Resource Discoveries, Learning, and National Income Accounting

Questions about the ultimate size of mineral and energy resource endowments and the degree of fiscal prudence which should be exercised by countries engaged in resource extraction have become central for many developing countries during the recent resource boom. To explore these questions, this paper develops a model of optimal resource extraction and discovery that combines two polar assumptions: (i) that discovering a resource today drives up the cost of future resource discoveries, and (ii) that extracting resources yields knowledge that reduces the cost of discovery. Although the model shows that resource discoveries should be valued at marginal discovery cost in measures of national saving and income, the ultimate size of the resource that can be exploited is the result of the interplay between rising discovery costs and accumulating knowledge. Empirical tests of the model show that the resulting income estimates would be extremely volatile for many extractive economies, owing to the lumpiness of resource discoveries. Two alternative accounting approaches, based on Hicksian concepts, yield more intuitive and less volatile income estimates. The question of fiscal prudence for extractive economies hinges on how optimistic countries are about the risks in future mineral and energy markets, and how far into the future these countries are willing to project optimistic trends when making decisions about how much to consume and how much to save of current resource revenues.