EAERE/FEEM/VIU European Summer School

Applications of Computable General Equilibrium
Models in Environmental Economics

Lecture Notes by Thomas Rutherford

June 25 to July 1, 2006

LECTURE 1: Decomposing the Integrated Assessment of Climate Change

This lecture deals with integrated assessment of climate policy measures, i.e. models in which both economic and climate mechanisms are incorporated to provide a framework for cost-benefit assessment of climate policy. In this lecture, I go through a decomposition approach for integrated models which is based on a linear approximation of the climate system. In our formulation the economic and natural science components are processed independently on different time scales. Turnpike properties of the Ramsey growth model can be exploited to provide a precise representation of post-terminal emissions and to reduce the economic horizon required to accurately approximate transition paths. Germaine to the economic assessment of climate policies, our decomposition accommodates formulation of the economic model in a complementarity format and thereby provides a means of incorporating second-best effects that are not easily represented in an optimization model.

Lecture notes are provided here.

Files for the associated introductory computing exercises are provided in this zip archive.

The paper which provids the basis for this lecture can be downloaded from


and related computer progams (in GAMS) are available from http://www.mpsge.org/mainpage/dicemodel.zip

LECTURE 2: Stochastic Programming in a Complementarity Format: Programming Tools and their Application to Climate Policy Design

This lecture introduces a new set of programming tools for representing stochastic equilibrium problems in GAMS. I outline the logic of stochastic optimization with recourse and show how these ideas are extended to complementarity programming. I illustrate this problem class through a model which explores the tradeoff between technology subsidies and carbon taxes in climate policy. Lecture notes describing the computational tools are provided here.

Files for the associated introductory computation exercises are provided in this zip archive. The file climate.gms represents an extension of the introductory climate model from Lecture 1 to a stochastic framework.

Lecture notes for an application of these techniques to climate policy are provided here.

The directories include GNUPLOT plotting routines for which source code and extensive documentation may be downloaded from gnuplot.sourceforge.net.