Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
The UK’s official measure of unemployment stood at 4.2 per cent in the three months between June and August this year. The definition is precise. It means that of the population aged 16 or above who are either working or actively seeking employment and ready to start in the next two weeks, 4.2 per cent did not have a job.
A longstanding problem with this official definition is that it does not accord with most people’s view that you are unemployed if you are of working age and do not have a job. That should make us think about how we regard unemployment. But the more immediate crisis is that the official figure is probably nonsense.
Should we be pleased that the latest official unemployment rate is down from 4.3 per cent in the previous official publication? Should we worry that the latest rate is 0.2 percentage points higher than it was three months ago? Or should we feel relief that the latest vintage of data was better than the previous one when the quarterly rise in unemployment was measured at 0.5 percentage points?
The answer is we have no idea what we should feel because we cannot have confidence in any of these official figures, even though they come from the august Office for National Statistics.
The survey that traditionally underpins the main labour market data has seen its response rate decline from almost 50 per cent a decade ago to 15 per cent this year. So bad is the data that the ONS stripped the figures of the “national statistics” kite mark on Tuesday this week, but due to the weird way the law prescribes these things, the figures are still “official statistics”.
The data crisis is deeply problematic for economic policy. The ONS cannot currently tell us with any certainty whether the labour market is loosening or tightening, so the Bank of England will have more difficulty than usual in setting interest rates. It cannot tell us whether the UK really has been an international outlier after the pandemic in seeing a sharp rise in the number of sick people not seeking work, potentially highlighting the damaging economic effects of high NHS waiting lists. It cannot provide any guidance on whether tax changes have caused a collapse in the self-employed.
It cannot do these basic tasks because its figures might well be rubbish.
It was obvious to me last month that the unemployment data could not be trusted. But instead of abjectly apologising for a problem it has known about since early 2021 when Michael O’Connor and Jonathan Portes first highlighted strange figures coming from the labour force survey, the ONS published an upbeat blog on Tuesday. According to the UK’s statistical agency, we should be pleased it is “transitioning to better methods of data collection fit for the digital age”. That was extraordinary.
Voters should demand a solid underpinning of fact to guide political discourse. Ministers have a right to be furious that the UK labour market figures are chopping and changing and might well paint an unnecessarily negative picture of the economy, just as the previous vintage of gross domestic product figures did. But they should also feel a little ashamed that the international reputation of the statistics is deteriorating and that is partly down to a lack of funding for a very small and cheap government department.
Money is not everything, however, and it is for the ONS and its regulator the UK Statistics Authority to stop covering up the horrors in our labour market data and state clearly what all policymakers and close watchers of the data know.
Right now in the UK, we do not know what is happening to unemployment, labour force participation and the fortunes of different groups in the jobs market. In important respects, our economic policymakers are flying blind.