A tale of two unemployment statistics

1.6 million people were unemployed in the UK in October-December 2016, according to the latest figures from the Office for National Statistics (ONS). At 4.8% the unemployment rate is currently around the same level as in 2005.  But while today’s headlines will focus on the unemployment figure, there is another number lurking in the ONS UK Labour Market Release: the ‘claimant count’.

The claimant count is the number of people receiving benefits because they are unemployed….

In practice, it is the number of people claiming Jobseeker’s Allowance or Universal Credit, where the receipt of their Universal Credit is dependent upon them seeking work. During November 2016 there were 0.8 million people claiming such benefits in the UK, 0.8 million fewer than the total number of people who were unemployed.

Not all unemployed people are able to receive unemployment benefits. There are various reasons why they might be ineligible, for example because they are in full-time education or over State Pension age. On the other hand, some people claiming unemployment benefits may be employed for a small number of hours each week.

….while ONS unemployment figures are based on survey data

The ONS calculates the unemployment figure by asking a representative sample of the UK population about their employment status. If a survey respondent is out of work but has looked for work in the past four weeks and is ready to start work within the next two weeks, they are counted as unemployed. The ONS then takes the number of respondents in their survey who are unemployed and scales this up to get an estimate of the number of unemployed people nationwide. We can be happy that this figure is reliable.

The claimant count on the other hand is an administrative statistic. Rather than being an estimate, like the unemployment figure, it is based on the actual number of people in receipt of unemployment benefits.

We can be more confident about claimant data for Parliamentary constituencies….

While the unemployment figure is reliable for the whole UK population, our confidence in the figure diminishes when we use the survey to estimate the number of people unemployed in small populations.

For an MP wanting to know what’s happening to unemployment in their constituency an estimate from the ONS’ national survey is too unreliable. This is because the survey would have asked few people in the local area and therefore there isn’t enough information to make an accurate estimate.

The claimant count however is counting the number of people who are actually claiming an unemployment benefit and is therefore accurate even when looking at small populations.

As the claimant count is accurate at small geographic areas it can serve as an alternative measure for unemployment in a small area, such as a Parliamentary constituency. The Library therefore publishes the claimant count for each parliamentary constituency in our briefing paper People claiming unemployment benefits by constituency.

….but does the claimant count give the full story?

The large gap that currently exists between the claimant count and unemployment level should make us uneasy.

In the past using the claimant count as a local measure of unemployment probably told us more about the actual level of unemployment in that area. This is because there was a closer relationship between the claimant count and the unemployment level nationwide.

The chart below shows the number of people who are unemployed and the number claiming an unemployment related benefit since 1971.These two lines hug each other fairly closely till the mid-1990s when a gap between them begins to widen. At their closest, in August 1993, the difference between the unemployment level and the claimant count was around 15,000.

The largest gap between the two measures, in October 2011, was 73 times this, with a difference between the claimant count and unemployment standing at over a million.

That’s a big difference!

The difference is particularly stark when we compare the unemployment level and claimant count during the last three recessions. During the recessions of the 1980s and 1990s as the level of unemployment rose the gap between the unemployment level and the claimant count shrunk. During the most recent recession the two figures both increased but they didn’t get any closer.

There’s no simple explanation for the divergence between unemployment and the claimant count in recent decades, although some of it will be down to changes in benefit rules such as the introduction of Jobseeker’s Allowance in October 1996.

As we do not have accurate and timely unemployment figures for local areas we can’t tell how different the two figures may be in any given constituency (although a comparison could be made for early 2011 using Census data). The national difference does give us a clue that they are likely to be quite different!

Still, the claimant count offers useful insight into how unemployment is changing locally

All of this might make you think: if the claimant count is so different to the unemployment level, why are we still using it?

Even though there is a gap between the two figures, the charts show the claimant count still tends to follow the same pattern as unemployment nationally. The claimant count therefore still provides us with a useful insight into how unemployment is changing at a local level. It also allows us to make some comparison between local areas across the country.

Want to know more… Read the Library’s latest briefing paper on Unemployment statistics… People claiming unemployment benefits by constituency, January 2017.

Notes on the statistics used in this blog

All figures have been seasonally adjusted. The claimant count figure prior to May 2013 is those claiming Jobseeker’s Allowance only. Data on unemployment and the claimant count are as published in ONS’ UK Labour Market statistical bulletin.

Picture credit:     Job Centre Plus by Richard McKeever. Licensed under CC BY 2.0 / image cropped.