The Economics of Attribute-Based Regulation: Theory and Evidence from Fuel-Economy Standards

This paper analyzes "attribute-based regulations," in which regulatory compliance depends upon some secondary attribute that is not the intended target of the regulation. For example, in many countries fuel-economy standards mandate that vehicles have a certain fuel economy, but heavier or larger vehicles are allowed to meet a lower standard. Such policies create perverse incentives to distort the attribute upon which compliance depends. We develop a theoretical framework to predict how actors will respond to attribute-based regulations and to characterize the welfare implications of these responses. To test our theoretical predictions, we exploit quasi-experimental variation in Japanese fuel economy regulations, under which fuel-economy targets are downward-sloping step functions of vehicle weight. Our bunching analysis reveals large distortions to vehicle weight induced by the policy. We then leverage panel data on vehicle redesigns to empirically investigate the welfare implications of attribute-basing, including both potential benefits and likely costs. This latter analysis concerns a "double notched" policy; vehicles are eligible for an incentive if they are above a step function in the two-dimensional fuel economy by weight space. We develop a procedure for analyzing the response to such policies that is new to the literature.