Ed Miliband and Liam Byrne want to get tough on 'evil' benefit scroungers. This is not evidence-based policy making.

Let's look at the evidence, in the form of table 12g of the ONS's survey of subjective well-being.

This shows that the unemployed are significantly less happy than those in work. In answer to the question 'overall, how satisfied are you with your life nowadays?' the average unemployed person replied 6.3 on a 0-10 scale, whereas the average employed person replied 7.5.

This is a big difference - equivalent to six standard errors. It means the unemployed are less happy than people who are divorced, widowed or who have a longstanding illness or disability.

The overwhelming majority of unemployed people, therefore, far from being content to scrounge, are pretty miserable. Any leftist party worth anything would have this as it basic premise.

In this context, the possibility that some people don't want to work should actually be welcomed by any utilitarian. The fact is that mass unemployment is probably here to stay. If a few people don't want to work, therefore, this increases the chances of other unemployed people finding a job, and thus moving out of deep unhappiness.

Let's look at this another way. Let's assume that a million 'scroungers' were to move into work; this is an utterly incredible assumption, given that there are only 455,000 job vacancies, and given that there's no evidence that there are so many scroungers. But bear with me. By how much would this raise GDP?

If they were all to get full-time minimum wage jobs - earning ''11,000 per year - aggregate labour income would rise by ''11bn - less than 0.8% of GDP. Add in the fact that it's profitable to employ them (ex hypothesi) and we're talking a rise in GDP of around one per cent.

But this is puny compared to loss in GDP caused by the financial crisis. Andrew Haldane at the Bank of England estimated (pdf) this to be 10%. And it's looking unlikely that we'll recoup this through faster growth any time soon.

The cost of 'scroungers', then, is an order of magnitude smaller than the cost of bankers.

The evidence, then, suggests that 'scroungers' are, at worst, a mild problem. So why expend political capital fretting about them? The nicest thing I can think of is that this is another example of how moralizing is displacing rational policy-making.