Tuesday, December 21, 2010

Wherefore art thou, median voter?

Just been reading a pamphlet on attitudes to inequality in the UK by the Joseph Rowntree Foundation, published last year. It makes fascinating, if slightly frustrating reading. People in Britain have mixed, not to say incoherent views about inequality.

On the one hand, people seem relaxed about high incomes if they are 'deserved' because of hard effort or particular talent. And on the other people are often very critical of the poor if they are heavily dependent on benefits. All of this sounds pretty bad for redistribution.

However, there are some interesting caveats to this conclusion. First, respondents in the JRF research had an interesting view on what constituted a 'deserved' high income. In fact, when asked whether earnings of £150,000 were in most cases 'a fair reflection' of an individual's contribution only 28% of respondents agreed, whilst around 50% felt this was not true in most cases. This suggests many people are unaware that £150,000 is a fairly standard remuneration for high status jobs in finance, business, the law and medicine (especially in London), rather than the kind of figure mostly reserved for successful entrepreneurs or football players. Interestingly, just short of half of respondents felt £150,000 was not a morally justifiable salary, since anyone should be able to get by with less.

Second, respondents were more willing to support welfare benefits if they were more available to people on average incomes rather than being simply targeted at the very poor. There was broad support for progressive policies on tax and welfare, but resentment at those who appeared to be playing the system and not taking opportunities for work.

Finally, even though a representative sample was taken with respondents from all income groups, the vast majority regarded themselves as being in the middle of the income distribution. People seem to have no sense of where they lie in the economic hierarchy. This point is confirmed by other research, revealing widespread ignorance of what particular professions usually earn, almost always underestimating the average salaries in professional jobs. The reason for this is most likely because people know the kinds of salaries earned by people they know, and most of the time this means salaries in a similar ballpark to theirs. Therefore, middle income voters don't know that much about how much the rich make, or for that matter the poor.

This has all kinds of implications. First, a useful corrective to the Meltzer/Richard model - if median voters don't know what the income distribution is, they can't correctly estimate their net benefits from redistribution. Ironically, in a very divided society median voters may be even more ignorant about the distribution, and therefore less capable of perceiving the worth to them of redistributive policies. This could offer some insight into why the model is less successful in explaining preferences in unequal societies than in more equal ones.

Secondly, it has implications for social democratic strategy. In particular, Labour should set about making the welfare state more universalistic, to the extent that it can. The current arrangements leave the median voter pretty much out of the welfare net, but very often average earners perceive welfare recipients as relatively advantaged despite their workless status. In fact, in high cost areas people can be better off not working, and that is clearly a policy designed to irritate median voters.

Of course, this is not easy when there is no money around. But there are good reasons for reducing welfare targeting, not the least of which is people's increased willingness to pay for benefits when they think they might receive some. Increased tuition fees and removing child benefit from high-rate taxpayers move in the opposite direction - Labour could regain some popularity by promising to reverse them.