Is There A Resource Curse?

Liberia is in the news as Chevron has announced plans to begin oil exploration there later this year. That has led to predictable (and quite reasonable) worries about the country succumbing to the so-called resource curse, in which natural resource development ultimately hurts, rather than helps, the country involved.

The news reminded me of a paper published earlier this year in the American Political Science Review that I’ve been meaning to blog about. In “Do Natural Resources Fuel Authoritarianism? A Reappraisal of the Resource Curse”, Stephen Haber and Victor Menaldo argue against conventional wisdom on the subject. Specifically, they argue that resource extraction does not make democracy less likely. Indeed many variations of their statistical analysis lead to “results that suggest a resource blessing”.

The authors appear to identify at least two problems with past analysis that explain why they come to different conclusions. Most empirical studies of the resource curse work by comparing large sets of countries. Those studies need to control for a variety of variables, like income and demographics, in order to isolate the impact of resource wealth on political institutions. But things like income surely aren’t independent of resource development. That invariably makes the statistical analysis fraught.

Most interestingly, though, the authors point out that resource development is not only a “cause” of institutional development – it’s a consequence. States do not develop resources simply because they’re there – they develop them because political decisions are made to do so and because the business environment is right. It’s quite plausible to believe that authoritarian states are more likely to go full tilt into developing their natural resource than democracies are in the first place. (That might be because governments need the rents to pay off their citizens, or because public opposition can’t get in the way.) If that’s true, though, one may find a correlation between resource development and autocracy that doesn’t actually reflect a resource curse.

In order to cut through these problems, the authors develop extensive histories of the countries they study. Doing that, rather than simply looking at cross-country snapshots, allows them to do a better job of sorting out cause from effect. That ultimately leads them to reject the resource curse hypothesis, and to conclude that a resource blessing often results. I can’t say that I’ve worked carefully enough through their (quite sophisticated) statistical techniques to know whether I ultimately agree with them, but the work is intriguing regardless.

So what does this all mean for Liberia? Two things, I think. Liberia scores in the middle of the pack on most Freedom House scores (though its anticorruption and transparency scores, which are particularly relevant to the resource sector, are weak). If it can get better before an oil boom comes on in full force, resource extraction might reinforce that trend; if it can’t, all bets are off. Second, Haber and Menaldo’s analysis is statistical: while it might say that on average there isn’t a resource curse, that should be little reassurance for any particular country that’s diving into extraction. Liberian policymakers would be wise to get the right institutions in place to protect their country and people from bad outcomes.