The ‘hidden data’ that could boost the UK’s productivity and job market

News – Blog

The ‘hidden data’ that could boost the UK’s productivity and job market

28 June 2021

Authors:
Hazel Klenk & Fay Sadro – the Learning and Work Institute
Jack Orlik, Amy Solder, Rhys Herriott and Sarah Mcloughlin – the CareerTech Challenge team.

A new report from Learning and Work Institute, Nesta and Challenge Works highlights the complexities (and opportunities) of using labour market data to support adults plan their careers.

Download the full report from Nesta’s website (PDF). If you are interested in discussing the findings from the report please contact us at [email protected].

The Skills for Jobs white paper has become a strategic centrepiece for UK government. In this white paper, the Department for Education sets out a plan to align the supply and demand for skills in the job market through a mix of measures that includes employer-led Local Skills Improvement Plans, new financial entitlements for training and ‘great careers support’.

If these approaches work, they could help bring the UK’s high level of ‘skills mismatch’ to best practice levels, boosting productivity by up to 5 per cent (OECD). They could also accelerate recovery from COVID-19 by giving workers the support they need to find new roles that realise their potential, make use of their skills and offer them a more secure future in a labour market that is being disrupted by new technology.

But how can ‘great careers support’ be delivered at a scale that has a real impact on the labour market? This was a central question that drove the CareerTech Challenge, a partnership between Nesta, Challenge Works and the Department for Education which sought to stimulate the development of new tools for career navigation, and uncover ‘what works’ to help adults develop new skills and find good work.

The barriers to ‘great careers support’

Of the 31 innovators supported by the CareerTech Challenge, 20 Prize finalists worked on building and testing new tools to make careers information, advice and guidance more accessible and tailored to the supply and demand for skills in local job markets. Their progress was documented and analysed through interviews by our evaluation partners, Learning and Work Institute providing a unique opportunity to understand the enablers and barriers for the development of ‘great careers support’. The research revealed an abundance of creativity among the innovators, and demonstrated the value of user testing, co-creation and partnerships with local stakeholders, such as FE colleges and employers. Yet it also uncovered a major challenge that is holding back innovation in the field – limited and inaccessible labour market data.

The deficiencies in the UK’s labour market data are illustrated by the experiences of the winners of the CareerTech Challenge Prize, the team developing Bob UK, a tool designed to provide instant, online careers advice and job recommendations based on information about local vacancies and the jobseeker’s skills. The developers attempted to source UK data that directly replicated data sources used to develop the version of Bob which has helped over 250,000 jobseekers in France. However, it became apparent that equivalent sources of data rarely existed. The Bob UK team was able to work around this issue by carefully combining alternative sources of data from a number of UK and non-UK sources.

Bob, the winning solution from the CareerTech Challenge Prize, curates a wide range of data to provide useful insights to jobseekers.

A chatbot provides advice about entering an industry impacted by the coronavirus pandemic.
A series of graphics identifying the automation risk of various jobs

Many other innovators experienced similar barriers, finding that the publicly available data that could help people to make more informed decisions about their careers is often incomplete, difficult to use and poorly described. The impact of this is significant. A shocking insight from the report is that one solution enabled careers advisors to base course recommendations on labour market information for the first time. Prior to using this tool, such information was too time-consuming for careers advisors to uncover and analyse for it to be of use, and job seekers were given advice that was not based on employer demand for skills.

How better data can help deliver great career support

To address this issue of hidden and missing data and unleash the productivity-raising potential of better skills matching, the report  makes a series of recommendations, including:

  • The creation of a central labour market data repository that collates publicly available information about the labour market.
  • Public data providers should review the quality and accessibility of the data they hold, and make it easier for developers to use.
  • The development of better skills and labour market taxonomies to facilitate consistency between sources and enhance data matching.

Nesta’ Data Analytics team has launched a number of initiatives that aim to fill gaps in the UK’s supply of labour market information:

  • Creating a skills taxonomy: With funding from ESCoE, the team created the first open data-driven skills taxonomy for the UK. This allows us to track the demand for skills, as well as map the distribution of skills across the UK. We are currently updating the taxonomy and hope to publish the next version towards the end of this year.
  • Mapping viable transitions for workers: In Mapping Career Causeways, supported by J.P. Morgan, the team used machine learning to measure the similarity in the skills and work activities required in more than 1,600 jobs. This information can be used to suggest ‘viable transitions’ to job seekers, based on their most recent role. The underlying algorithm can also identify the skills gap between any two jobs which can inform decisions around training. The code for this project is open and available on Github.
  • Measuring automation risk: With funding from The Gatsby Foundation, the team provided the first known estimates of automation risk for apprenticeships. Policy recommendations included increasing awareness about the types of tasks that were found to raise risk (such as routine and repetitive activities) and those that lower risk (such as tackling unstructured problems in changeable environments).
  • Providing free insights on skill demands: In partnership with the Department for Education, the team is creating an Open Jobs Observatory. The Observatory will contain free insights from online job adverts, with a focus on the skills requested by employers. The team is collecting the adverts with the permission of job sites. A pilot version of the Observatory will be launched this year.
  • Identifying green jobs: The team’s next focus is developing a methodology for tagging green jobs. At present, the UK has no way of identifying individual jobs in green sectors. Developing this methodology is a necessary step in transitioning towards a greener economy.

As well as the CareerTech Challenge Prize, Challenge Works is continuing to support innovation which supports people adapt to a rapidly changing labour market:

  • The Rapid Recovery Challenge to find and scale tools and services that improve access to jobs and money for people within the UK – with a specific focus on those hardest hit by the economic shock resulting from COVID-19. The £3 million Challenge is funded by Nesta, in partnership with JPMorgan Chase Foundation, the Department for Work and Pensions and the Money and Pensions Service.
  • The European Social Innovation Competition includes a challenge prize to incentivise, support and reward social innovation that will help people and organisations identify, develop and strengthen the skills they will need to adapt and thrive in a changing world. The 2021 edition of the Competition, ‘Skills for tomorrow’ will award three €50,000 prizes for the best scalable social innovations that aim to contribute to job creation, growth and overall European competitiveness and power the European economy’s green and digital future.

You might also like…