When ideology overtakes engineering, the public pays the price

Ed Miliband’s department has approved gigawatts of new generation while the cables, substations, supply chain and planning system required to connect it remain years behind. The National Audit Office has exposed a target-first policy whose costs will fall on households, businesses and the British countryside.

Electricity cannot be generated by ministerial announcement. A development consent order is not a power station. A Contract for Difference is not a working turbine. A grid connection agreement is not a live cable. Yet Britain’s energy policy increasingly depends upon presenting these entirely different things as though they were interchangeable.

Ed Miliband inherited an ageing network and a dysfunctional connections queue. He did not create every weakness in the system. But he made the decisive choice to accelerate the political destination without first establishing whether the physical route could be completed. The 2030 deadline came first. The land, engineers, factories, substations, planning decisions and transmission lines were expected to fall obediently into place behind it.

They have not.

The “National Audit Office’s investigation into the electricity transmission programme” (https://www.nao.org.uk/reports/upgrading-the-electricity-transmission-network/) is devastating precisely because of its restrained language. Auditors do not normally accuse governments of allowing ideology to overtake reality. Instead, the NAO records that the grid programme was not founded upon a completed whole-system assessment of the available options; that the upgrades themselves were not supported by an overall business case and value-for-money appraisal; and that many were selected because they were already in development, rather than because the Government had demonstrated that they represented the best solution.

That is ideology translated into administrative procedure: choose the deadline, select what is already on the shelf and compel the physical system to catch up.

Our “project-by-project audit of every substantive energy DCO approved during the last 18 months” (sandbox:/workspace/scratch/6d47a5ae73c0/DESNZ_energy_approvals_true_connection_dates_brief_2026-09-13.pdf) identified 23 approvals between March 2025 and September 2026. Twenty-one were generation projects representing approximately 15.5GW of proposed capacity; the other two were carbon-capture infrastructure. Yet not one of those 23 projects carries a publicly verifiable, firm physical connection date before 2030. Every new generation project that could be reliably matched to the September 2026 transmission register showed zero megawatts connected.

That does not prove that every scheme will fail. It proves something more politically embarrassing: the Government cannot substantiate its own delivery story.

Byers Gill has a contracted connection date in 2031. Fenwick is shown in 2032. Dean Moor is in 2037. Peartree appears to be around 2033. The first or principal stages associated with Five Estuaries, Outer Dowsing, North Falls and Dogger Bank D begin in 2030 or later. Springwell is phased and depends upon the proposed Navenby reinforcement, which has yet to secure all the planning authority required to make that timetable real.

Across Lincolnshire and Yorkshire, the position is not one of a single delayed cable but of interdependent reinforcements, staged connections and constrained substations. In the Creyke Beck area, more than 13GW of proposed generation is competing against existing export capability of less than 7GW. Grimsby-to-Walpole construction extends towards 2033. The principal reinforcements needed to move Scottish renewable power south also reach beyond 2030. Norfolk, Suffolk and Kent face their own delayed or undecided transmission corridors.

Oaklands is the strongest possible objection to this conclusion—and even that example proves the point. Its developer has targeted generation in late 2027 or early 2028, and construction activity has begun. But the Drakelow connection documentation indicates that the new transformer needed for firm capacity is expected in October 2029. Earlier export may be possible only through a technical-limit arrangement subject to curtailment. Oaklands might produce before 2030, but a possible curtailed connection is not the same thing as verified firm export.

The Government’s auction headlines are equally misleading. Allocation Round 6 awarded about 9.65GW, Allocation Round 7 approximately 8.44GW and AR7a another 6.23GW. Some AR6 and AR7a award rows relate to capacity already operating, and the ledgers contain separate project phases. More importantly, a revenue contract does not manufacture a transformer, obtain a local substation consent, reserve a cable vessel or energise a transmission line. It establishes a potential commercial route to market if the physical project can be delivered.

Morgan and Mona demonstrate the gulf between consent and reality. The projects moved through development and obtained consent, yet “EnBW subsequently withdrew from both” (https://www.enbw.com/investors/news-and-publications/enbw-will-not-continue-offshore-wind-projects-mona-and-morgan.html), recording a €1.2 billion impairment. It cited the absence of a CfD alongside supply-chain cost increases, higher interest rates and continuing implementation risks. Other ownership options may still be pursued, but the lesson is unavoidable: Whitehall can approve a wind farm without making it financeable or buildable.

The NAO’s wider findings make the position worse. Of 56 ongoing transmission projects regarded as essential, 50 are forecast to energise later than the date NESO originally considered optimal. Thirty-two are either off track or face significant risks against their agreed delivery dates. Four are already expected after 2030. Only 17 of the wider group of projects examined were actually under construction.

Seven particularly critical reinforcements could add as much as £6.7 billion a year to constraint costs if they are not delivered by 2030. These include Sea Link, Eastern Green Links 1 and 2, Yorkshire Green, Norwich-to-Tilbury and the reinforcements through Brinsworth, Chesterfield and High Marnham. Several remain off track; the remainder carry significant delivery risks.

The consumer pays for this mismatch. Thermal constraint costs increased from £600 million in 2018-19 to £1.9 billion in 2025-26. The NAO reports that they could reach between £6.6 billion and £7.8 billion in 2030 if the required upgrades do not arrive. This is money spent compensating generators, replacing power that cannot be transported and operating around bottlenecks created by the failure to build generation and networks in the correct sequence.

Consumers are then charged for the attempted cure. Ofgem estimates that up to £70 billion may be invested in the transmission network between 2025 and 2031. NESO has identified approximately another £89 billion of possible reinforcement beyond 2030. The transmission component of a typical household bill is forecast to rise from about £44 in 2025-26 to approximately £104 by 2030-31.

The public therefore risks paying three times: first for the contracts used to encourage new generation; secondly for the network required to connect it; and thirdly for the constraint payments and substitute generation required when the first two do not arrive together. Ministers announce the capacity. Consumers finance the consequences.

The countryside also pays. Farmers and communities are being asked to surrender land for pylons, converter stations, substations, construction compounds and cable corridors on the promise of an urgent national requirement. Yet some of the generation those assets are intended to serve is delayed, commercially uncertain or scheduled beyond 2030. The state is exercising the powers of emergency while operating to the discipline of aspiration.

Energy security suffers too. Nameplate renewable capacity is not firm capacity available at the winter peak. Until sufficient networks, storage and dependable generation are physically operating, Britain must retain gas-fired and other dispatchable plant to cover periods of low renewable output and local transmission constraint. A government that boasts about approved gigawatts while neglecting deliverable megawatts risks leaving the country with two systems: an expensive renewable system that cannot always move its power and a conventional system that must remain available to rescue it.

This is why even major energy suppliers and developers are beginning to break ranks. Their frustration is not difficult to understand. They are expected to reserve scarce equipment, employ specialist teams and commit billions against politically fixed dates while grid access, planning decisions and network designs continue to move underneath them.

In August, “EDF called for a strategic review of the Pathway to 2030 transmission programme” (https://www.edfenergy.com/about/an-electric-britain/transmission-costs-need-greater-transparency-and-strategic-review). EDF estimated that construction costs had risen to two or three times the transmission owners’ original forecasts. It said its own generation projects were experiencing average connection delays of two years, with the worst delayed by more than five. It also warned that the public information available was insufficient to determine whether individual projects still offered value for money.

This is not the complaint of an anti-renewable campaign group. It is the intervention of one of Britain’s largest generators and electricity suppliers, arguing that consumers are being asked to fund enormous commitments without adequate evidence, transparency or scrutiny.

Equipment manufacturers and contractors face the same disorder. According to the NAO, some grid-equipment costs have risen to five times their 2022 levels. Global demand means transformer, cable and manufacturing slots must be reserved years in advance. Yet the previous regulatory process often prevented transmission owners from placing firm orders until individual cost approvals had been received. Government created an accelerated deadline while preserving a procurement sequence incapable of meeting it.

Even when workers and equipment are available, access to the live network is becoming a bottleneck. During 2025-26, 15,300 of 23,400 applications for system-access outages had to be cancelled or rearranged. Every cancellation disrupts construction programmes, strands contractors, wastes booked resources and pushes costs further upwards. Suppliers are being asked to price certainty into a system governed by uncertainty.

Meanwhile, accountability dissolves across DESNZ, Ofgem, NESO and the transmission owners. The NAO found no single individual accountable for the complete portfolio. Central oversight teams have limited formal authority and depend heavily upon influence and voluntarily supplied information. DESNZ does not publish a regular, reconciled project-level account showing which schemes are consented, financed, under construction, connected or dependent upon unfinished reinforcement.

That absence of transparency is politically convenient. “Approved” sounds like “built”. “Awarded” sounds like “operating”. “Connection date” sounds firm even when it is conditional upon works that have neither consent nor a secure construction programme. The language allows ministers to bank the headline years before the country receives the electricity.

Britain unquestionably needs to modernise its grid. It needs new generation, better north-south transfer capacity and faster planning decisions. But necessity is not a blank cheque, and urgency is not an excuse to reverse the proper order of engineering.

The Government should publish one auditable delivery register showing, for every major project, its planning status, financial investment decision, supply-chain commitments, enabling works, firm connection capacity, curtailment conditions and realistic first-full-export date. No minister should be permitted to count a project towards 2030 merely because it has received consent or a revenue contract. Where the network dependency is not consented, financed and scheduled, the project should be described honestly as post-2030 or at risk.

Above all, energy policy must return to first principles. Start with the electricity the country will require, including at the winter peak. Establish the firm capacity, fuel security, storage and resilience needed to meet it. Decide where generation can be located at the lowest whole-system cost. Then consent and construct the generation and network as one coordinated programme.

The NAO does not say that clean power is undesirable. It says, in the careful language of public audit, that the Government has not demonstrated that its chosen programme is the best-value route, that delivery is already slipping and that consumers are exposed if it fails.

That is the consequence of allowing ideology to overtake reality. Ministers collect the announcements. Developers collect permissions they may be unable to use. Network companies collect regulated investment. The public receives higher bills, industrial decline, scarred countryside and a system that still requires conventional power to keep the lights on.

The true test of an energy policy is not how many gigawatts Ed Miliband can approve on paper. It is how much dependable electricity reaches British consumers, when they need it, through infrastructure that actually exists, at a price the country can afford.

On that test, the Government’s grand 2030 experiment is not merely behind schedule. It has yet to prove that it was physically coherent in the first place.