Commentators say scale of work was ‘underestimated’ by both sides – but smaller scale pilots offer opportunity to boost collaboration between central and local government
The list of the 19 councils initially chosen to take part in the trials – Photo credit: GDS
More than half of the councils taking part in the local government trial of the Government Digital Service’s flagship identity assurance scheme GOV.UK Verify have dropped out.
The trial aims to assess how councils can link into the identity assurance service, which allows certified companies to check a person’s identity using payment histories, such as mortgages or credit checks, and photographic ID.
It is part of GDS’ wider efforts to expand use of the service – it has pledged to get 25 million people signed up to the service, which went live for central government in May last year, but currently has just 1.2 million registered accounts, by 2020.
The pilots launched last summer after a detailed application process that saw more than 30 councils bid to be part of the work. Nineteen were chosen to test Verify for one or both of two services: applying for a residents’ parking permit and for a travel concession pass.
However, ten of those councils have now left the trial entirely, while two of the six that signed up to both of the pilots have dropped out of one.
There are now just five councils trialling Verify for travel passes, down from 11 at the launch; while there are seven piloting Verify for parking permit applications, down from 14 at the launch – the current figure includes Cambridge Council, which signed up after the initial announcement.
Of the six councils that had signed up to both trials, Hillingdon and Brighton and Hove have left both, while Camden has left just the permits trial and Southampton just the travel pass trial.
Newcastle, Wigan, Barnet, Canterbury and Chelmsford are no longer in the permit trial, while Essex, Luton and Central Bedfordshire are no longer taking part in the travel pass work.
When asked by PublicTechnology for their reasons for leaving the trial, two councils – Hillingdon and Southampton – indicated that they were happy that their existing systems. Southampton said that its existing online verification service for bus passes “provides a similar functionality to the Verify solution”.
Brighton and Hove – which left in February, just before the pilot entered alpha phase – said in a blogpost that it was “a great project but currently the timing isn’t right for us”, as the digital team “has a lot to deliver this year”.
It added that its plans for a virtual permits service “stands to benefit from a tie up with Verify at a later date”.
Camden made a similar point, saying that it had already invested in master data management, which it was looking at “fully integrating into the next phase of Verify”.
Wigan, Chelmsford and Newcastle councils all issued the same statement: “We are not participating in the current phase of the GOV.UK Verify local authority pilots. We remain in contact with GDS on further GOV.UK Verify developments and hope to include the system in local services in the future.”
Steep learning curve
The dropout may also have to do with the amount of work involved in the trials, which is understood to have been a steep learning curve, with both sides underestimating what it would take to get 19 councils to the stage where they can use Verify.
“Councils are used to procuring not building tech. Councils mostly lack the skills to do discovery, work in sprints and collaborate cross border,” said Adam Walther, project director at FutureGov.
“Equally Verify local have probably underestimated the complexity of working with this sector, and lack some of the design, delivery and political skills.
“To move beyond central government to support other parts of the public sector requires more humility, better design and reaching out to partners by everyone involved.”
However, he added that having fewer authorities involved might benefit the pilot in the long-run: “The learning curve for councils can be too great for many of them to be supported by a small Verify team. Start small then scale.”
Matthew Cain was head of digital at Buckinghamshire County Council – which remains enrolled on both trials – when the work began, and he echoed Walther’s comments.
He said that some of the initial requirements and expectations were unrealistic – for instance on the level of tech spend available to councils and in asking for roles that “didn’t even exist in the authority” – but that the pilots had given local government and GDS a crucial opportunity to understand each other.
“Verify is a potentially transformative platform,” Cain said. “But in order to make headway, there needs to be more understanding on both sides.
“GDS are in the foothills of a journey that started five years ago – but in five years’ time the local government landscape will look very different. There’s a limited period of time for local government and GDS to seize those opportunities.”
Lowering the assurance levels
Meanwhile, there are some signs that the Verify local team in GDS are adapting the approach to better suit local government.
Kat Sexton from Cambridgeshire County Council – which joined the trial at a later date – told the Socitm spring conference last week that GDS was working on allowing Verify to offer a lower level of assurance that someone is who they say they are.
She acknowledged that for some central government services – for instance where people are using it to verify their identity in order to be paid benefits by government – it was necessary for there to be a high level of assurance about their identity.
“But for something like a parking permit, do we need to have that level of assurance? Since then, GDS have gone away and…they’re actually creating a lower level of assurance, which is great because we’ll be [keen to use] that,” Sexton said.
The plan to develop a lower level of assurance will also make it easier for some central government services to use Verify – and increase the number of services connected to it, which currently numbers just 12.
For instance, GDS leader Kevin Cunnington has previously said that HMRC’s lack of enthusiasm for Verify is because it works at a higher level of assurance than the tax authority needs, and which is provided by its own identity assurance scheme, Government Gateway.
But getting buy-in from departments like HMRC and the Department for Work and Pensions – which run a large number of citizen-facing services – is essential for Verify to boost user numbers and demonstrate it offers value for money,
The National Audit Office has said that without HMRC users, there would be a net cost of £40m associated with Verify – but the tax authority is developing a replacement for when Government Gateway ends in March 2018, and the apparent competition between the services has been widely criticised.
“Some 138 public services use Government Gateway, and the government needs to decide whether it is backing Gateway or Verify,” said Daniel Thornton, programme director at the Institute for Government. “This is not an area where competition helps.”
Meanwhile, Walther said that HMRC’s refusal to use Verify would not help GDS’s case with local government, as “it gives some in local government an added excuse not to use it”.
However, he added that trying to increase user volume in local government was “tricky because Verify would scale one council at a time” – and that there were more pressing issues that just increasing numbers.
“Verifying parking permits is one thing. Verifying the network around a vulnerable child, or a recipient of Adult Social care support who had Alzheimer’s is quite another,” Walther said.
“This should be about a longer term systemic change about how residents get services in the digital age. Ultimately residents don’t care if central, local or district provide the service. They need to get stuff done. The mindset challenge is bigger than the volumes one.”
The Cabinet Office and GDS were approached for comment on the trials, but a spokesperson said they would not provide a running commentary on the pilot during the pre-election period.