The mysterious non-claiming unemployed

Throughout the financial crisis there’s been a puzzle gnawing at me, which seems critically important – yet has been barely mentioned. It’s glaringly obvious when looking at the BBC news reports after every release of the unemployment figures, the latest version of which is this (see also here):

I’m not talking about those aspects that the media & commentators have focused on: the very high unemployment total, or even how this figure itself is an underestimate of ‘underemployment’ (see here, here and here). Instead, I want to know: why do so few unemployed people claim unemployment benefits , compared to the last recession?

Clarifying these numbers

To quickly explain what these two numbers in the chart above mean:

  • The ‘unemployment count’ in this case is taken from the Labour Force Survey, and categorises people as unemployed if they have been actively seeking work, AND are available to start work if a job is offered.
  • The ‘claimant count’ is official data on the number of people claiming unemployment benefits (now called Jobseekers’ Allowance or ‘JSA’). So it’s about benefit claims rather than unemployment.

So the issue is that there are now more people available to start work and actively seeking work, relative to the numbers claiming unemployment benefits.

One problem with the figures above is that they are absolute numbers rather than rates. If we go direct to the National Statistics data (assembled from two tables), then we see that unemployment in 16-64 year olds was 10.0% in Jun-Aug 1992, compared to 8.3% in the same period in 2011 (via a low of 4.8% in 2004-5). In contrast, unemployment benefit claims among 16-64 year olds was 12.3% in 1992 – 2.3 percentage points higher than the unemployment level – while in 2011 it was 5.8% – 2.5 percentage points lower than unemployment per se.

Some answers, some questions…

Now, some differences between the unemployment rate and the claimant count rate are perfectly understandable. Nearly 300,000 students in full-time education are counted as ‘unemployed’ on the above definition, but it is likely that many of these do not sign on; there are likely to be short delays in becoming unemployed and claiming benefits; some unemployed people claim other benefits (and some non-unemployed people claim JSA, e.g. some sick and disabled people); and various other reasons besides.

But what could possibly explain the sharp change over time in the balance between unemployment and JSA claims?  Two explanations seem most likely: either unemployed aren’t entitled to the benefit any more, or they are entitled but aren’t claiming it.

Entitlement: you can see in the chart above that the two lines start seriously diverging after 1996. This is when the old ‘Unemployment Benefit’ was replaced by JSA, which changed the eligibility of the benefit. [You can see a summary of the two types of eligibility to JSA on p8 of this report].

In the only recent comment on the divergent trends that I can find, Prof Steve Fothergill (whose work I regularly use) says:

“Quite a lot of people who are out of work find they don’t qualify for Job Seekers Allowance. It’s now means tested so if you have a partner in work you don’t get anything. It’s forcing individuals to be supported by other household members and households are pushed down closer to the poverty line.”

Take-up: however, restricting eligibility is not the whole answer. According to the official analysis [p144], the take-up of JSA has fallen sharply since the last recession:

[As these figures get updated you can find them here]

Now take-up rates are incredibly difficult to calculate – partly because benefits eligibility is so complex that it’s hard to figure this out from survey questions; partly because people are asked to give proof in real life that they aren’t asked in a survey; and partly because people are surprisingly bad at reporting if they’re claiming benefits (and which benefits they’re claiming) in surveys. So the Department for Work and Pensions sensibly says to be careful in using these figures.

Nevertheless, given the evidence above, it seems likely that there’s been a genuine, noticeable drop in JSA take-up even since 1997 – which might have followed an earlier drop 1992-1997, given that the gap between unemployment and the claimant count started before the take-up figures begin.

This raises FURTHER questions though, about why take-up has gone down. One possibility is due to stigma – something that I’m investigating in a project with Kate Bell and Declan Gaffney, so more on this in coming months.  Another possibility is that people are claiming different benefits, which I might come back to in a future post.  Or it might be because the incentive to claim unemployment benefits has gone down. Peter Kenway’s JRF report on unemployment benefit levels shows the decline in payment rates like this [p12]:

As I’ve previously written on the blog though, the low unemployment benefits are supplemented more by other benefits in the UK than in other countries – although still, the collective package is not generous in comparative terms.

