Technology Spillovers and Land Use Change: Empirical Evidence from Global Agriculture

We estimate the effects of agricultural technological progress on cropland expansion at various geographical resolutions, from the country level to the world as a whole, while formally accounting for the international interdependence of national supply responses. Evidence for these effects has thus far been scant, contributing to polarized perceptions about the potential for improving agricultural technologies as a means to slow down deforestation. We find that, in most countries of the world, growth in total factor productivity (TFP) is either uncorrelated or is positively associated with cropland expansion. Yet worldwide TFP growth have been an important source of global land savings. The divergence between the country-level and the global results is explained by the changes in production patterns as countries interact in international markets. Our preferred point estimate of the elasticity of global cropland to global TFP growth is -0.34. Moreover, we estimate that satisfying food demand from 1991 to 2010 without observed TFP growth would have necessitated an additional 173 million hectares, or close to 10% of the area covered by tropical rain forests.