Monday, April 14, 2008

Not a Puzzle

Re the previous post, Peter commented:

Very excellent post on income inequality. But it is not a puzzle unless you think that some general sense of opinion polls translates into policy. But we know it does not. Most poeple want gun control in the US and we don't get it. So there is a big intermediate variable between general opinions and policy outputs, or rather a set of variables clumped under the label political processes: political parties, institutions, the rules of elections, like primaries and superdelegates, political resources, the role of other issues , campaign contributions, media control, lobbying etc. These processes refract opinions through a prism which shapes the outcomes. On those grounds, it is not surprising that public opinion worries about inequality but not much is done about it. We see this appearing in all sorts of issues like health care, etc. This is why I discount all findings and arguments that start with "public opinion says..." There are too many steps in between. Researchers have been comparing countries by looking at the connections between political institutions ( like the electoral law, or the legislature), interest groups and policy outputs. There is a lot to be learned there that helps resolve the puzzle!


Peter is of course correct. When I wrote the original post, in fact, I thought of asking readers to propose their own solutions to the puzzle: what aspects of our respective political systems do you think account for the different outcomes regarding income inequality? But there is even more to it than that. The survey itself does not really get below the surface of attitudes toward inequality. Other surveys would yield different views of public attitudes in the US and France. For instance, hostility to specific redistributive measures such as the estate tax is higher in the US. So even before we get to institutional differences, we can question the premise that public opinion is quite similar. Another blog post yesterday illustrates the way in which the framing of questions about "fairness" (a value related to "equality") can elicit quite different attitudes from the same population: here is Mark Thoma commenting on Kwame Anthony Appiah. The data cited by Kenworthy can thus be viewed as asking "which income distribution do you regard as fairest?" But if the researcher were then to ask, "Do you think it would be fair to raise marginal income tax rates to 80 percent on those making more than X in order to achieve the distribution you regard as fair?" conflicted feelings would rapidly emerge. So Peter's remark could be extended as follows: "Outcomes are shaped by institutions, but what shapes institutions? Might it perhaps be that in every society there are certain contradictory opinions about what is desirable, and that in the interest of stability institutions evolve (or are designed) in such a way as to keep value conflicts off the table as much as possible?"

No comments: