers must demonstrate readiness and, increasingly, put money at risk to retain access to the queue.

The challenge for a regulator is to distinguish projects that are likely to materialise from demand that is cheap to reserve but unlikely to. The answer may depend on how much a developer is required to risk, and whether that risk measures commitment, financial capacity, or both.

A relatively cheap option on scarce capacity

The queue matters because connections are scarce. New transmission and distribution infrastructure takes time to build, while demand for connections can arrive much faster. What matters is not simply how much generation or demand Britain needs, but who gets access to the network, and when.

Economically, holding a queue position resembles an option: a right, but not an obligation, to connect. The developer obtains a potentially valuable claim on future network capacity while committing relatively little capital upfront. If the project becomes viable, the queue position is valuable. If circumstances change, the developer can abandon it. The upside of retaining a favourable connection can be significant, while the cost of holding the position may be limited.

This does not mean projects entering the queue are necessarily speculative. Large infrastructure projects are uncertain by nature, and a project can be commercially genuine while still being several years from construction. The problem arises when the incentives of individual developers do not align with the interests of the system as a whole. Each developer has an incentive to secure capacity early because losing a queue position could be costly. Collectively, this can produce a queue substantially larger than the set of projects that will ultimately be built.

The consequence is not simply a long waiting list. A large volume of uncertain projects makes the queue a less informative signal of where network capacity will actually be needed.⁵ If network planners cannot easily distinguish committed projects from those that may later disappear, investment decisions have to be made with less certainty of future demand. The scale of the queue illustrates the challenge: between November 2024 and June 2025, total contracted demand in the connections queue rose from 41GW to 125GW.⁵

Contracted demand in the connections queue tripled in seven months

This is the underlying problem Ofgem's reforms are trying to address. Rather than asking developers to prove that a project will definitely be built, which no regulator can know in advance, the system can make holding scarce capacity more costly. The cost of retaining a place should change the calculation for projects whose commitment is weak, while remaining manageable for projects that are sufficiently likely to proceed.

Proof versus pledge

A regulator can test whether a project is credible in two different ways. The first is to ask for evidence: has the developer secured an end-user, ordered long-lead equipment, and demonstrated the financial and technical capability to build? Ofgem's proposed data-centre milestones do exactly this. The second is to make the developer put money at risk. If a project is abandoned under the relevant conditions, the developer can lose the security; if it reaches energisation as required, the proposed data-centre fee is returned.⁶

The economic logic is different. Evidence-based screening asks the regulator to observe whether a project has reached particular stages of development. A financial commitment instead makes the developer reveal something about its own expectations. A project whose prospects are weak has more reason to avoid tying up capital or risking forfeiture than one that expects to proceed.

This resembles a costly signal. Where one party knows more about the quality of an investment than the other, requiring a cost to demonstrate commitment can make it harder for projects with a low likelihood of proceeding to imitate those that are genuinely ready. But the signal is imperfect. A developer with deep pockets may be willing to pay even when its project remains uncertain, while a genuinely viable project with tighter financing may struggle to absorb the cost. The fee can therefore distinguish projects partly by their willingness and ability to bear the financial cost, rather than by their underlying likelihood of proceeding.

This creates a further risk. If financial commitments become too large, access to scarce grid capacity may increasingly favour developers with stronger balance sheets rather than those with the most viable projects. Smaller or more financing-constrained developers could therefore be screened out even where their projects are capable of proceeding, potentially concentrating the queue among a smaller group of well-capitalised firms.

Ofgem is not choosing between screening and financial commitment. Its data-centre proposals use both: developers would have to provide evidence of progress at specified milestones while also securing the commitment fee. The two mechanisms test different things: whether the project is actually progressing, and how much the developer is willing to risk to retain its place.

Three fees, three triggers

Ofgem's three commitment mechanisms are aimed at different versions of the same problem.

Of the three mechanisms, the PCF most directly addresses the weakness of the old queue. Rather than requiring the network to determine which projects are credible, a rising cost of retaining capacity gives projects that are no longer progressing an incentive to withdraw. Its relatively modest starting fee also limits the risk of otherwise viable projects being excluded simply because they have less access to capital. It has not yet been activated, however: when NESO first published its activation metric on 3rd August 2026, it stood at 0MW.⁸

The data-centre proposal creates a much stronger financial signal. This may be useful where individual projects can place very large demands on the network, but the size of the commitment also exposes the weakness of using finance as a proxy for readiness. The mechanism therefore risks testing balance-sheet strength alongside genuine commitment.

