Abstract

We examine the impact of agricultural productivity growth on deforestation in the tropics. We first develop a dynamic model of forest conversion to farmlands, in which conversion is costly and market potential for the agricultural output is limited – both of which affect the amount of land and fertilizer used. The model emphasizes that it is the growth of agricultural productivity, rather than its absolute level, that determines deforestation patterns. It predicts that a decline in fertilizer price growth (i.e. an increase in agricultural productivity growth) leads to increased deforestation when market potential is high, and to decreased deforestation when market potential is low, providing clarity on a key debate (the Jevons vs Borlaug hypothesis). Second, we test the theoretical predictions using forest loss data spanning the entire tropics at a spatial resolution of 0.5 x 0.5 degrees between 2000 and 2022. Combining information on soil characteristics and global variations in fertilizer prices, we identify local exogenous changes in fertilizer costs. Our findings indicate a negative correlation between fertilizer price growth and deforestation, particularly in areas with high market potential. We find no evidence of a positive relationship, even in areas with low market potential. Additionally, we find no evidence that protected areas mitigate the negative relationship between fertilizer price growth and deforestation.

Co-authors : M. Couttenier and S. Desbureaux

Practical information