Save content
Have you found this content useful? Use the button above to save it to your profile.
Reform required for digital IDs to be effective | accountingweb

Digital ID scheme will face many hurdles


Recent calls by Tony Blair and William Hague for a UK digital ID scheme have been met by cries of “Big Brother”. The real problem, however, is that the scheme’s claimed benefits are unrealisable, says Bill Mew.

28th Feb 2023
Save content
Have you found this content useful? Use the button above to save it to your profile.

Successive UK governments have championed the idea of “digital government”, with long-standing ambitions to use digital, data and technology (DDaT) to deliver fundamental improvements in the relationships between the citizen and the state. However, the kind of transformation and reinvention of government that would have enabled us to deliver the digital age has never happened. Instead, senior civil servants – champions of cautious, iterative and cost-effective evolution – have prevented any kind of real revolution from ever happening. Consequently, processes in most government departments have remained siloed and largely unmodernised. 

Much of the digitisation that we have actually seen has just been superficial – the addition of a digital interface to existing systems and processes. “Departments have failed to understand the difference between improving what currently exists and real digital transformation, meaning that they have missed opportunities to move to modern, efficient ways of working,” commented the House of Commons Public Accounts Committee.

Siloed and inefficient

This form of cautious digitisation has led to a level of success, with the number of tax returns filed online growing from 38,000 in 2001 to 11m in 2021, but behind the and HMRC websites the core mechanisms of government have remained largely unchanged – siloed and inefficient.

Much of the lack of integration has been down to the competing fiefdoms of large central government departments at one end of the spectrum and the fragmentation of operations across multiple entities in local government and the NHS at the other.

There have been isolated successes where DDaT has been used effectively to drive real change, with the coalition government seeing a notable wave of transformation that secured the UK a brief appearance at the top of the UN’s global league table for e-government. Things have since stalled and the UK has fallen steadily down the rankings.

This growing digital/policy divide is detailed in Fracture, a new book by Jerry Fishenden (@notuk) which explains how a gulf has arisen between the capabilities of the government and the needs and expectations of its citizens. He examines how citizens and government could collaborate to break down existing institutional boundaries, enabling the underlying causes of policy problems to be better understood and tackled. He also advocates sidestepping the stale dogmatic debates of central versus local government to move to a form of governance in which devolution is achieved through an improved national operating model. In this model participation, power, resources and decisions operate at the most relevant, appropriate and accountable level, whether national, regional, local or hyper-local.

Blair and Hague’s recommendations

The Tony Blair/William Hague plans for a national ID card system acknowledge that the success of any such scheme will be dependent on this kind of reform. Their report included more than 40 recommendations, such as:

  • limiting the Treasury’s power to manage science and technology investment
  • appointing “executive ministers” from outside Parliament to rewire Whitehall’s approach to science and technology
  • using AI to help teachers in schools and provide personalised support to pupils at home
  • offering tax breaks to stimulate pension fund investment in UK start-ups.

Both politicians failed, however, to acknowledge that the government is incapable of meeting most of the challenges that they outline because of the bureaucratic malaise. And the malaise persists because of their own failure to enact any real reform or digital transformation during their time in power. Now no longer in power, they are calling on an administration that is beset with significant problems of its own to enact a transformation that they themselves failed to achieve.

Ten big issues

While real reform is an essential precursor for any digital ID scheme, there are 10 big issues that also still need addressing.

  1. Carrot or stick? Are we going to make using the new digital IDs compulsory or optional for citizens, government services and commercial services (like banks) and if optional then what value propositions will do most to encourage their use?
  2. Do we need it? We already have commercial identity systems for payments and so on. Do we really need a separate system in parallel or can we exploit commercial systems that are already in widespread use? The NHS Track and Trace fiasco was an example of the government shunning a more effective commercial approach for ideological reasons.
  3. How pervasive and inclusive is it? If we accept that there’s a real need for a digital ID, then we will also need physical ID cards in parallel, enabled by changes in the law. Driving licences are frequently used, but are currently not actually a legal form of ID, and many citizens don’t even have a passport or driving license anyway. It is not a question of just combining the passport, driving licence and so on into one digital identity. We would need a whole new system that covers everyone, even those without access to technology and those still in their childhood.
  4. Integration The existing government ID scheme called Verify came in parallel with Pay and Notify, all of which were shunned by many government departments and never adopted in the commercial arena. Any replacement needs to be part of a suite of services that provide real value to drive up adoption.
  5. Alignment Do not underestimate the challenge of aligning everything around a single identity, given that we already have so many different unique identifiers in use, from driving licence number, NHS number and passport number, to a UTR with HMRC or national insurance number. These systems already include fraudulent identities as well as duplicates or aspects that don’t align, like out-of-date addresses or names (before and after marriage for example).
  6. Acceptance Any digital ID will require public support and parliamentary approval – previous ID card initiatives have been voted down in Parliament. 
  7. Systems overload Any new scheme will need to replace existing identity systems or they all end up being used in parallel. The current total is 46.
  8. Security An all-pervasive ID will need to be secure (to prevent data breaches) and fraud-free (to prevent fraudulent applications). Current systems don't fare well here.
  9. Privacy There are very real Big Brother concerns, especially if biometrics are incorporated. You can reset a password, but not your DNA, fingerprint or face.
  10. Implementation A scheme of this importance and breadth will need to be implemented flawlessly, not a mess like universal credit or the Track and Trace app.

