If you are trying to predict the outlook for unemployment it is, of course, essential to get the starting point right. This blog therefore runs through the latest available official data and employment surveys. A follow-up piece will outline my own forecasts.
The pandemic has raised question marks over all the usual measures of unemployment and there is still plenty of confusion (particularly in the media) over what they mean.
The ‘claimant count’ has soared, but this partly reflects a relaxation of the rules which means that a large number of people who are actually still in work are now eligible for unemployment-related benefits. This figure is therefore (almost certainly) an overestimate of those without jobs.
In contrast, the headline unemployment rate, derived from the Labour Force Survey (LFS), probably understates the deterioration in the labour market. In part this is because people who are on furlough still count as employed even if they are not doing any work.
What’s more, people who are no longer actively looking for work, perhaps because they do not expect to find a job, do not count as unemployed.
For the record, though, the LFS measure of the unemployment rate for those aged 16+ was 4.1% in the three months to July, and 4.4% in July alone (although the single month data are based on a smaller sample and are more erratic). This corresponds to about 1.4 million unemployed in the three months to July, rising to 1.5 million in July itself.
At face value, the LFS numbers might suggest that hardly any jobs have been lost. Unemployment is certainly still low by past standards: the headline rate peaked above 8% in the wake of the last global recession in 2008-09. However, other measures give a clearer picture of what has been happening to employment.
A far better guide to the number of job losses is the experimental PAYE data, which are also timelier (they are already available for August). These suggest that UK payrolls fell by about 706,000 between February and August, with by far the biggest fall taking place in April.
Note that this is a net figure: the number of people who have lost jobs was larger than this, but some of them (or others) will have found new ones. What’s more, the main reason for the fall is that employers have cut the number of people they are hiring, rather than increased the number of redundancies. Indeed, with the exception of April, the number of people leaving jobs each month has been lower than usual.
Indeed, LFS data on redundancies show only a small rise, at least relative to the 2008-09 recession. (Note also that there are normally tens of thousands of redundancies every month, even in good times.)
The final piece of the puzzle is how many people are still on the government’s furlough scheme. Probably the best guide here is the ONS Business Impact of Coronavirus Survey (BICS), covering 7th to 20th September, which suggested that 9% of the workforce remained on furlough. The PAYE data suggest that total private payroll employment was about 28.3 million in August, and 9% of this is about 2.5 million.
Note this 2,5 million figure will include many people who are already working again part time, while more than half (56%) of businesses are voluntarily topping up wages – two factors which suggest the demand for labour is healthier than the headlines suggest.
The true figure is probably also now lower – probably closer to 2 million. In part this is because some of these people may also already have lost their jobs. However, other business surveys (including the PMIs and the BICS) suggest that the economy has continued to recover, despite the disappointing GDP data for August.
Above all, recruitment surveys show both a pick up in online job ads and in new hiring, and a reduction in the pace of redundancies.
In short, the UK labour market has probably entered the autumn in much better shape than many of the recent headlines suggest – though this period will undoubtedly see unemployment rise further.
Plain-speaking Economics 