Electric Vehicle Subsidies: Cost-Effectiveness and Emission Reductions

Résumé

This paper studies the environmental performance of electric vehicle subsidy programs in Canada. I leverage changes in the provincial-level subsidies to study their short-run impact on sales and charging station deployment using a natural experiment setting. My findings suggest that subsidies are very effective at increasing electric vehicle adoption, but failed to induce additional charging station installations in the short-run. I rely on a structural estimation of the demand for cars and the supply of charging stations to evaluate the environmental impact of subsidies. My results suggests that Canadian rebate programs led to an increase in adoption of 93%, and an increase in the size of the charging station network of 19%. I take these results as additional evidence of weak network effects. I propose a unified framework to conduct a cost-benefit analysis. I estimate the marginal cost of abating carbon emissions to be between $311 and $423 per ton, well above conventional estimates of the social cost of carbon. Part of the reason behind these high estimated costs is that half of the subsidies went to infra-marginal consumers who would have purchased an electric vehicle whether or not rebates are available. Finally, I evaluate the performance of two alternative policies: an income threshold on eligibility and a cash for clunker program. I find that the additional emission reductions tied to the removal of clunkers are crucial for improving the environmental performance of rebate programs.