Modeling Land Supply and Demand in the Long Run

The goal of this paper is to improve modeling of the supply and demand of land in the GTAP framework to enhance suitability of the model for climate change policy analysis and, specifically, analysis of green house gas emissions driven by land use and land use change. On the demand side, we begin with a dynamic general equilibrium model that predicts economic growth in each region of the world, based on exogenous projections of population, skilled and unskilled labor force and technical change. Economy-wide growth is, in turn, translated into consumer demand for specific products using an econometrically estimated, international cross-section, demand system that permits us to predict the pattern of future consumer demands across the development spectrum. This is particularly important in the fast-growing, developing countries, where the composition of consumer demand is changing rapidly. These countries also account for an increasing share of global economic growth and greenhouse gas emissions. Consumer demand is translated into derived demands for land through a set of sectoral production functions that differentiate the demand for land by Agro-Ecological Zone (AEZ). In equilibrium, supply of land by AEZ to every land-using activity adjusts to meet the derived demands for land. The paper devotes considerable attention to alternative approaches to modeling the supply of AEZ land to different land-using activities in the economy. We address the issue of land mobility across different uses via sequence of successively more sophisticated models of land supply, beginning with a model in which land is perfectly mobile and undifferentiated, and ending with one in which land mobility across uses is governed by a nested Constant Elasticity of Transformation function which also accounts for the heterogeneity of land within AEZs. Here, landowners solve a sequential revenue maximization exercise in which land is first allocated between forestry and agriculture, then between grazing and crops, and finally, amongst competing crops. The supply elasticities are consistent with econometric estimates of these parameters. We find that the most realistic representation of land supply results in baseline land rental changes in forestry and grazing that appear excessive. This likely stems from limitations of the current model, such as absence of unmanaged land and the lack of forestry input-augmenting productivity growth in forestry processing sector. Having identified these limitations, we plan to address them in future versions of this paper.