The world of work has changed dramatically over the last decade. The employment rate in the UK has seen record highs in recent years, and the way that people work has become much more flexible, with a large increase in self-employment and part-time working. The UK employment and jobs statistics, produced by the Office for National Statistics (ONS), are hugely valuable because they help us understand what the labour market looks like and how it is changing over time.

Now, more than ever, these statistics need to be able to keep pace with a changing environment: COVID-19 has caused an unprecedented challenge to the labour market and wider economy. It is crucial that we are able to measure and understand the impact, as the public needs statistics that help them understand the scale and nature of the changes. We have been impressed by the agility and pragmatism of ONS in its response to the COVID-19 outbreak so far.

Today we have published documents covering the trustworthiness, quality and value of the employment and jobs statistics produced by ONS and statisticians in the Welsh Government, the Scottish Government and the Northern Ireland Statistics and Research Agency, to examine their value for all users across the UK.

Our assessment is an interesting case study of the common issues that statistics producers face. It really is a microcosm of the UK statistical system. We came across many recurring themes from our other regulatory work including: the potential of administrative data to generate insight and fill data gaps, the need for clear and prominent information about statistical uncertainty, and coherence of data and statistics. This blog explores these themes and may offer some wider lessons for producers of official statistics.

Using administrative data sources to fill gaps in insight

To understand how the labour market is changing, we need statistics that capture all aspects of employment and jobs. There are still a number of data gaps, for example, measures of job quality and data on the self-employed. ONS is already taking positive steps to explore opportunities to fill data gaps, such as its collaboration with HMRC to use their Pay As You Earn (PAYE) Real-Time Information (RTI). Access to these data offers huge potential to enhance the value of the statistics because it captures detailed information on all employees on the PAYE system. ONS should focus its efforts on understanding how RTI data can be integrated with existing data sources to maximise the value of the statistics. Active and ambitious leadership from ONS is vital to achieving this change. It would be helpful for users if ONS published regular updates on these statistical developments to enhance the transparency of its plans.

Being clearer on uncertainty in the estimates

The employment and jobs statistics are mostly based on household and business survey data, which means that they are only estimates of the true number of people of in employment and the number of jobs. An indication of the level of uncertainty around the headline estimate is useful because it allows users to understand whether changes over time or differences between countries or regions are meaningful. In other words, are the changes real or are they an artefact of the way that households or businesses are sampled? Our view, endorsed by the users we spoke to, is that uncertainty is not properly reflected in the messaging in ONS statistical bulletins. This means readers might jump to the conclusion that the figures in the bulletins and tables are more precise than they really are. We have asked ONS to integrate information on uncertainty and explain reasons for changes.

Ensuring statistics and data sources are coherent

Coherence reflects the degree of similarity between related statistics and the insight that can be achieved by drawing them together. The coherence of the employment and jobs statistics was identified as a key strength in both our assessment of ONS’s statistics and our review of labour market statistics in the devolved nations. The statistics are consistent over time, comparable across geographical areas and the definitions are in line with international best practice, which reflects their quality. However, we heard from some users that there are issues with coherence of certain data sources, including the headline estimates from the Labour Force Survey (LFS) and Annual Population Survey (APS). This is a challenge for ONS, especially at the current time when COVID-19 is dramatically changing the way LFS data are collected. To enhance the quality of the statistics, it is important that ONS and its partner statistics teams in the devolved nations work effectively together to enhance the value provided to users of statistics in their own nations and across the UK.