The “impossible precision” of regulation: How policies dynamically interact with people heterogeneity and learning

Margherita Saraceno (Università degli Studi di Pavia)
Giorgio Rampa (University of Pavia)

Abstract

The present dynamic Bayesian-learning model frames the problem of a policymaker that, given a socially optimal goal to pursue, hardly achieves its goal with precision because regulation effects results from a complex interaction between regulation and individual beliefs, compliance costs, and subsequent individual choices. Showing how heterogeneous agents decide –time by time– whether to comply with given rules based on their conjectures, the available information, and their private costs, we study the possible conjectural equilibria and prove that the policymaker can pursuit optimality by acting on various structural parameters of the model (corresponding to various kind of regulatory interventions) to align the conjectural equilibrium level of welfare losses to the optimal one. However, the policymaker must be aware that any kind of regulatory intervention implies dynamical effects that depend on both people’s conjectures and learning staying behind individual choices. Finally the precise achievement of any policy goal cannot be taken for granted, while requiring time and perseverance.

Download the file