Should Rules Be Simple

Working Paper: CEPR ID: DP515

Authors: Paul Levine

Abstract: The principal argument of the paper is that in an incomplete information setting, where the private sector lacks information on goverment objectives and has to learn about the policy rule by direct observation and estimation, simple `sub-optimal' rules may outperform the more complicated rule that is optimal under complete information. This result is demonstrated by simulations using an overlapping contract rational expectations model. The paper thus provides some formal reasoning to support arguments for simplicity associated with credibility and the need for the private sector to be able to monitor policy.

Keywords: simple rules; credibility; learning; rational expectations

JEL Codes: 023; 026; 321


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
precommitment (D70)better outcomes (I14)
discretion (Y60)worse outcomes (I14)
private sector's learning about policy rules (L52)economic outcomes (F61)
incomplete information (D89)simple rules outperform complex rules (C52)
private sector learning (J24)performance of simple rules (C52)
initially optimal rule becomes complex (C61)performance of simple rules improves (C52)
simple rules (Y20)better welfare outcomes (I30)
private sector's perception of government behavior (H11)credibility of rules (C70)

Back to index