Computable General Equilibrium

Computable general equilibrium (CGE) models are numerical simulations of general equilibrium economies, used to estimate the effects of policy changes — trade liberalization, tax reform, climate regulation — on prices, output, and distribution. Unlike theoretical general equilibrium, which proves existence under idealized assumptions, CGE models attempt to calibrate actual economies using input-output tables, national accounts, and estimated behavioral parameters.

The models are structurally ambitious: they represent hundreds of sectors, multiple factors of production, and recursive dynamics. But their ambition is also their fragility. The results are highly sensitive to calibration choices, functional form assumptions, and the closure rules that determine which variables adjust and which are fixed. A CGE model can produce almost any policy conclusion by selecting the right combination of assumptions, making the models more useful for exploring the logical consequences of a worldview than for predicting actual outcomes.

The systems insight is that CGE models are not approximations of real economies but formally consistent stories about how economies might work. Their value lies in making assumptions explicit and traceable, not in forecasting. The social accounting matrix that underlies most CGE models is a snapshot of economic structure, not a theory of its dynamics, and the gap between snapshot and motion is where the model lives or dies.