Learning Temporal Preferences

Abstract

We analyse households' responses to an unanticipated change in consumption opportunities and evaluate their implications for the nature and formation of preferences. We study the tariff experiment conducted by South Central Bell where local telephone measured tariffs were introduced for the first time in Louisville, KY. Households were given the choice to remain in a flat rate scheme or switch to the new measured tariff scheme. The results of the analysis support models where consumers react to a change in the environment in the direction predicted by theories of rational investment in information. Households learn rapidly to undertake optimal decisions, and react to potential savings of seemingly small magnitude, typically about $5.00 per month. We find no support for models where consumers' responses are determined by inertia or impulsiveness, including systematic tendencies to undervalue future wants common to models of hyperbolic discounting. From a methodological viewpoint, the analysis shows how the appropriate treatment of predetermined endogenous variables and state dependence turns out to be crucial for interpreting the data.

Keywords: Learning, rationality, dynamic discrete choice panel data analysis, dynamic consistency

JEL Classification: C23, C25, D83, D90