However, the largest challenge of all will be winning over major government departments, such as HMRC, that have been the main barriers to reform in the past. HMRC in particular has always resisted any identity schemes that cannot be used by both individuals and organisations (as it needs to deal with both). If ID cards, as expected, are only used by individuals then it will inevitably need to retain at least one other identity system for organisations to use and so may well drag its heels on this initiative.

Resistant to change

Unfortunately HMRC is not the only major government department that appears as resistant to such change as ever, even though the ageing and unreformed systems that they operate are struggling to cope with evolving demands. Bureaucratic malaise can have this effect, as HMRC has seen with Making Tax Digital (MTD). Our large central government departments are unfortunately unlikely to welcome the idea of data sharing and collaboration with each other or of ceding power to regional or local government to allow the integration of the tax systems with those that manage benefits or business rates.

Privacy campaigners may well already be complaining about the potential over-reach inherent in returning to a system of identity cards that were abandoned by Churchill 70 years ago. However, the biggest complaint that citizens should actually have is that the promised benefits of digital ID will never be realised as long as vested interests in all corners of government resist any real reform. 

The government has already invested a billion pounds of taxpayers’ money on a series of identity systems, as covered in these recent articles.

  • Government struggles with £1bn identity problem, which looked at the 45 ways of accessing UK government services and the multiple supposedly unique identifiers from driving licence number, NHS number and passport number to UTR with HMRC, and national insurance number.

Before we pour good money after bad into a new digital ID card system (on top of the many identity systems that already exist) we need to focus not only on achieving real reform across government, but also on the 10 issues above.

Replies (6)

Please login or register to join the discussion.

By Hugo Fair
28th Feb 2023 19:47

" HMRC in particular has always resisted any identity schemes that cannot be used by both individuals and organisations (as it needs to deal with both)" ... and they have a point.

I'm not about to break my habit of a lifetime by defending HMRC, but I've sat in interminable meetings with them (hosted by HMT) where they've wrestled with multiple concepts of how to authorise a digital transaction from a body that is anything other than a solitary individual (so not just 'organisations' but partnerships and various hybrid entities).
Those meetings concluded without any progress let alone recommendations on the basis, as far as I recall, that multiple 'unique' ID types was a non-starter ... so a set of rules would need to be constructed in order for verification of an entity to be derived via attributes associated with each individual (who might be a designated officer for instance).
The problem rather obviously remained that these attributes would need to be ... wide ranging (multiple types of entity) + verified at a point in time + maintained via some historical database (as humans have a habit of changing their roles/allegiances to non-human entities).

Without disagreeing with any of the other points made above (mandation vs optional is always a good starting place) ... has some wonk finally proposed a workable solution for unique identities?

Thanks (2)
By dancingbubbles
01st Mar 2023 16:52

Big issue No. 10 : a scheme of this importance and breadth will need enough redundancy and robustness such that it will still function despite the inevitable flaws in its implementation. It's idiotic to think you can make something flawless.

Thanks (0)
By Mr J Andrews
01st Mar 2023 17:51

Why not go along with the Bliar/ Vague recommendations ?We don't seem to have had a new quango for yonks.

Thanks (0)
By moneymanager
02nd Mar 2023 11:23

For "siloed and inefficient", read, more securely contained on a strict "need to know" basis.

If an accidental release of data CAN happen it WILL happen, no Big Brother conspiracy needed for that even if it is as plain as a pikestaff.

Thanks (0)
By NotAnAccountant2
02nd Mar 2023 13:10

Grrr, I could almost cut and paste my answer on the Digital Coin thread to this one.

All of the issues around this have been studied to death. If anyone is interested in a readable but comprehensive introduction (pre blockchain) then this is a good read:

I would describe this book as "not too heavy in maths" but others might disagree.

Any of Schneier's essays are well worth a read too, linked from that site.

As soon as HMRC, BoE, HMG start bringing in consultants linked to large firms, then the agenda will be driven by "how can we monetize this and extract money from the public for our employer" instead of "how can be optimize this to provide best value for the country as a whole."

Sadly, there is probably "one right answer" to the first goal - so it will look "easy" if you listen to those people, while academics with no agenda will bicker endlessly over the (unavoidable) tradeoffs of one implementation over another and how they impact the "greater good".

Thanks (0)
By listerramjet
02nd Mar 2023 14:57

When politicians talk about digital ID they are really talking about control. Digital transformation doesn’t need it. You are right about how Whitehall works though.

Thanks (1)