A final thought

There’s an interesting discussion to be had about what this all means, and whether this is a sign that our social security system is failing in the midst of the financial crisis. But what seems particularly strange is how this isn’t a topic of discussion – a mystery on top of a mystery, to my mind…

20 responses to “The mysterious non-claiming unemployed”

  1. Hi Ben,

    Thanks for your very interesting articles, which I’ve been following.

    I would add that the behaviours and systems of Jobcentre Plus staff and the organisation hold people back if they don’t absolutely have to claim (eg they have a partner who is working or some savings). I’m an employment adviser and the reports of JCP ‘services’ to their customers is quite shocking. I don’t understand why JCP staff are not trained to at least level 3 advice standard: impartial, knowledgeable, informed about equalities barriers and legislation and proactive.

    I was myself made redundant in 2010. There is absolutely no way that I would go anywhere near JCP to claim. Unfortunately, most of my clients are not able to avoid it. They are desperately looking for work and work experience and almost never get the assistance they need from JCP. Quite the opposite. Many of them suffer from nervous health conditions directly as a result of longterm exposure to this statutory hazzard…



    • Thanks for the comment Sheila. Just to clarify – what exactly are the problems with JCP? As far as I understand it’s some combination of (i) incorrect advice or generally a poor service; (ii) discriminatory behaviour by ethnicity/gender/disability; (iii) stigmatising attitudes to claimants in general; (iv) excessive intrusion into people’s personal lives. But which bits do you think are worst about it, and which stopped you using JCP?

      • You’ve summed it up, really. I think the core problem is political will. Jobcentre Plus is poorly resourced and the message to them from politicians is to try to do their jobs with poor management, minimal training and chaotic communication and quality assurance systems. If it were a business it would have stopped trading and been taken over by competitors. it is not a business, thankfully, and I think the key way to improve it is from the top down. From Government, in other words. By recognising the value of the welfare service to society and investing in it. I’ve tweeted some interesting IPPR links just now on that, by coincidence.



  2. Great blog, Ben.

    Ontario has seen a similar drop in social assistance claim rates despite high unemployment. One possible explanation that we’ve been considering is that social assistance rates are so low that people who are eligible are instead living on credit for as long as they can. Credit is still easily available and it is possible that people are maxing out their credit cards before turning to social assistance. We’ve heard anecdotally from caseworkers that a lot more people have significant consumer debt when they apply for support.

    Just one more thing to consider!

    • Important to consider, in amongst all the other influences! There’s some interesting research from the US showing that most people delay claiming benefits for a little while, so maybe it’s this that changed. I’ll come back to this in future posts, definitely…

  3. I think the main problem is entitlement, especially that of couples where one works and the other is no longer entitled to benefits. I worked with a young person whose partner was working at minimum wage, her partner was also young so on a low minimum wage. Between them they were ‘earning’ more on benefits before he got work. If they lived in separate homes they would be better off but costing the taxpayer more money – 2 lots of housing benefit etc. It is mad, work should pay more than staying on benefits – and the answer is not to reduce benefits! Some of this shortfall may come from immigrants who are here legally but not working yet and not eleigible for benefits. (please note this is not a ‘immigrants stealing our jobs’ dig, I am happy to have a diverse population)

    • I’d assume that the number of unemployed legal migrants not entitled to benefits is small (partly because some people lose their entitlement to be in the country if they lose their job – which happened to some friends of mine). But interesting to check this, and it certainly does exist – a South American guy I spoke to for my PhD had been in a motorbike accident and was bedridden for 10 months, was not entitled to benefits, and somehow his family got through this with help from other family and friends. In general I need to have a closer look at changing entitlement conditions over time, and the numbers in each disentitled category. So thanks for mentioning it!

