Using modelling to understand journey times and jobcentre use

Downloads

• RGS-IBG case study Jobcentre journey times (.pdf)

Challenge

Understanding the geography of jobcentres is critical to making decisions over how and where to deliver services, and is a recurring challenge as policy is revisited under changing conditions and governments. The 2015 Spending Review required the Department for Work and Pensions to co-locate more JCP offices with other services and cut 20% of its estate.

Subsequently, the Work and Pensions Committee held a 2016 inquiry into the future of Jobcentre Plus, and in 2017 the DWP published plans to change how it used Jobcentre Plus offices and benefit centres around the UK. The DWP operated around 1.5 million square metres of office space, and suggested that 300,000 square metres could be vacated.

Plans to close, merge or move offices would have implications for staff and for service users, and decisions would be better informed with an understanding of impacts on accessibility and journey times.

Solution

The National Audit Office (NAO) is a Parliamentary body which scrutinises public spending. The NAO submitted a memorandum to the 2016 Work and Pensions inquiry, highlighting journey times to jobcentres by area across the UK as well as statistics on unemployment.

As a follow-up, the NAO created a model of journey time to jobcentre locations around the UK. This drew on four key data sources:

• Output areas, the key building block of the Census and used widely to structure demographic data. The model used 171,372 English output areas, each containing on average 125 households.

•  Jobcentre locations from DWP data.

• UC and JSA claimant numbers from ONS’s Nomis database.

• Travel time between each output area and Jobcentre location, using the DfT’s metholodogy; this does not include service frequency.

The model included datasets both for the existing jobcentre locations in England as of 2015, and iterations of the DWP’s proposed cuts to jobcentres. In 80 cases the model used schools, GP locations and centres of employment as proxies where the DfT’s 2016 data set did not match jobcentre locations.

Ultimately, the NAO research combined and mapped travel time estimates from 171,372 English output areas (each around 125 households) to the nearest ten jobcentre sites, producing an interactive tool which can be viewed below and on the NAO website.