Resource and Energy Economics is has a manuscript on the economic growth associated with natural resource rich areas:

James, A. & Aadland, D., 2010, The curse of natural resources: An empirical investigation of US counties, Resource and Energy Economics.

From the abstract:

Research consistently shows that natural resource dependence tends to be associated with lower economic growth. However, the studies typically focus on differences across nations or states. We fill a gap in the literature by testing the so-called resource curse at a more disaggregated county level. Our results show clear evidence that resource-dependent counties exhibit more anemic economic growth, even after controlling for state-specific effects, socio-demographic differences, initial income, and spatial correlation. A case study analysis of Maine and Wyoming, and the counties within, highlight the growth effects of specializing in natural resource extraction.

I like the idea they have here, of testing counties, but I have some problems with the execution. They argue that looking at counties lessens the problems associated with differences in state regimes, which is true–there are over 3,000 counties.

They cover several theories on the resource curse, including “Dutch Disease”–where specialization in sectors unrelated to manufacturing minimizes that sectors’ growth potential via returns to scale and production externalities. There’s also what you might call the ‘trustafarian’ idea, where resource endowments cause people and governments to be overconfident in their economic returns, so they do not invest other sectors or human capital to the degree they should.

The authors also briefly bring up institutional differences, include a weak discussion of the role of civil strife. Here is where I have problems.
In other fields, such as sociology, the “curse of oil” or diamonds hinges on how the resources tie the economy to past colonial economic arrangements (with global corporations instead of imperial governors), concentrating wealth into privileged classes (often market-dominant ethnic majorities) who then use control over the military to enforce their economic hegemony. This cycle systematically impoverishes the remainder of the population and suppresses other possible avenues of economic growth.

It’s for me hard to tell what effect James and Adland really find here with their association between lower growth rates in resource dependent counties in the US in terms of testing these theories. We don’t really have an explanation here; it seems quite clear to me that in the presence of property regimes, different sorts of resources have different returns, and there’s no reason to believe that value or profit stays in the place it is created: ranch owners are corporate and asset wealthy; ranch workers are most assuredly neither.

I’m also not sure that we can treat agriculture as a natural resource endowment in the way diamonds and oil are.