Will Universal Credit push claimant unemployment above 1.5 million? And if so, what do we do about it?

27 February 2018

Last Wednesday’s employment figures were once again all (fairly) quiet on the jobs front.  As we reported, employment rose a bit, the numbers out of work fell a bit (with unemployment rising while ‘economic inactivity’ – those not available or looking for work – fell by a lot), and earnings growth was unchanged.

But beneath these headlines, we also saw continuing signs that the gap is closing between the ‘claimant’ measure of unemployment – which counts benefit claimants who are required to be available and looking for work – and the Labour Force Survey measure (which is a larger count of all of those unemployed, regardless of whether they claim benefit).  Historically, the survey measure has been higher than the survey measure since the 1990s (when Jobseeker’s Allowance was introduced), and in the last decade the gap between the two has grown. However in the last two years, while LFS unemployment has fallen by 200,000 (or nearly 15%), the Claimant Count has gone the other way – rising by 70,000 (around 10%).  By coincidence, the last two years has also seen the rollout of the Universal Credit ‘Full Service’.

The rise in the Claimant Count and the rollout of the full version of Universal Credit are undoubtedly linked.  We’ve flagged this in our monthly briefings for the last year or so, and the Office for National Statistics have pointed this out in their frequent revisions to the official Claimant Count series.  But how high will it go, what’s driving it, and what should we do about it?

On how high it might go, a simple look at the trends in the Claimant Count for offices where Full Service has rolled out and those where it hasn’t gives some signs of what’s going on.  Figure 1 below shows this – with the count broadly flat in non-Full Service areas, while in Full Service offices it has risen slowly but steadily – up by about 80,000 since rollout began at the end of 2015.  So in effect, all of the rise of the Claimant Count in the last two years appears to be explained by Universal Credit.

Figure 1 – Changes in Claimant Count unemployment in UC and non-UC offices (as at December 2017)

Source: NOMIS and Learning and Work Institute analysis

But this only tells us part of the story.  Full Service has so far only reached one third of Jobcentre Plus offices (233 out of 700, with the remaining two thirds moving over this year).  And because rollout has only recently ramped up, most offices have either not yet got UC or haven’t had it for very long.

So we’ve looked at the 233 Full Service offices alone, and then looked at what’s happened to the Claimant Count in each office before and after implementation.  We’ve then averaged this out across offices.  The results are in Figure 2 below, and are stark – with the count more than doubling in Universal Credit Full Service sites in the two years after implementation.

Figure 2 – Change in Claimant Count in Jobcentre Plus Offices before and after UC Full Service rollout

Source: NOMIS and Learning and Work Institute analysis

There’s an important health warning with this data, in that inevitably the further you get from zero the fewer the number of offices that are being included.  Only three offices have had UC Full Service for more than two years, and only one in five have had Full Service for more than one year. However, two fifths have had it for at least six months (and of course the pre-rollout data covers all 233 offices).

So while these figures will evolve over time, the trends are clear and unambiguous – for the two years before Full Service rollout caseloads are flat, and after it rolls out offices see a significant growth in their Claimant Count.  (And this happens in virtually every office where Full Service rolls out – no other factors, like seasonality or local economies, could explain it.)  It seems entirely plausible that as the remaining two thirds of offices go live, and as time goes on, the Claimant Count could easily double compared with what it was before rollout began – so 1.5 million claimants or more, by the end of 2020.  For context, this would be higher than at any point since the ‘double dip’ in late 2012.

So what’s driving these changes?  In short, it must be either more people under Full Service meeting the Claimant Count definition (available and looking for work), people spending longer on the count, or both.  The House of Commons Library produced a short briefing last year which argued for the first of these – more people on the count – and particularly, people who would have claimed only Housing Benefit or tax credits, and those awaiting assessment for Employment and Support Allowance.  This seems plausible, and it might also be contributing to the recent rises in the LFS measure of unemployment – with more people who were previously ‘economically inactive’ now available and looking for work.

However, it feels unlikely that this could explain all of the growth.  In particular, if this did explain all of it then you would expect the growth to level off.  Instead (worryingly) although the growth in the count appears to slow down, even those offices that have had Full Service for over a year are still seeing their caseloads grow.  So if it isn’t just more people signing on, could it also be fewer people signing off?  The short answer is that it’s impossible to know – because data on off-flows and durations isn’t published (unlike for Jobseeker’s Allowance), and the government’s assessments of the impacts of Universal Credit all pre-date the rollout of Full Service (in fact the only evidence that we have of the effectiveness of UC comes from the earlier ‘live’ service).

So this leads on to my final point, on what we do about this.  First, we need to know and understand what is driving the growth in the count.  So far the government has been tight-lipped on Full Service – with no new data analyses, evaluation findings or impact assessments.  But with the scale of growth now so great, and the real possibility that the count might double over the next three years, government needs to open up its data – on inflows, outflows, durations and characteristics – and give its assessment of what is happening.

Secondly, these rises have implications for Jobcentre Plus staff, resourcing and the services that they provide.  Jobcentre Plus is around one third smaller than it was the last time the Claimant Count hit 1.5 million.  In addition, the government anticipates that up to a further one million UC claimants who are in low-paid work will be required to see a work coach.  We know that people spend longer on benefits when advisers see them less frequently, so these increases in caseload need to be matched with increased investment in work coach support.

And so thirdly, most importantly, we need to ensure that as caseloads increase, those individuals are getting the support that they need.  In principle, the fact that the gap between the claimant and survey measures is closing is no bad thing – we’ve often argued that the gap was too wide, and too many unemployed people were not getting the support that they need.  But in practice, this means ensuring that claimants can access the right employment support in the right ways at the right time, whether that’s through JCP, the Work and Health Programme or other services.  So far, all of the focus in the debate on Universal Credit has been on the difficulties and delays that many people have faced in getting their awards.  But these figures should be a reminder that in the long run, the success of these reforms depends just as much on how well it supports people to get into and get on in work.


Tony Wilson

Director of Policy and Research