The battery OTCF takes a different approach because its trigger depends on oversubscription across an entire technology. This allows the mechanism to respond as pressure on the queue changes, but oversubscription itself says little about the viability of an individual battery project. A project does not become less credible simply because many similar projects have applied for connections. The fee may discourage speculative applications when the queue greatly exceeds expected system need, but it is a relatively blunt way of distinguishing between projects within that queue.

None of these mechanisms creates a conventional market for grid capacity: developers cannot simply outbid one another for a connection. Instead, Ofgem is adding financial signals to an administrative allocation system. Their success therefore depends on whether those signals make the queue better reflect which projects are actually likely to proceed. If they instead primarily reveal which developers can most easily bear the cost, the information problem has not disappeared; the queue is simply being filtered by a different criterion.

How much is a promise worth?

If the purpose of a commitment fee is to discourage developers from holding capacity they are unlikely to use, its effectiveness depends partly on its size. A trivial security may do little to change behaviour. A very large one may deter projects for reasons unrelated to whether they are genuinely ready to connect.

Ofgem's proposed data-centre fee makes the trade-off unusually visible. The proposed security of £237,500 to £712,500 per MW is equivalent to around 2.5% to 7.5% of the regulator's benchmark average project cost.⁹ For a 40MW project, that implies £9.5m to £28.5m of financial security. That is large enough to change the economics of entering or remaining in the queue. But the relevant cost is not necessarily the full amount forfeited. If the project reaches energisation under the rules, the security is returned. The burden comes instead from the opportunity cost of tying up capital, arranging a guarantee or letter of credit, and accepting the risk of eventual forfeiture.

The challenge is one of mechanism design. If the evidence requirements are too weak, or the commitment too small, the queue stays crowded with projects that have little chance of reaching construction. If the commitment is too high, the regulator ends up screening not only for commitment but for access to capital. There is no guarantee that the projects screened out will be those least likely to proceed. The optimal security is therefore not the one that removes the most projects, but the one that improves the information contained in the queue.

This shifts the issue from the cost of a grid connection to the cost of retaining a claim on scarce future capacity when the project behind that claim remains uncertain. And the network still has to be built: a better queue can improve allocation, but it cannot create substations, transmission lines or connection capacity that do not exist.

A more informed queue

Britain's grid queue is not simply a problem of too many projects. It is a problem of information. Developers know more about the prospects of their own projects than the regulator does, while a queue position can have value long before a project is ready to build. Ofgem's response seeks to change the incentives around holding a place in the queue. Progression requirements provide evidence of readiness; commitment fees make developers put something at risk.

On balance, the reforms are likely to improve the queue. Requiring projects to demonstrate readiness and put capital at risk should make speculative or poorly developed applications less attractive, leaving the network with better information about which projects are likely to proceed. But financial commitment should complement, rather than substitute for, evidence of project readiness. Otherwise, Ofgem risks replacing a queue that was too easy to enter with one that disproportionately favours developers best able to finance a place in it.

Ofgem should therefore place greater weight on observable progress, such as meeting development milestones, and use commitment fees as a supporting signal rather than the principal test of credibility. The aim should not be to make queue positions expensive enough to eliminate excess applications, but to make them conditional enough that retaining one reveals useful information. That would not solve Britain's shortage of physical grid capacity, but it would give the network a better basis for deciding which projects that capacity should serve.

Footnotes

1. Ofgem, Ofgem moves to free up grid capacity by tackling excess battery projects (opens in a new tab), 17th September 2026.

2. Ofgem, Summary Decision Document: TMO4+ Connections Reform Proposals (opens in a new tab), 15th April 2025.

3. Ofgem, State of the market: Energy Infrastructure Highlights January 2026 (opens in a new tab), 27th January 2026.

4. Ofgem, Minded-to Consultation: Connection and Use of System Code (CUSC) CMP470: Introducing an Oversubscribed Technologies Commitment Fee (opens in a new tab), 17th September 2026. ²

5. Ofgem, Demand connections update (opens in a new tab), 6th November 2025. ²

6. Ofgem, Curate: Demand Connections Reform (opens in a new tab), 29th July 2026. ²

7. Ofgem, Decision: Connection and Use of System Code (CUSC) CMP448: Introducing a progression commitment fee to the Gate 2 connections queue (opens in a new tab), 8th December 2025.

8. National Energy System Operator, Progression Commitment Fee (PCF) (opens in a new tab), 3rd August 2026.

9. Ofgem, Ofgem acts to free up grid capacity by tackling speculative data centre projects (opens in a new tab), 29th July 2026.