FactCheck: Is the Work Programme worse than nothing?
“These figures are the first indication that the Work Programme has had a promising start in what’s been a very difficult labour market.”
Chris Grayling, 9 July 2012
The latest unemployment figures show that the number of people out of work for more than six months has increased by 49,000 to reach 1.41 million.
The government’s answer to the problem of long-term unemployment is the Work Programme, the biggest ever welfare-to-work scheme in UK history.
Under the scheme, private companies are tasked with finding jobs for the unemployed, and are paid by results, with each “prime” provider getting up to £14,000 for everyone who stays in a job for two years.
But the biggest question at a time of chronic high unemployment has always been: is it working? The government refused to let the prime contractors publish their early performance statistics, and we were forced to rely on leaked documents to get a glimpse of how well some companies were doing.
But today ministers broke the resounding silence with the first eagerly-awaited statistical release about how the earliest entrants have fared.
The Department of Work and Pensions publication covers the first cohort of 28,600 unemployed people who joined the Work Programme when it kicked off in June last year.
It says 48 per cent of them had a break in their benefit claim at some point during their time on the programme. Some 24 per cent had a continuous break in benefits claims of 13 weeks. And 14 per cent had a continous break of 26 weeks.
So about a quarter of people who joined the Work Programme a year ago stayed off benefits for at least three months in a row.
Clearly, these numbers could be worse. A significant number of people moved off benefits in the opening months of the Work Programme.
But then we would always expect some people to come off benefits. That’s the nature of the job market. Even if the government did absolutely nothing to help them, a certain percentage of people would find work off their own bat.
The Work Programme can only be said to be “working” – or having any effect at all – if providers are doing better than this “non-intervention” rate.
We know DWP came up with a figure for the non-intervention rate, because it was used to calculate the success rates that the Work Programme providers have signed up to deliver. If the companies don’t achieve results above the baseline rate, they will struggle to make a profit.
There was some reluctance to share this magic number with FactCheck, and indeed we don’t think the department has ever published the non-intervention rate.
But the National Audit Office (NAO) did spill the beans in this report, published earlier this year.
The watchdog said DWP had told them it would have expected 28 per cent of jobseekers to find a job without help from the Work Programme. That’s an average that cuts across various groups of jobseekers of different ages and characteristics over the lifetime of the programme.
So when we learn today that only 24 per cent of the first Work Programmers have come off benefits for at least three months, that puts things in an unflattering perspective.
The government would no doubt respond that we would not be comparing like with like by putting those numbers side-by-side, and that’s perfectly true.
For a start, it’s not as if the 24 per cent mentioned in today’s release have even actually found jobs. They’ve simply stopped claiming benefits, for whatever reason. For many that reason will be failing to sign on, moving abroad or going into education or training.
ONS figures for the last three months show that on average, just under half the long-term unemployed people who came off Jobseeker’s Allowance actually did so because they found work, according to analysis by the Social Market Foundation.
That makes today’s numbers look even more pessimistic – but we should be careful not to jump to conclusions about this data.
It’s only fair to point out that the National Audit Office questioned the government’s methodology in coming up with the non-intervention rate, saying it was based on research done when the economy was in better shape.
So it could be that the benchmark has been set unrealistically high. Although if that’s the case it won’t help the government dodge accusations of failure.
If civil servants messed up when they set the non-intervention rate too high, they will have made it almost impossible for their private sector partners to make a profit, jeopardising the long-term future of the whole project.
Most importantly, the scheme has only been up and running for a year, and Work Programme providers have up to two years to work with jobseekers.
As with any newly launched project of this scale, we would expect performance to increase steadily over time, which is what happened in the case of the Flexible New Deal, Labour’s payment-by-results predecessor.
In fact, today’s data is so limited – by DWP’s own admission, in fairness – that it’s difficult to draw any concrete conclusions. We’re talking about a very early snapshot of a very small group of people, and the outcomes are not directly comparable to the first official performance indicators that will be published in the autumn.
So while we’re going to leave Mr Grayling in the middle of the FactCheckometer, we also don’t think there’s enough in this release to back up shadow work and pensions secretary Liam Byrne when he says: “It is a cut and dried case that the Work Programme is sinking under the weight of unacceptably high unemployment.”
Mr Byrne might be in for a field day later this year if the Work Programme providers are doing worse than the government’s own estimate of what would have happened if ministers had just sat on their hands. If that turns out to be the case the Work Programme will begin to look like a multibillion-pound flop.
More neutral commentators like ERSA, the industry body for the welfare-to-work industry, said today’s figures mirrored their own research into early outcomes but added: “This is a very narrow release of data from which the overall success of the Work Programme cannot be judged.”
We’re also reluctant to criticise the government for releasing some information – limited though it is – since we are among the commentators who have been calling for them to come clean.
By Patrick Worrall