Contracts for Difference (CfD) Allocation Round 7: Subsidy Control Transparency Database estimates

What this is about

DESNZ has published the methodology behind its Subsidy Control Transparency Database entries for CfD Allocation Round 7. The document explains how subsidy estimates were calculated and why they have been rounded — to the nearest £5 billion at the aggregate level and £1 million per project. The rounding itself is the story: it is an official admission that the government cannot estimate the cost of its flagship renewable energy programme to within £5 billion.

AR7 also marks a structural change. Contracts for fixed and floating offshore wind, onshore wind, and solar have been extended from 15 to 20 years. This is not a technical adjustment. It extends the period over which consumers are locked into administered strike prices, suppressing the wholesale price signal for an additional five years per contract. A 20-year CfD awarded in 2026 runs to 2046. That is a generation-length commitment to paying generators an administratively determined price regardless of what the market would otherwise produce.

Key points

The £5 billion rounding band. The total AR7 subsidy estimate has been rounded to the nearest £5 billion. This means the actual estimate could be anywhere within a £5 billion range — and even that range is itself uncertain. DESNZ acknowledges the drivers: future wholesale electricity prices, weather-dependent generation volumes, and bidding behaviour across the market. These are not minor unknowns. They are the core variables that determine cost. Rounding to the nearest £5 billion is not conservative presentation; it is a concession that the estimate is barely meaningful.

Project-level estimates rounded to £1 million. Individual project subsidy estimates are rounded to the nearest £1 million, with DESNZ noting that "uncertainty is greater for project level estimates than for the total." Load factors — how much a wind farm or solar array actually generates relative to its capacity — are singled out as a significant driver. This is correct: a 5 percentage point difference in load factor over 20 years changes the subsidy cost by hundreds of millions for a large offshore wind project.

20-year contracts for the major technologies. AR7 extends contract duration from 15 to 20 years for fixed bottom offshore wind, floating offshore wind, onshore wind, and solar. The rationale is not given in this document, but the effect is clear. A longer contract reduces the generator's financing risk by extending the revenue guarantee. The cost of that reduced risk is borne by consumers, who pay the difference between strike price and reference price for five additional years. In any period where wholesale prices fall significantly below strike prices — as has happened historically — the additional five years represent a direct transfer from bill-payers to generators.

The "offsetting downward pressure" caveat. DESNZ includes a defensive paragraph noting that the subsidy amount "is not the same as the net impact on consumer bills" because renewables place "offsetting downward pressure" on wholesale prices. This is the standard merit-order argument: adding zero-marginal-cost generation pushes down the clearing price. It is true in a static snapshot. It ignores the dynamic costs: curtailment payments when generation exceeds demand, balancing costs when wind drops, transmission reinforcement to connect remote wind farms, and the cannibalisation effect where each additional MW of subsidised renewable generation erodes the market revenue of every other generator — including other renewables. The net bill impact is a different and larger calculation than DESNZ acknowledges here.

Undiscounted, 2024 prices. The estimates are presented without discounting, which inflates the headline figure. A pound paid in 2045 is worth substantially less than a pound paid in 2027. Presenting undiscounted totals makes the subsidy look larger than its present value. This is an odd choice for a department that elsewhere uses discounted figures in its impact assessments. It may reflect the difficulty of choosing a defensible discount rate over a 20-year horizon, or it may simply be the convention for the Subsidy Control Database.

The valuation formula. Estimates use the AR7 Allocation Framework valuation formula, which applies standardised assumptions about load factors, wholesale prices, and generation profiles. This formula was designed for the allocation process itself — determining budget headroom and selecting projects. Using it for transparency reporting is pragmatic but circular: the government is reporting costs based on the same assumptions it used to award the contracts.

What happens next

This is a transparency publication, not a consultation. There is no response mechanism and no policy decision pending. The database entries are now live and will be updated as projects reach financial close, begin generating, or are terminated.

The more consequential question is whether this rounding band becomes a recurring feature. If AR8 follows the same pattern — 20-year contracts, estimates rounded to the nearest £5 billion — then the CfD programme will be committing tens of billions in consumer-funded subsidies with cost estimates that cannot be stated more precisely than the GDP of a small country. At some point, "inherently uncertain" becomes "we do not know what this costs." AR7 is closer to the latter than DESNZ would like to admit.