The government needs to offer much more clarity on how it plans to up departmental data sharing in its Digital Economy Bill, MPs have been told.
Mike Bracken, former GDS boss, said the government uses bulk data too much – Photo credit: Dods
The bill, which entered its public bill committee stage yesterday, aims to make it easier for departments to share data with each other, as well as improving access to data for research and the production of national statistics. The government has emphasised that it aims to use this to make public services better and more efficient for citizens.
Although it met with cautious approval from some groups initially, it has come under fire from privacy advocates and opposition MPs, with many people calling for more details, particularly on how government will define the ‘public good’ when deciding whether or not to share data.
These comments were echoed by a number of witnesses at yesterday’s Digital Economy Bill Committee hearing, who asked for more clarity on the section of the bill relating to digital governance, with concerns around privacy, sharing of data and public trust.
Among them was former Government Digital Service chief Mike Bracken, who said that he did not believe there was an assurance that the government had sufficiently considered safeguards on privacy in the bill.
He said that the sentiment behind the bill was “to be supported” – calling it a “positive, forward step” – but added it lacked detail on how data can and should be shared around departments.
“It is not clear yet that the current sharing agreements within government are appropriate,” said Bracken, who is now chief digital officer at the Co-op.
The bill, Bracken said, “seems to reinforce the primacy of Whitehall’s willingness to share more data in ways that it has been sharing data over time” – rather than making more fundamental changes.
“The government uses bulk data too often when what is actually required is only a small amount of data by another government department. There are different mechanisms that can do that more safely,” he said, for instance by querying a dataset using an application programming interface rather than sharing an entire dataset.
“I suspect it is that willingness to share very large sets of data in different ways for the convenience of government departments and agencies that is the root cause of the unease around the data sharing part of the Bill,” Bracken said.
He added that it appeared that some elements of the bill were “driven more by the operational structures of Whitehall and government agencies than by the needs of users accessing that data”.
This would go against the GDS principle of user need, and is a point that has been previously been picked up on by technologist Jerry Fishenden, who earlier this year said that the bill did not ask whether better services could be created by redesigning them to meet users’ needs, rather than trying to “reverse engineer” a solution through data sharing.
“The fact that service design (and hence data) is fragmented across organisations is a reflection of services designed around organisational structures and their needs, rather than citizens,” he wrote in a blogpost.
Meanwhile, MPs asked witnesses how government could ensure the public believed it would make good use of their data, after a number of surveys have indicated trust in government is low. A recent study by the Information Commissioner’s Office found that just 36% of people trusted government departments with their data.
Jeni Tennison, the chief executive of the Open Data Institute, said that more clarity on how the data will be used was the only way to improve public trust, saying that the current arrangement are “opaque at best”.
It was important to communicate the plans in a clear and understandable way, she said. “We would like to see a lot more transparency about what existing measure there are, and how these [new] measures fit in with those measures, so people can get to grips with the way data is flying through government,” she said.
The MPs also pressed the pair on how the government could avoid another care.data scenario, referring to the government’s failed attempt to improve the sharing of patient data within the NHS.
Tennison answered that the way to do avoid another scandal would be to put more provisions around openness and transparency into the bill.
“It’s a very difficult space between balancing the right to privacy and the public good, which we can get from the use of data,” she said. “It is a fuzzy and difficult one, one we are going to be working through for many years, but having transparency and openness about it enables us to have an informed debate about where we are making that balance.”
In a later evidence session, Hetan Shah, executive director of the Royal Statistical Society, made a similar comment, saying that the bill “missed a trick” on the issue of privacy and public trust by not highlighting areas of excellence.
The Office for National Statistics, he said, had a very good track record of keeping data secure and being transparent in its work, as well as criminalising the proceedings of misuse of data, but this “has not been put on the face of the bill”.
“A tremendous amount could be done to reassure by taking what is already good practice and putting it on the face of the bill,” he said.
Statistics and research
A further discussion point was on the use of administrative data – information routinely collected by government through its normal work, such as tax data – for research and national statistics.
Both Shah and Charles Bean, who led a review into the UK’s statistics earlier this year, warned that the UK was falling behind other countries, including Canada, New Zealand and the Scandinavian countries, in its use of administrative data to create national statistics.
The idea is that, by combining administrative data with existing statistical sets, such as the census, you gain a better sample size – often the whole population rather than the subset assess in surveys – and can cut costs by carrying out surveys less often.
Shah used international students as an example.
“At the moment there are Home Office data in one place, the Higher Education Statistics Agency holding useful data in another place and there are labour market data held in a third place,” he said. “You could bring all those things together to actually track the impact and the numbers and so on, which at the moment we just do not have a good handle on.”