DWP data sharing pilot aims for single source of the truth

The Department for Work and Pensions (DWP) has begun work on a cross-department data project.

Financed through HM Treasury’s Shared Outcomes Fund, which incentivises departments to work collaboratively across departmental boundaries and challenging policy areas, the pilot aims to improve understanding and testing the feasibility of meeting the respective labour market data needs of different departments.

The project aims to make data accessible and discoverable in a way that enables monitoring and data governance. This will involve new technical procedures such as shifting from batch movement of datasets to what the DWP describes as “in-place data shares, where data is shared from a single source and reused many times without unnecessary copying”.

Working in collaboration with the Open Data Institute (ODI) on data trusts and the Alan Turing Institute’s (ATI), the project aims to explore the legal, digital and data protection framework necessary for a collective engagement. According to the DWP, the pilot aims to help departments proportionately share their data, while rigorously protecting the privacy of those whose data it is.

The project’s other main goal is to show government bodies how to use this data to improve public services, as well as respond earlier as dynamic changes in the labour market happen.

The DWP said that connecting and making better use of the government’s different labour market datasets is vital to informing and improving citizen outcomes, iterating policies based on wider evaluation and delivering the best value for taxpayers’ money.

Discussing the project, Paul Lodge, chief data officer at DWP, said: “Currently, data-sharing arrangements between departments are lengthy to put in place or amend, and resource intensive. As a result, the subsequent insight is often well-behind the pace of change, especially in something as dynamic as the UK labour market.”

According to Lodge, this makes it difficult for government analysts and service delivery teams to understand what data is held by others or to confidently navigate their governance processes. “As a result, the department data silos across government make it hard to deliver joined-up interventions at the speed needed for our ambitious policy agenda,” he added.

“To be effective, a data trust requires the data to be accessible and discoverable, and the onward extraction of any results need to be monitored and governed. That means asking new technical, procedural and legal questions of our data to see whether a data trust can help to deliver greater collaboration or what changes would be needed.”

He said that the ability to share DWP data was critical during the pandemic and to support the Covid-19 vaccination programme.

“Now the same is true for our efforts in supporting the labour market and its continued recovery,” he said. “DWP is committed to maximising employment across the country to aid the economic recovery. It’s also the case that other government departments need granular insights into everyday labour market developments, especially those affecting local areas, specific sectors and disadvantaged groups.”