Attitudes towards redistribution

 

This book by Charlotte Cavaillé is forthcoming, so this post is based on this excellent podcast, which is well worth an hour or so of your time (and/or buy the book when it’s out).

Although inequality can be measured in many ways, here I want to focus on one particular measure: the share of income going to those at the top of the income distribution (1% or 0.1%). In the UK this started rising from the early 1980s to the mid-2000s, but it hasn’t risen significantly since then. (It rose from about 6% to around 15% for the 1%, and about 2% to around 6% for the 0.1%. Figures from this IFS paper, discussed here.) Yet when people are asked whether (see figure 2) “government should redistribute income from the better-off to those who are less well off”, the percentage saying yes (about 50%) is much the same today as it was in the early 80s. Admittedly this question doesn’t specify who the ‘better off’ are (more on this latter), but nevertheless the combination of growing inequality at the top with unchanged views on redistribution is interesting.

Here I found Cavaillé’s framework for thinking about attitudes to inequality very helpful. The first point is that if views about redistribution from the top 1% were governed by self-interest alone, the proportion wanting more redistribution should be 99%. However Cavaillé argues that attitudes to redistribution are governed only partly by self-interest but also by views about fairness. We are a social creature after all, rather than just individualists out for ourselves. Furthermore she argues that when it comes to redistribution, views about fairness are divided into two: “redistribution from” and “redistribution to”.

This might seem counter intuitive when thinking about a survey question that combines both aspects. But if you think about it, redistribution does involve two acts: taking away (redistribution from) and giving (redistribution to). Cavaillé convincingly argues that the way most people think about fairness when thinking about taking away is rather different to fairness when giving to others. In general terms, different attitudes about ‘redistribution from’ tend to go along economic left/right lines, but attitudes about ‘redistribution to’ are more correlated with socially liberal or conservative mindsets.

To be more concrete, experimental evidence from behavioural economics suggests the dominant idea concerning ‘redistribution from’ is proportionality: have those earning higher incomes earned (in a moral sense) those better rewards? Proportionality is used by people far more often than notions of equality. Whether incomes under capitalism are deserved or not will be correlated with where people are on an economic left/right spectrum, but they are also influenced by elite discourse about the extent to which rewards are justified.

In contrast, notions of fairness concerning ‘redistribution to’ involve social solidarity and free riding. Again in behavioural economics experiments ideas of reciprocity (help others until they start to free ride) dominate concepts of need. Cavaillé suggests that social liberals tend to be more optimistic about those who are in receipt of redistribution and welfare, while social conservatives obsess more about free riding, and are unconvinced that the state can prevent this.

Cavaillé uses these ideas to explain changes in attitudes in a number of countries, including why support in the UK for redistribution has fallen or at best stayed constant while incomes at the top have risen so dramatically. First she points out that following Thatcher’s election victory in 1979, the debate about whether incomes produced by the UK’s capitalist system were fair or not, a debate that had been prevalent in the 1960s and 70s, largely disappeared. Instead dominant narratives became about wealth creation and incentives, both of which were generally and selectively used to refer to those earning high incomes. This was continued under Blair, who was famously relaxed about high incomes.

This meant that attitudes to redistribution shifted from thinking about ‘redistribution from’ to thinking about ‘redistribution to’, particularly under the Blair/Brown government where social support for the poorest increased substantially. This was a gift to the political right, and particularly to the right wing press, which produced endless stories about scroungers sponging off the welfare state. This helped to make attitudes towards redistribution more unfavourable in the first decade of this century. To put it another way, the left wing social conservative, whose views on redistribution would always be conflicted, thought more about ‘redistribution to’ and free riding, even though inequality at the top was rising.

Still, doesn’t self interest count for something? How much it counts for depends a lot on information. Better off social liberals may often be in favour of redistribution until the moment they realise how much their taxes will need to increase! I would argue that systematic information and debate about top incomes is very thin on the ground, and in particular is unlikely to reach the less well informed who are often left wing social conservatives. (There is a strong positive correlation between the amount of education people have received and social liberalism.) In particular, very few people realise how much they have become personally poorer as a result of the growing incomes of the 1% (assuming, as seems reasonable as a first approximation, that this is a zero-sum game). To put it simply, if today the 1% get nearly an extra 10% of national income compared to the post-war period, then the 99% have on average 10% less income.

At this point we need to address the problem that survey questions talking about rich and poor, although they provide useful information about changing attitudes over time, may be too general to pick up views about the very well off: the top 1% and especially the top 0.1%. After all, even the right wing press carries stories about ‘fat cats’, even if they tend to be more about those in the public rather than private sectors. In this recent opinion poll, for example, 66% of voters say that the wealthy do not pay their fair share of tax, relative to just 6% who say they pay too much. Using the term ‘wealthy’ rather than ‘better-off’ may tap better into views about the top 1%, but note also that this is only a question about ‘redistribution from’, and avoids talking about where any extra tax might go.

Let me summarise by using this analysis to suggest what those (like myself) favouring greater redistribution from the top 1% need to do to convince others. The first thing is to focus on the very top of the income distribution, and be explicit about how much the rise in income going to the 1% has made everyone poorer. The more information people have, the more self interest will kick in. Along the same lines, stress that greater incomes for the 1% have been accompanied by lower, not higher, growth rates. Second, stress that CEO pay is not determined ‘by the market’ (which might make some believe it reflects effort or contribution), but is instead set by other CEOs or well paid executives and board members. Finally, when asked about ‘redistribution to’ (as will inevitably happen), focus on areas of public spending where there is less perceived scope for free riding, like the NHS. Some of this is intuitive, and probably bread and butter for those who campaign on this issue, but I found it useful to see how these lessons follow straightforwardly from Cavaillé’s framework.