The applied (or computable) general equilibrium (CGE) model is one of today's standard tools of policy analysis. As with all economic policy tools and prescription drugs, its use requires great caution. Nevertheless, in all cases where the repercussions of proposed policies are widespread, there is no real alternative to CGE. If it is not used explicitly, the tools used will contain implicit implications for remote implications, if only to deny them.

The ability to deploy CGE models is the outcome of research going back at least 130 years and involving very disparate lines of inquiry. Economic theory and the vastly improved availability of economic data have played basic roles. But other research inputs have been equally crucial: improvements in computing power and the development of algorithms for computing equilibria. The decisive step in the last direction has been the pioneering work of Herbert Scarf. If one examined a time series of development and publication of applied general equilibrium models, I am sure that there would be a marked régime change following Scarf's paper (1967) and especially his monograph (with the collaboration of Terje Hansen) (1973).

Let me give a partial account of and reflections on the development of applied general equilibrium models. This is not a true scholarly account but relies primarily on my own impressions over the years. It originated as an after-dinner speech and should be regarded as a written version of one.