The Work Programme, the government’s vast welfare-to-work scheme, was supposed to be an engine of good news. It has been cited on a number of occasions by David Cameron as the shining example of radical innovation in a notoriously difficult area of policy and a firm rebuttal to the Labour charge that the government is somehow complacent about unemployment.

Billions of pounds are being made available in contracts to private and voluntary sector organisations in exchange for their expertise in placing benefit claimants in work.

Crucially, the service providers are paid by results – meaning, after a small “attachment fee”, they only get their money when their clients have jobs. People deemed harder to employ – generally those who have been out of work for longer – carry a premium. This is supposed to act as an incentive for providers to concentrate their efforts on the stubborn cohort of the long-term unemployed.  (A weakness in predecessor programmes was deemed to be that providers got paid for finding jobs for people who would have found them anyway and ignoring those who most needed help – the practice known in the industry as “parking and creaming”.)

A second aspect of the Work Programme deemed vital by government and providers is the “black box” approach. This means, in essence, that the Department for Work and Pensions won’t dictate the methods used to place people in work. Providers are meant to innovate and compete. The better formulae – the devices contained in the black box – will, in theory, succeed and their designers can then get more work and make more money. Naturally, the DWP does not (knowingly) tolerate cruel, illegal or fraudulent methods in the black box. The system is meant to drive imaginative, local solutions to a famously intractable problem.

As a theory it could all sound rather splendid: harness market forces alongside the noble ethos of the voluntary sector, underwritten by the DWP budget, to get the long-term unemployed back to work. The practice is proving tricky for a number of reasons. One is that lines of accountability are hard to police in a vast inter-locking network of different providers operating in different regions. This flaw has been exposed in the case of Jubilee crowd stewards [2] allegedly being asked to sleep under London Bridge – and foregoing wages – in order to gain experience of crowd management. The chain of command from the DWP to a prime provider to a secondary provider to an actual employer means it is hard to say what the case actually expresses about the policy. Whose bad decision was it and to what degree does that express a systemic flaw? 

The same issue is raised by recent allegations of fraud at A4e, once a major beneficiary of DWP contracts [3], although it must be pointed out that the accusations relate to bits of A4e’s past practice and not its Work Programme activity. The point is that a private company, doing work on behalf of the government, is accused of wrongdoing. Had the whole thing been run in-house at the DWP, a minister would be called to answer for it. Now there is a danger of accountability leaking through the gaps.

But by far the biggest problem is the labour market itself. As I have noted before, the Work Programme was designed, and its funding arrangements set, with an eye to fiscal and labour market forecasts from the Office for Budget Responsibility (OBR). These forecasts have all subsequently been revised in a more pessimistic direction. Even before the revisions, many observers and industry insiders expressed concern that the funding model was unrealistic.

An important reservation was that small providers – the ones most likely to actually innovate and know the job market terrain in which they work – could never manage with the kind of cash flow constraints that the DWP insisted on when negotiating contracts. So a handful of giant companies got the prime contracting work and then sub-contracted out the actual business of placing people in jobs – and the financial risk -  to smaller players, often charities. At least one charity has pulled out [4]. Others are rumoured to be on the brink.

A good account of the flaws in the model, based on past records of non-state providers meeting their targets for getting people into work, was published by the Social Market Foundation [5] in August 2011.

What is becoming increasingly clear is that the Work Programme was conceived at a time when the main problem with unemployment was thought to be difficulty in matching people to jobs, training them and motivating them to take what was on offer. Those are still issues in some areas and some cases, but much deeper structural problems with the labour market are now apparent. So too are regional variations that mean there simply aren’t vacancies to be filled.

But the feature of the labour market that seems to be causing most problems for the image of the government policy is the decline in decently-paid low- and semi-skilled jobs alongside a vast expansion of unpaid work in the guise of “experience” and “internships”. This too was the defining feature of the Jubilee steward story. For the employer (and presumably the Work Programme provider) it seemed quite reasonable to offer unpaid work as a precursor to paid work. This is well-established in the jargon as one of the “pathways” back to labour market participation. But that concept relies on the assumption that people need coaxing off a cosy life on benefits. Many are far more preoccupied by the urgent need for wages.

This too was the problem with the government-sponsored work experience scheme (not the same as the Work Programme) that caused a minor scandal last year [6]. Companies were accused of employing “slave labour” – welfare-claimants who were given to understand that their benefits would be docked if they didn’t show up. The DWP vehemently denied that such a sanction was official policy.

Defenders of the policy argued then too that “work experience” was an essential staging post on the route back to actual work. Opponents pointed out (amid more lurid claims) that the scheme was essentially providing a taxpayer subsidy for the companies that would otherwise have had to recruit people to stack shelves etc. and pay them. The government’s welfare-to-work policies are meant to match people with actual vacancies, but in the absence of demand from employers they are creating perverse incentives for people to work without wages.

It is important to disentangle two things. On one hand, there is the original ethos of a policy that emerged from many years of frustration with government’s constant inability to find work for people who were claiming benefits even when the economy was growing and, by many measures, there were jobs to be had. Second, there is long-term downward pressure on wages at the bottom end of the labour market, compounded by stagnation, a global shortage of demand, low investment, public sector cuts and only modest private sector job creation. In such conditions, the best welfare-to-work policy conceivable would run into difficulty. No wonder the Work Programme, very far from perfect, is in trouble. But even if it fails in a downturn, something very much like it will still end up being re-invented for the recovery.