Projecting Land-Use Change in the Dynamic GTAP Framework

We developed a model that predicts patterns of land-use change at the global scale over the long run to enhance our understanding of land-use related greenhouse gas emissions. We started from a dynamic GE model that projects economic growth in each region of the world and augmented it with the most important determinants of supply and demand for land. The most important driver of the demand for land is consumer demand for food and forestry products, which we model with econometrically estimated AIDADS demand system. To account for the influence of technology on the demand for land, recent empirical estimates and forecasts of technological change in the agricultural sectors are included into the model. Technical change in forestry processing sectors is another key factor in determining demand for land used in forestry sector and outcomes of the competition for land between forestry and agriculture, and this is determined through the soft link with the Global Timber Model. To capture intensification in land using and processed food sectors, we introduce various substitution possibilities in production that reduce derived demand for land. The implications of alternative assumptions made with respect to the specification of land supply are explored through a sequence of successively more sophisticated models of land supply. We begin from very simple model where land endowment is fixed, homogenous and perfectly mobile, and end with one which realizes that land endowments are heterogeneous, where mobility of land across uses is restricted by a nested Constant Elasticity of Transformation (CET) function within each land type, and land can be expanded by converting unmanaged land into production land. The latter is modeled as an investment decision whereby new land is accessed only when present value of returns on land in a given region is high enough to cover the costs of accessing the new land. We conclude the paper with comparison of the model results against historical trends.