07.12.20 Big Tech

What now for Downing Street’s data revolution?

Dominic Cummings never realised his vision for a digital government. But the cogs he set in motion are still turning

Since coming into power, Boris Johnson’s government has elevated data as a way to fix the “profound problems at the core of how the British state makes decisions”. But the gap between rhetoric and reality has only widened since the pandemic hit, not helped by a marked refusal to work transparently.

Rushed attempts to “fix” Covid-related problems through data – such as the A-level algorithm debacle which disadvantaged poorer students by standardising grades, and the first version of the contact tracing app – have failed and had to be ditched.

As the government prepared its blueprint for what Whitehall will look like over the next four years, the Bureau teamed up with The Economist’s 1843 magazine to uncover a power-grab happening behind closed doors, the extent to which No 10 is using personal data, and the private companies that stand to benefit.

Back in January, Boris Johnson’s chief aide Dominic Cummings set the tone with his infamous “weirdos and misfits” blog post calling for “unusual” data scientists and software developers to overhaul government decision-making. (Two recruits later left after allegations of racism). Cummings has now departed No 10 – but it’s not clear how much of his vaunted digital transformations could stick.

As the pandemic has continued to dominate British life, threats to shake-up the civil service have ramped up. In June, Cummings warned “a hard rain is coming” because Covid-19 had “exposed fundamental problems in the Whitehall machine and that many officials now accepted the need for radical changes”.

Yet for all the noise from No 10 around data, when it comes to actual details of the mooted reforms, there is surprisingly little public information. Meanwhile, contracts have been awarded to controversial data analytics companies such as Palantir and Faculty AI to create an NHS data store of patient information, without following the usual rules around transparency due to the pandemic. Palantir has since won a post-Brexit border contract, overseeing data from multiple different departments.

Dominic Cummings at a Downing St press conference. Richard Pohle/Pool/AFP via Getty Images

Who will guard the guards?

One troubling development of this government’s approach to data policy is a seeming lack of interest in regulating who controls data.

For example, a leaked document outlining civil service modernisation plans seen by the Bureau states it is interested in “shifting the default operating principle from data protection to data sharing” – indicating it sees those principles as mutually exclusive. Cummings had previously described EU data protection regulations as “a legal and bureaucratic nightmare”.

Questions over governance and oversight will be even harder to answer as a result of the pandemic. Earlier this year, Steve Barclay, chief secretary to the Treasury, said that “a key focus” of the autumn long-term spending review would be “addressing legacy IT and investing in the data infrastructure we need to become a truly digital government”. Again the details of how that might operate remained lacking – and the spending review was subsequently abandoned for a one-year look ahead focused almost exclusively on protecting jobs from the coronavirus crisis.

Taking a central lead on improving data is not necessarily bad. The Public Accounts Committee, responsible for overseeing government spending, found Whitehall “has barely scratched the surface of what it needs to do so it can use data to deliver joined-up public services and increase efficiency,” in a report at the end of last year.

Overcoming complicated challenges involving a lack of government-wide data standards, ageing IT systems, fragmented leadership, and a civil service culture that does not support sharing data across departments will require “firm pressure from the centre of government," it said. That is arguably necessary for the functioning of a modern state.

But, as the body also notes, it must be an open process to ensure “the upholding of the public’s trust in how government uses their data”.

Gavin Freeguard, head of data and transparency at the Institute for Government, said the secrecy around data usage is a major problem. “Some of the concerning themes are definitely around transparency, governance and accountability – we need better information on what data is being used, by whom, to what end,” he said.

To build a fairer, more democratic society, we all need to know what isn’t working and who is responsible. Our investigations expose failings and wrongdoing so those with power can be held to account. We have more stories like this and you can help us investigate

Find out more

“The A-level fiasco shows the importance of those things around automated or algorithmic decision-making – we need to know what data goes in, how decisions are made and for people to have a right to understand and appeal decisions.”

An obsession with “tech solutionism” – where data is seen to solve any problem rather than starting with the problem first – is another issue, he said. “With key figures at the centre of government enthused about data, with the National Data Strategy open for consultation, with plans for leadership of data and digital government taking shape, there’s a real opportunity to make data work better for the public benefit. But that’s only going to happen if the government is transparent and accountable and brings the public with it.”

Header image: Boris Johnson delivers his May address on the government’s ‘roadmap’ out of lockdown. Credit Hollie Adams/Bloomberg via Getty.

Our reporting on Decision Machines is funded by Open Society Foundations. None of our funders have any influence over the Bureau’s editorial decisions or output.