It’s not as easy to count the number of ‘working people’ as you might think
How many people are employed in the UK today? This is not just a question in a pub quiz for boffins. It is one with important practical implications.
Changes in the number of people employed is a key piece of information which the Bank of England’s Monetary Policy Committee takes into account when setting interest rates.
If employment is rising strongly, for example, this suggests that the labour market is tightening. The bargaining position of workers is improved, and so wage increase will rise. This in turn will lead to higher inflation. And the prospect of higher inflation reduces the scope for interest rate cuts.
Last week, the governor of the Bank of England, Andrew Bailey, raised serious concerns about the reliability of official labour market data when appearing at the House of Commons Treasury Select Committee.
The Resolution Foundation argued last week that in fact the total number of jobs in Britain is almost one million more than the Office for National Statistics (ONS) estimate in its Labour Force Survey.
In complete contrast, in August the Bureau of Labor Statistics (BLS) stated that the American economy had created nearly one million fewer jobs over the previous year than it had originally believed. Once a year, the BLS revises its numbers, and this was the result.
The monthly announcement of the change in the numbers employed is followed with great interest in the US. It is not just the policymakers setting interest rates in the Federal Reserve. Wall Street follows the number avidly as a valuable snapshot of the state of the economy.
The fundamental problem is that the economy does not exist in a scientific laboratory: we cannot put it on a set of scales to measure how big it is.
This is not a question of incompetence by the statisticians who compile economic data either here or across the Atlantic. Both the ONS and BLS are careful, thoughtful bodies.
The fundamental problem is that the economy does not exist in a scientific laboratory. There, estimates can be measured with great precision. But we cannot, for example, put the economy on a set of scales to measure how big it is.
Instead, national account statisticians – to give them their full title – use a wide variety of information to put together a picture of the economy, whether output, inflation, employment or whatever.
But it is of necessity an estimate. Estimates can, and do, change as more information becomes available. And they still could be inexact. Outside of the financial markets, we can never know for sure the “correct” value of any economic variable.
The ONS survey of the labour market has experienced problems, of which they are fully aware, since the pandemic. The response rate to the survey has dropped from 39 per cent to just 19 per cent. The Resolution Foundation has made their own attempt to measure employment. But this, too, may be wrong.
Estimates of the growth of GDP made for any particular quarter can and do vary over time by one per cent or more. Given that in recent years the economy has barely been growing over the course of an entire year by one per cent, these revisions are huge.
For example, in the second quarter of 2008 – the quarter immediately prior to the recession sparked by the financial crisis – the ONS initially estimated a slowdown in the economy. They believed there had been no growth compared to the first quarter of the year.
Now, instead of a mere slowdown the ONS estimates a sharp fall. In other words, the Great Recession began three months before people thought it did at the time.
The ONS is making increasing use of big data, and this will definitely help. But at the same time, there is no substitute for the old-fashioned method of getting out and about and actually seeing what is happening.
Paul Ormerod is an economist at Volterra Partners LLP