Appearing before MPs recently, the head of the Office for Statistics Regulation warned that departments are not meeting the openness ambitions of their senior leaders, particularly for policy impact evaluations
Government departments’ behaviour on data transparency fails to match the aspirations of permanent secretaries and other senior leaders, Office for Statistics Regulation chief Ed Humpherson has told MPs.
Humpherson told a session of parliament’s Public Administration and Constitutional Affairs Committee that it was a “simple hygiene thing” for departments to make supporting information about policy announcements available to MPs, the media and the public. He said there was a risk to the scrutiny of policy development if government data could not be openly assessed.
“It really shouldn’t happen that there’s some really good analysis that supports a policy statement [but] the policy statement comes out with numbers and you as a member of parliament, or the media, or a member of the public can’t find the numbers,” he said. “The thing that, I think, I am concerned about is that whenever we talk to senior officials about this, we call it intelligent transparency, we get sign-up.”
Humpherson added: “Senior officials, permanent secretaries absolutely endorse and recognise it. What we then find is cases where, in a specific moment, a particular department concludes that it’s not in its communications interests, or they forget to make the underlying data available. So, I would really want to see those commitments that we hear from senior officials being much more publicly made, and much more embedded into their practices and processes.”
Committee member Ronnie Cowan questioned the credibility of any claims that a department had simply forgotten to publish a particular dataset.
He told Humpherson that PACAC’s recent investigation into the future of the government estate had found the Cabinet Office had made several claims about the economic benefits of relocating civil service roles from London to regional hubs without publishing the underlying research to support its estimates.
Cowan also referenced charity Full Fact’s concerns about Home Office ministers citing unpublished operational data that could not be reconciled with published official statistics.
He asked Humpherson whether OSR guidance from 2022 encouraging departments to proactively publish data that was being cited by ministers had achieved its aims.
The OSR head said he believed the guidance had been successful “in part”, but acknowledged that the Cabinet Office and Home Office examples were “disappointing”.
- Met Police creates head of data rights post to ‘ensure transparency and build trust’
- Transparency: Court rules that junior officials’ names should not be redacted from public legal challenges
- MPs urge better use of people data to create ‘transparency and efficiency improvements within government’
“I hadn’t been aware of the Cabinet Office example, and now I am aware… we’ll have a look at that,” he said. “The core principle is that if an assertion is made publicly on the basis of data, those data should be equally available to all.”
Humpherson said the OSR had worked with the Home Office with some success.
“Over time, the analytical leaders in that department have increasingly embedded into the Home Office’s processes the standard expectation that numbers are made available,” he said. “They are, largely speaking, complying with that.”
Impact evaluations ‘a weak area’ for transparency
Humpherson told PACAC members that publication of policy impact evaluations was “one of the weaker areas” for government transparency and an area where the OSR planned to push.
“There will be some evaluations which are done which show that policies haven’t worked as intended, but I think that as long as government is willing to make the bad news available, then people will be confident in the good news when the evaluation says there is good news,” he said.
Humpherson said the Troubled Families Programme, begun under the coalition government, was a project where the late-running publication of evaluation had wrongly prompted negative expectations.
“As a result of that delay in releasing the evaluation, external users suspected that the evaluation had a bad news story – because it wasn’t being released,” he said. “The evaluation was eventually put out into the public domain and it’s a really strong and effective evaluation.”
He said the point was that impact evaluations needed to be “available, accessible and transparent”.