More empirically though, what does workfare set out to achieve and is it capable of succeeding? The logic of workfare is as follows. Imagine individual A. A is unemployed and there are job opportunities available for him. However, once he accounts for all his existing income plus the benefits he receives from the government, A will only be £20 per week better off with a job than without one. A makes the decision that 25 hours a week in work is not worth an extra £20 and decides to continue claiming benefit. Contrast this with individual B. The pay-off from work will be higher for B but there are no job opportunities available.
Now, if the government introduces workfare (which it is now doing), it would follow that A and all those like him will now stop claiming benefit and take a job. While the pay-off from work will still only be an extra £20 per week, the option of remaining on benefit is no longer attractive: it will involve working the same hours for less money than what could be earned in the labour market. Meanwhile, B and his equivalents will move into workfare which could (theoretically) improve their job prospects.
In this situation, the government is able to make significant savings. Claimants like A stop claiming benefit and start paying tax, while B and co increase their prospects of finding a job.
While there are all sorts of (and rightly so) moral problems with this approach, there is the even bigger question of whether it would have any chance of working. The success of workfare depends entirely on a large group of people (the As) being voluntarily unemployed; yet in such a difficult labour market - where the vast, vast majority of people do want work - the logic of workfare just doesn't add up to reality. Instead of trying to deter people from using the welfare state, it is far better to give them what they need, whether that is personalised support, intensive counselling, training schemes or (heaven forbid) an old-fashioned benefit while they simply wait for a job to come up.