Resource Wealth, Innovation and Growth in the Global Economy

Pietro F. Peretto, Duke University
Simone Valente, ETH Zurich


The distribution of primary resources across countries is an important determinant of trade patterns: economies endowed with natural resources that can be processed into factors of production export such resource-based commodities and import manufacturing goods from resource-poor economies. This asymmetric trade structure creates relevant interdependencies: while resource-poor economies specialize in manufacturing by force of nature, they gain from trading non-primary goods demanded by resource-rich countries specialized in primary production.

There is increasing evidence that trade variables, especially natural resource abundance and export concentration, are important determinants of economic growth. The links between resource abundance and economic performance have been studied to some extent in relation to Dutch-Disease phenomena — that is, situations in which a sudden increase in natural endowments harms economic growth and yields adverse effects on income levels in resource-rich countries — but deserve a more systematic analysis. In the last decade, several economists argued that a "resource boom" may generate a productivity slowdown because greater natural endowments induce a reallocation of labor and capital toward resource-intensive sectors: this process goes to the detriment of other sectors that exploit more intensively human capital and technological innovations and create knowledge spillovers that ultimately drive economic development. Similar reallocation mechanisms are often invoked to explain the so-called “Curse of Natural Resources” — the fact that many resource-rich countries exhibit low income levels and slow growth.

The empirical case for the resource curse hypothesis, however, is not built on firm grounds: the original studies documenting a negative correlation between natural resource abundance and economic performance identified resource abundance with the ratio of resource exports to gross domestic product. This is an imperfect proxy for physical abundance and is more a measure of specialization. According to more recent empirical work, if we measure resource abundance with stock-based indices — which are much better proxies for the size of endowments — the data strongly reject the resource curse hypothesis since resource abundance is positively correlated with growth and income levels.

The above considerations suggest that economic analysis should look more deeply into the transmission channels between resource booms and economic growth, with special regard to (i) the structure of the supply side of modern economies and (ii) the role of asymmetric international trade. We pursue this aim by building a two-country model of endogenous growth where resource-rich economies export both final goods and resource-based intermediates, and import final goods from resource-poor countries. Natural endowments are not directly consumed but exploited by resource-processing firms that sell intermediate inputs to manufacturing sectors producing final goods. In each economy, the growth process is driven by innovations that increase the productivity of the manufacturing sector. The price of resource-intensive goods heavily depends on the elasticity of global demand — the sum of domestic and foreign demands for the processed resource — which determines the consequences of endowments shocks. The fact that the primary sectors of resource-rich countries are vertically related to both domestic and foreign final sectors, in turn, implies that the elasticity of the demand for resource-based intermediates reflects the characteristics of the technology employed by final producers. A major implication is that the response of income levels and resource booms to increases in the size of resource endowments depends on whether labor and the raw resource are complements or substitutes in the production of resource-based intermediates. If they are substitutes, a resource boom generates a mild reduction in the resource price so that aggregate resource incomes — that is, resource price times extracted quantity — increases in the resource-exporting economy and the higher demand for manufacturing goods stimulates innovations and thereby enhance economic growth in the short-medium run. Resource-poor economies, in turn, will experience higher wages and positive feedback effects on growth as they export final goods to finance resource imports. If labor and the raw resource are complements, instead, a resource boom generates a strong reduction in the resource price so that aggregate resource incomes decline in the resource-rich economy, inducing a temporary slowdown in productivity growth as well as negative feedback effects on the resource-poor economy.

These conclusions suggest three questions that deserve empirical scrutiny. First, the response of employment in primary sectors to resource-endowment booms may substantially differ from the predictions of the Dutch-Disease theory: if an increase in the resource endowment yields higher resource income, this is an incentive to innovations and productivity growth in final sectors. Second, asymmetric trade matters for growth: given the existing interdependencies, it is important to investigate at the empirical level how the growth performance of resource-poor countries responds to resource booms in resource-rich economies. Third, the central role of the elasticity of substitution between resources and labor suggests analyzing in detail whether regular technological biases exist in the production process of resource-based industries.