Optimal Timing of Electric Vehicle Subsidies

There is geographic variation in the environmental damages from electric vehicles (EVs) relative to gasoline internal combustion engines (ICEs) because power grids have different mixes of generation capacity. On the surface, this cautions against subsidizing EVs before the grid is sufficiently decarbonized. However, in this paper I show that EV subsidies are best introduced before the time when EVs become cleaner than gasoline ICEs for two reasons related to the dynamics of decarbonization and technology diffusion. First, because EVs are durable goods, their lifetime damages can be less than those of gasoline ICEs even if the static emission-per-mile comparison disfavors EV adoption at the time of the subsidy. More importantly, policies boosting technology diffusion have positive spillover effects. As marginal emissions of the power grid decline in the long run, more EV adoption produces environmental gains in the process. I simulate an empirically calibrated EV diffusion model, calculate the discounted lifetime damages of EVs versus gasoline ICEs, and examine EV subsidies enacted in different years. Even when EVs are initially more polluting than ICEs, I find that the environmental return from the policy-induced EV diffusion process decreases when governments delay intervention