Disaggregating the resource curse: Is the curse more difficult to dispel in oil states than in mineral states?

The hypothesis of the natural resource curse has captivated the economics profession, and since the mid-1990s has generated a large body of policymaking initiatives aimed at dispelling the curse. In this paper, we evaluate how the effect of resource abundance on economic growth has changed since these policies were first introduced by comparing the periods 1970-89 and 1996-2008. We disaggregate resources into oil, gas, coal, and nonfuel mineral resources, and find that disaggregation unmasks diverse effects of resources on concurrent economic and institutional outcomes, as well as on the ability of countries to transform their economic and institutional infrastructure. We consider resource dependence and institutional quality as two channels linking resource abundance to economic growth in the context of an instrumental variables (IV) model. In addition to exploring these channels, the IV framework enables us to test for the endogeneity of the measures of resource dependence and institutional quality in the growth regressions, paying particular attention to the weakness of the instruments.