top of page
Search

Data centres are about to relearn the 2014 Capacity Market's hardest lessons - the expensive way

chris16485
7 hours ago
3 min read
Converting the Capacity Market Reform into Data Centre Reform
Converting the Capacity Market Reform into Data Centre Reform

The UK Capacity Market has been running since 2014. It has verified emissions declarations, audited technical claims, and a decade of enforcement precedent behind it. It has stiff penalties for underperformance, because the lights might go out in the middle of winter. Ofgem's new data centre connection reform - Curate - asks developers to self-certify with a director's letter and nothing else. That gap should worry anyone about to sign a commitment fee.


The verification gap

Under the Capacity Market, a false declaration is a termination event. Emissions data goes through one of four Independent Emissions Verifiers. Project spend is certified by an Independent Technical Expert. Todays proposed Curate has none of this. A data centre developer self-declares status, references a definition buried in the Cyber Security and Resilience Bill, and that's it - no named verifier, no disclosed audit right, no published penalty for getting it wrong. Ofgem is asking a market to behave responsibly using a governance model lighter than the one it already runs next door.


A better fee reference point exists

Ofgem's proposed commitment fee - proposed at 2.5–7.5% of capex, or roughly £237,500 to £712,500 per MW - is a blunt instrument. A flat percentage doesn't distinguish a marginal project from a bankable one.


On the other hand, the Capacity Market's User Commitment Methodology does exactly that: it prices commitment against actual risk profile, not a fixed multiple of spend. If Ofgem wants the data centre fee to filter genuine investment from speculation rather than just taxing everyone equally, this is the more credible design reference - and it already has a decade of operational history to draw on.


Two open questions nobody's addressing yet

1️⃣what happens when a government strategic-project designation reallocates grid capacity after a neighbouring developer has already paid a six or seven-figure commitment fee to secure their project? No published Ofgem or DESNZ material addresses this conflict. It's not hypothetical - it's the direct consequence of running a national strategic-demand mechanism alongside a paid queue.

2️⃣ there's no confirmed timing yet for a National Data Centre Strategy that would resolve this, and the Flex Technical Taskforce's findings - due to the AI Energy Council this autumn - remain unpublished. Developers are being asked to commit capital now, against a policy framework still being written.


The curtailment % you're being handed is not neutral

Non-firm and ramped connections carry a curtailment forecast from NESO or NGET. In our experience, those forecasts skew (sometimes very) conservative - network operators have every incentive to over-predict curtailment hours and none to under-predict them, given the compensation exposure and embarrassment if they get it wrong. No published methodology or track record lets developers check this. The fix is straightforward: commission independent third-party grid dynamic modelling before accepting a non-firm offer, rather than pricing risk off a single, unverified number.


Where this leaves developers

Curate borrows the Capacity Market's instincts - commitment, milestones, forfeiture - without yet borrowing its safeguards. Until Ofgem closes that gap, the burden of verification sits with developers, not the regulator.


We've written a full briefing covering all fourteen sections of the reform - DCCF mechanics, self-build and independent transmission ownership, statutory powers, N-1/N-2 standards, and the complete picture developers need before committing capital.


 
 
 

Recent Posts

See All

Comments


bottom of page