Inequality and Growth

“Tyler Cowen”:http://www.marginalrevolution.com/marginalrevolution/2007/07/is-economic-ine.html points to a very interesting new paper by Daron Acemoglu and his colleagues (PDF – it was “here”:http://econ-www.mit.edu/faculty/download_pdf.php?id=1510 this morning, but this link isn’t working for me any more; see “here”:http://www.isop.ucla.edu/cms/files/acemoglu_bautista_querubin_robinson.pdf for a slightly earlier version) on the relationship between political and economic inequality. Tyler’s gloss is that this provides general insights into the “meme” of whether economic inequality is bad for growth, and concludes that “at least from that data set, the real problem seems to be rent-seeking behavior through the political process.” Thus, unless I misunderstand him (which is possible; he may just be blogging in shorthand), he is saying that this paper provides significant evidence suggesting that economic inequality isn’t the cause of slower economic growth; instead, political inequality and rent-seeking are at fault.

This, however, isn’t how I read the paper. Acemoglu and his colleagues seem to me to be quite careful _not_ to extrapolate broad conclusions about the general relationship between inequality and economic growth from their findings (they do believe that there are some important implications for the development literature, but that’s not quite the same thing). Instead, what they argue is that _under conditions of weak institutionalization_, some degree of economic inequality may be a good thing. Roughly speaking, their causal argument is that under these conditions powerful land-owners may serve as a counterweight to politicians who have access to the powers of a repressive state, preventing the latter from grabbing as much property to themselves as they would be able to otherwise. When economic inequality is lower, power is diffused among many actors, who can’t organize against the state as successfully because of collective action problems (here they borrow from Robert Bates’ research on Kenya).

This is a very interesting, and potentially very important result for the study of development, if it’s borne out elsewhere. However, its implications for the general meme that Tyler is talking about seems to be more limited. It’s likely to tell us interesting things about what is happening in countries where there is a more or less naked power struggle going on between different interests to define property rights etc. However, on the authors’ own interpretation, its relevance to countries where these struggles are more sublimated is at best uncertain. Acemoglu and his co-authors specifically distinguish their findings from previous work on the nineteenth century US South, where there _is_ evidence of a strong link between economic inequality and adverse outcomes. They suggest that the key difference between the two cases is their respective levels of political development (institutions in the South did place some constraints on what politicians could do that weren’t present in Columbia, thus vitiating the need for a countervailing force against the state).

In contrast, the relationship between inequality and economic outcomes appears to be different in strongly institutionalized environments such as the United States. Here, political institutions place certain constrains on politicians so that having an independent landed elite is not necessary as a check against politicians and does not necessarily create a tendency for better outcomes. Rather, and possibly consistent with the US evidence, in such environments greater inequality may have negative economic or political consequences (for example, in the extreme via “political capture,” capture,” as in Acemoglu and Robinson, 2006b, or Acemoglu, Ticchi and Vindigni, 2006) [quote from earlier draft – will update tomorrow if nec.]

Thus, the paper is really an argument about what we are likely to find in countries with more-or-less openly predatory states, not about economic inequality’s more general effects for growth. Regardless, this is an interesting and potentially very important paper that’s well worth reading.

[Crossposted from “Political Science Weblog”:http://www.henryfarrell.net/polsci/]