The Conceptual Flaw in the Microeconomic Foundations of Dynamic Stochastic General Equilibrium Models

First-generation dynamic stochastic general equilibrium (DSGE) models have been criticized for their lack of financial markets but, more perceptively, for their barter properties. This note explains why the second of these criticisms is fundamental. All DSGE models are built on frictionless, perfect barter, Walrasian microeconomic foundations. Introducing money and banks into such models converts them into a ‘friction’ contra the fundamental principle that monetary exchange is more efficient than barter. This insoluble difficulty with the microeconomic foundations of DSGE models arises because theorists ignore the Hahn problem that applies to all monetary models based on Walrasian general equilibrium (GE) microeconomic foundations. The Hahn problem reveals three things. First, a perfect barter GE solution always exists in any ‘monetary’ model erected on Walrasian GE microeconomic foundations. Second, inessential monetary features are easily attached to perfect barter microeconomic foundations but as easily removed, leaving the perfect barter solution intact. Third, attaching such inessential additions leads to logical error; the misuse of language that produces invalid conclusions. A second-generation DSGE model that is intended to increase understanding of financial crises is then examined to show that it suffers from the Hahn problem; it converts banking and financial markets into ‘frictions’, and words and economic concepts take on different meanings. That renders the new DSGE model impossible to interpret or use as a basis for advice on monetary policy.