  4. UK isn’t alone in having a big gap between ILO unemployed and unemployment benefit receipt. U.S., New Zealand and Canada have even bigger gaps. France also has surprisingly few claimants though a higher % of ILO than UK (perhaps reflecting longer scholastic careers). International comparisons are hazardous of course for same reasons UK trends are tricky to analyse. For a forthcoming report Kate and I looked at the data from the World Social Security report for comparable (i.e. wealthy) nations. An interesting factoid that rather bears out Steve Fothergill’s point is that in this sample of 13 countries, only New Zealand and Australia have lower % of unemployed receiving a *contributory* unemployment benefit than UK. New Zealand and Canada do not have contributory unemployment benefits.

      • This is when I get very glad that for the project on benefits stigma, me and Declan are working together – the quick response is really interesting. Looking forward to seeing these numbers on the blog at a later stage!

  5. I think Sheila Heard’s comments nail it – whether or not JCP are deliberately awkward, they are incompetent. The hoops that you need to jump through are numerous and the paperwork that is ‘lost’ (on more than one occasion) similarly so. I can verify this having worked at different CABx. People either give up or become disenchanted with the system. What I think will be very interesting is that when Universal Credit is introduced, take-up of all benefits should rise exponentially as no-one should fall through the gaps as they do now. However, I suspect the reality is that whoever administers UC may make the hoops even more numerous and higher to jump through, thereby leaving many completely devoid of any help.

    • What happens with Universal Credit will be interesting – and it’s one of the few recent changes that could (in principle, ignoring everything going on around it) be quite helpful.

      But on the other hand, I’m worried about how the different parts of the system interact. Ben Richards and John Hills at LSE have recently written a paper – see the blog posts by John Hills and Gavin Kelly – on how there are massive effective tax rates due to the system of university fees bursaries, despite the attempt at simplification in UC. We’re a long way from some of these goals, still…

  6. Anecdotal of course, but I’ve twice been out of work long-term and not even applied for JSA, because I had money saved up and it seemed dishonest (as well as being a massive pain in the hole and waste of time I could have spent doing something less awful than filling in their forms, such as setting fire to my head). That would surely account for a few non-claimants, although I doubt it’s very many.

    Interesting stuff though either way.

  7. Important to go back further than your chart: to the late 70s/early 80s. that is the point when a trend began which – from memory – peaked around 86 – 88. It was exactly the reverse of your exam question above ie the claimant count persistently exceeded ILO unemployment, principally because a large number of people who were signing on were not in fact looking for work.
    That led to a series of public policy changes, notably
    – in OECD terms a more active labour market policy in which sanctions and employment programmes together became more widespread….the work programme being the latest manifestation;
    – institutional change to deliver that policy more effectively…..formation of the Employment Service in 86 and Jobcentreplus in 2001…notable for bringing jobsearch support conditionality testing and benefit payment for the first time in a model that has been recognised internationally as somewhat more successful than commentators above imply;
    – a track record of making JSA harder to get and with an assertive and interventionist regime wrapped around it.

    The effect of all this is to make signing on for JSA a really unattractive option, and indeed a benefit which it is hard to qualify for. That in turn explains that at the low point of the last three recessions the claimant count has peaked at successively lower levels. This is seen as a policy success because it reduces benefit expenditure and increases flows off benefit into work: in other words it has worked well in labour market and public expenditure terms. I do appreciate however if you view this trend through a classic social welfare lens you might view it as less benign…..

    • Thanks salientwork – interesting point, and when (if!) I look at this further I’ll be sure to go back to the mid 80s. I’ll also think at the same time about why we even have unemployment benefits – if we just value this in terms of public expenditure, then we’d get rid of the benefit completely… But setting out what other goals it aims to achieve is not always easy (see also the post coming up on Tuesday btw).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: