The UK government has launched a major consultation on its digital identity scheme without providing a single estimate of what it will cost taxpayers. In a consultation document released yesterday, the Cabinet Office said determining the cost was impossible at this stage because policy decisions that would materially impact the costs involved had not yet been made.
This represents a striking abdication of fiscal accountability. In December, the Office for Budget Responsibility estimated the scheme would cost £1.8 billion over three years, flagging it as unfunded. The government promptly rejected that figure, but has offered nothing to replace it. When pressed in Parliament yesterday on the cost, the Chief Secretary to the Prime Minister, Darren Jones, responded that ministers "do not know the answer" and preferred honesty to "announcing a budget for something that then gets massively out of control." The sentiment is fair; the execution is not. Refusing to estimate costs is not fiscal prudence; it is fiscal evasion.
The consultation itself reflects how far the scheme has retreated from the government's initial ambitions. Prime Minister Keir Starmer announced mandatory digital ID in September, with employees unable to work without one. That generated fierce resistance, including three million people signing a parliamentary petition against the scheme. In January, the government backed down, making the digital ID voluntary whilst keeping digital right-to-work checks themselves compulsory. Employers can still accept passports or eVisas instead.
The government is now packaging this retrenchment as enlightened design. The new framing emphasises that digital ID should be "something people will want to get, rather than something they must have." The system will store minimal data: name, date of birth, nationality, a facial image, and a unique identifier. Biological sex and gender will be excluded, as will most addresses. The system will build on GOV.UK One Login, which already serves 122 public services.
There are legitimate arguments for this approach. Voluntary schemes can reduce surveillance concerns compared to mandatory ones. Minimal data collection protects privacy. The private sector has already built digital identity services that work; the government's role could reasonably be limited to ensuring interoperability and access for those excluded from commercial offerings. The consultation process, running until 5 May, followed by a "People's Panel" of 100 to 120 randomly selected citizens, suggests genuine engagement beyond the usual tick-box exercise.
But the scheme still carries real tensions. Civil liberties groups warn that voluntary systems can become mandatory through function creep, particularly as governments identify new uses. Even without central databases, decentralised systems can enable surveillance if unique identifiers link data across platforms. Critics note that function creep is almost inevitable; what starts as right-to-work verification readily expands to benefits access, licence applications, and beyond. Once the infrastructure exists, political pressure to extend it invariably follows.
The government's refusal to cost the scheme publicly is harder to defend. Darren Jones frames the consultation as the necessary precursor to budgeting. This is understandable but incomplete. Parliament has a right to know, before endorsing the principle, what different design choices would cost. Rough orders of magnitude are not impossible; they are merely being withheld. This invites the legitimate suspicion that government wants to move forward without scrutiny of the fiscal case.
The genuine complexity here is real. Digital identity systems offer genuine efficiency gains; Estonia has saved citizens an average of five days a year in paperwork. They also create genuine risks of state overreach. Reasonable people disagree on where to draw the line. What is not reasonable is asking Parliament and the public to support a scheme while refusing to explain its cost. The consultation should proceed. But so should transparent budgeting.