We continue to be impressed by the pace at which organisations have adapted their data collection and dissemination processes to provide necessary information on how the COVID-19 pandemic is affecting various aspects of our lives. The move from face-to-face surveys to telephone surveys has created real challenges for the production of key economic indicators such as population and labour market statistics.
There has been heightened demand for real time information during the pandemic and ONS has played a pivotal role in meeting this demand, through the introduction of its Faster Indicators series and by accelerating its plans for greater use of administrative data. One example of this has been the publication of experimental statistics on earnings and employment drawn from HMRC’s Pay As You Earn (PAYE) Real Time Information (RTI) system. These statistics provide a timely insight into the impact of the pandemic on unemployment and earnings.
Whilst there have been positive developments with new sources of data for understanding the labour market, there are continuing issues concerning the Labour Force Survey (LFS) and the quality of labour market estimates derived from this survey during the pandemic. Towards the end of 2020, it became clear that the change in surveying methods for LFS had affected the level of non-response bias, and in particular, affected the housing tenure breakdowns. The data showed a fall in the proportion of renters in the sample compared with an increase in the proportion of owner occupiers. To mitigate this impact, ONS introduced housing tenure weighting into the LFS weighting methodology which it explained in an article published alongside the October labour market release.
The tenure weighting has offset some of the impact of the move to telephone interviewing, however, there remains further issues which require attention. For example, the ratio of UK born to non-UK born workers was highlighted in a recent ESCOE blog. ONS has been transparent about the challenges it is facing, which it set out in a recent blog, and we welcome its efforts to look at alternative sources of data. It is our expectation that ONS continues to work closely with its key users and other statistics producers to explore opportunities to address these issues, and it should continue to provide caveats in the statistics in the interim, particularly around the greater levels of uncertainty that exist in the data.
We will continue to monitor these developments and look to ONS to be transparent about its progress in addressing the concerns raised about the labour market outputs from the LFS.