This issue of Climatic Change includes a valuable contribution from William U. Chandler analyzing policies for limiting carbon dioxide (COz) emissions in China. In his analysis, Chandler employs the global energy/CO z model developed by Edmonds and Reilly (ER) at the Institute for Energy Analysis of Oak Ridge Associated Universities (IEA/ORAU).* The ER model has been applied extensively in numerous studies assessing the impact of different policies on future greenhouse warming, and it has emerged as the leading analytical tool for such purposes.** Recently the model has become available in compact form for personal computers, which will likely spread its use yet further. Nevertheless, despite the wide popularity and apparent flexibility of the ER model, it has certain structural characteristics that limit the nature and range of policies it can analyze. Researchers should be aware of this limitation because it happens to preclude comparative economic analysis of some very promising CO z abatement opportunities.
The Edmonds-ReiUy (ER) model is a partial equilibrium model that utilizes classical economics to balance energy demand and supply at market-clearing prices. Hence demand for energy services is determined endogenously, which is an important improvement over earlier global energy models.*** However, this does not mean that supply and demand are given equal emphasis in the model. The supply side in ER is specified in considerable detail, permitting fair economic competition among different supply options, including hydroelectric, nuclear, fossil fuels of different grades, solar photovoltalc, synfuels, etc., and the model then computes the CO z consequences of whatever particular supply mix results. In contrast, the demand side is not disaggregated at all for non-OECD countries (including China), and is disaggregated into only 3 gross sector levels for OECD countries (residential/commercial, industrial, and transport). This structure gives the model a number of fundamental limitations. First, it is impossible to