We should welcome this experiment with workfare

David Cameron has announced that Jobseekers who do not find sustainable work after they have been through the Government’s new welfare-to-work scheme, the Work Programme, will be made to do compulsory work in the community for six months. "Workfare" is what people normally call such schemes.

This is welcome news – and something I've been arguing for for ages.

It should be popular, too – our polls at Policy Exchange show that the public agree by a margin of six to one (80%-13%) with the idea that "people who have been out of work for 12 months or more, who are physically and mentally capable of undertaking a job, should be required to do community work in return for their state benefits."

In fact this experiment is less than that statement, as claimants unemployed by the end of their time on the Work Programme will in general have been out of work for as much as for as much as three years. (Some groups, like ex-prisoners, will have been referred to the work programme faster.)

It is commendable that the Government is going to pilot this scheme first, since evidence from the US has highlighted the advantages of localised testing in leading to successful policy innovation. One problem in Britain's political system is that we tend to perform experiments not locally, but on the nation as a whole.

Curiously the announcement seems to imply an automatic national rollout in 2013. That reflects a general habit in British government of assuming that new ideas are going to work. This undermines the purpose of the pilots – if it’s going to be done regardless of the outcome, why bother?

It is clear that workfare can be very effective – for example, a workfare component was involved in the US reforms – leading to welfare rolls falling as much as 94 percent. Those with attitudinal problems, people who ‘cycle’ in and out of work or those without work experience would surely benefit.

But the truth is we don’t know for whether it would be right for everyone. Given the potential cost of these schemes (though these are often exaggerated) it makes sense to test what is effective and what isn’t.

In this case quite a number of people may end up being part of this experiment. The minimum performance standard set for getting claimants into sustainable employment through the Work Programme is only around a third by 2013 (varying by group) and we may see up to a million claimants transitioning onto the Work Programme from Incapacity Benefit).

A full randomised control trial, building on evidence provided by the Employment Retention and Advancement trials run by the previous government, followed by a detailed, comprehensive evaluation, would be the best way.

There are other pitfalls with such schemes – in other countries such as Australia, governments have found that claimants can get comfortable in their ‘workfare’ roles and lose the motivation to leave and seek jobs in the real labour market. We also have to be careful to design such scheme so as not to displace existing unskilled and semi-skilled employment.

Some will claim that workfare bullies people out of the system. But as Polly Toynbee has written (admittedly a little while ago) in an article entitled ‘Workfare really works’:

More likely, many were claiming falsely. Either they already had full-time jobs paying them above benefit levels (we are not talking here about earning a little extra on the side) or they were well able to get jobs once pushed. The Low Pay Unit complains that many have been pushed into unsuitable work, but after two years, is that so unreasonable?

Quite. Workfare has huge potential to reduce unemployment. Advocates like me have to admit that we don't know as much about it as we would like, which is why we need to start with experiments.