ABSTRACT
Economists often cite simplicity as a leading criterion for evaluating economic models. Friedman (1953: 14), for example, praises models that “ ‘explain’ much by little” in the way of context-specific detail or adjustable parameters, an idea canonized in economic methodology by frequent discussion of economic models’ so-called parsimony.1 Strangely, though, few economic models include simplicity among the inputs in any agent’s objective function, or provide clear motivation for simplicity as a normative criterion in evaluating the effect of policies on individual and aggregate behavior.
