Projecting the global economy in the long run: A dynamic general equilibrium approach.

摘要

This thesis builds on the GTAP-Dyn (Ianchovichina and McDougall, 2001) dynamic general equilibrium (GE) model of the global economy. It presents model modifications and estimates of key economic relationships in order to improve the quality of the long run projections with the GTAP-Dyn model for purposes of global environmental policy analysis. In particular, parameters governing the degree of international capital mobility and the composition of regional capital stocks and regional wealth are estimated using country panel data sets. These econometric estimates are used to parameterize the model. The behavior of household savings is also modified to stabilize net foreign assets in the long run scenarios. These new features are important for the analysis of carbon leakage due to international capital mobility, following selective imposition of carbon taxes.; Future patterns of consumer demand are an essential part of any long run economic projections. Due to its flexibility in representing changes in the consumption bundle as incomes rise, the AIDADS demand system is built into the model. Use of AIDADS is particularly important in the fast-growing, developing countries, where the composition of consumer demand is changing rapidly. To implement this demand system in the computable GE model, this thesis develops a precise calibration procedure.; Changes in patterns of consumer demand for food and forestry products drive changes in land use, which are a key part of any greenhouse gas emissions baseline scenario. In this work, the issue of land mobility across different uses is addressed via a sequence of successively more sophisticated models, beginning with a model in which land is undifferentiated and perfectly mobile, and ending with one in which land is heterogeneous and mobility across uses is governed by a nested constant elasticity of transformation function. Projected economic growth and demographics trends imply rising land rents. However, the extent of the projected increase in land rents in forestry and grazing appears excessive. This stems from several limitations of the current model, including the absence of unmanaged land and lack of forestry input-augmenting productivity growth in the forest processing sector. Addressing these limitations will be a high priority for future research.