New Vehicle Feebates

New vehicle feebate programs encourage improved fleet-wide vehicle fuel efficiency; yet analyses of these policies have been limited to ad hoc proposals. In this paper, we exploit an extensive, multi-year dataset which includes more than 16 million observations to evaluate the welfare implications of a long-standing vehicle feebate program in the Canadian province of Ontario. We (1) show that second-best optimal feebates can be written as a function of new vehicle Pigouvian taxes; (2) find that Ontario's feebate program was welfare-enhancing relative to a no feebate scenario but that a second-best optimal benchmark would have yielded additional welfare while reducing fleet-wide emissions; and (3) that Ontarian consumers responded asymmetrically to fees versus rebates