Rebecca Evans, Cabinet Secretary for Economy, Energy and Planning
This week, the Office for National Statistics (ONS) published the latest monthly labour market statistics for Wales. These included estimates from the Labour Force Survey (LFS) of rates of employment, unemployment and economic inactivity.
The estimates from the LFS once again are not robust enough to say whether they reflect a real change in labour market conditions. The latest data shows there were no statistically significant changes for headline labour market indicators which means that we are unable to know whether the estimates reflect a real change in the labour market or are due to changes in the sample of people used to generate the estimates.
The LFS estimates are produced by asking a sample of thousands of people in Wales questions on a variety of labour market topics, crucially including information on both those working and not working and using those responses to provide information on the labour market as a whole. If we drew many samples, each would give a different result, known as sampling variability. One of the challenges faced by the ONS has been maintaining response rates and therefore, the size of the sample. For example, at a UK level, the achieved number of person interviews has fallen from around 81,000 in 2014 to around 47,400 in 2024.[1]
Smaller sample sizes can increase sampling variability and make it more difficult to make accurate estimates about the labour market. This is especially true for estimates for Wales, the other devolved nations and English regions.
The regularity of the LFS – with updates published each month as part of a wider suite of labour market data – make it one of the most timely sources of information on the UK labour market. However, the LFS estimates are particularly volatile and have been for some time, with large differences between time periods. This volatility has increased as the sample size has fallen which makes it difficult to accurately assess the state of the labour market in Wales when only using this data source.
Due to these issues, the LFS-based labour market statistics are no longer classed as accredited official statistics, instead they are now labelled as ‘official statistics in development’ until further review, with caution recommended when interpreting the data. Official statistics in development are statistics that are undergoing development and are subject to testing in line with the standards set by the Code of Practice for Statistics for trustworthiness, quality, and value.
Other LFS users, including the Bank of England and the Office for Budget Responsibility, have expressed concerns about the quality and have noted the volatility of the UK level estimates. However, the issues for Wales and other countries and regions within the UK are likely to be even greater due to the smaller samples used to generate the estimates.
Given the continuing challenges with the LFS, the Welsh Government recommends not solely relying on this data source to understand changes in the labour market, especially over recent periods. The best way of understanding the performance of the Welsh labour market is to consider longer term trends across a basket of indicators. These include data sources, such as HMRC real time information on paid employees, data on workforce jobs, the claimant count and the Annual Population Survey; a larger labour market survey which the Welsh Government boosts.
Evidence from these data sources suggests that the labour market in Wales has followed similar trends to the UK as a whole since the pandemic.
Further information on the challenges, updates and current badging of the LFS are available in a Chief Statistician’s update on the labour market estimates for Wales.
Given the challenges outlined above, the ONS is introducing a new Transformed Labour Force Survey (TLFS) which they plan to become the main data source for information on the labour market in the UK and Wales in the future. The ONS is continuing to test some further design improvements for the TLFS and have indicated that they will report on progress in Spring 2025. In the meantime, the ONS will continue to use the LFS as its lead measure of the labour market, however for Wales we continue to recommend that consideration should be given to the trends exhibited by other sources.
I recently met with the UK National Statistician and set out the current challenges with using the Labour Force Survey data for Wales. Welsh Government analysts are continuing to work closely with the ONS to make clear how important reliable data on the labour market for Wales is. We recognise the substantial work it is undertaking on the TLFS, which should bring improvements to labour market statistics across the UK. We also continue to call for the ONS to provide users of the current LFS data for Wales clear information on what reliance can be placed on them and what other sources can provide an indication of what is happening in the labour market in Wales.