Understanding the Long Run Determinants of Land Use in Global Crop Production

Cropland cover is expected to increase in the future given the projected growth on global demand for crops. Over the past century, crop demand has largely been driven by food demands, which were, in turn driven by a combination of rising population growth and increased per capita incomes, with the former dominating. However, with population growth slowing, and incomes rising more strongly in the developing world - where dietary upgrading is a significant factor - income growth has emerged as a relatively more significant driver of total food demand. At the same time, renewable energy policies across the world have introduced a new, dynamic element to the industrial demand for cropland. However, the drivers of long run cropland use are not limited to those directly related to crop demand. New trends suggest that crop yield growth may be slowing in critical breadbaskets around the world. In addition, some agricultural lands have been lost to urbanization and soil degradation. There is also a growing interest in setting aside land for biodiversity. Finally, there is considerable interest in terrestrial carbon sequestration which could place important new demands on the world's croplands. The co-existence of so many competing factors poses a challenge for properly assessing the long-run global supply and demand for croplands in 2050. In this paper, we quantify the contributions of each of these key drivers to the long-run global cropland use using a partial equilibrium model of global agriculture. To implement the model, we also construct a global database for key land use, agricultural and socio-economic variables from several sources. Our analysis starts with validation of the model over an historical period from 2001 up to 2007-05 (5 years). The model's prediction regarding cropland use is slightly lower than the observed change (+5.8% vs. +8.3%). Decomposition of the drivers of land use change shows that income growth (+3.9%) has a larger impact than population growth (+1.6%), biofuels (+0.2%) and urbanization (-0.3%). Technological change in crop, livestock and processed food production also have a modest impact on global cropland use over this historical period (+0.8%, -0.3% and +0.1%, respectively). We then turn our attention to projections of global land use over the coming decade. Here we project continued expansion in global cropland (13.7%). We then decompose the drivers of this projected change in land use. For example, if population grows at projected rates and yields grow at recent historical rates, then projected cropland use rises by just 1.7%. If we add urbanization, then global cropland use is slightly reduced (+1.5%). Because of the magnitude of the projected production targets, adding biofuels boosts global cropland use strongly (+4.2%) when added to the preceding factors. Adding technological change in the livestock and food processing sectors, as well as growth in per capita incomes boosts projected cropland use to 13.7%. In short, income growth is likely to be the critical driver of global cropland change in the future.