The effects of new energy vehicle subsidies on air quality: Evidence from China

Automobile exhaust emissions cause detrimental effect to human health by polluting air. The promotion of new energy vehicles (NEVs) to reduce vehicle exhaust emissions for human health has become an important tool to fight against air pollution. This paper examines the causal relationship between the NEV subsidies in China and air quality through a two-way fixed effects model and a theoretical model. Combining data on staggered prefecture-level city implementation of the subsidies with remote-sensing daily PM2.5 concentration data, we show that the subsidies for NEVs in China significantly improved air quality. The subsidy of NEVs in the private sector (EVPV) reduced PM2.5 concentration by 3% for the current year and 3.2% two years after. In particular, the effect of the EVPV implemented two years after increases by 5.7% with an additional 10,000 people, increases by 3.1% with an additional million civilian cars, increases by 0.3% with another one key corporation related to NEVs existed in the city, and in medium city, large city, megacity and megalopolis is significantly positive on air quality as compared with that in small city. In addition, the EVPV implemented for the current year and two years after reduce PM2.5 concentration by an additional 1.6% and 3.4% if air quality worsens by 25% in the city, respectively. The subsidy of NEVs in the public sector (EVPB) increased PM2.5 concentration by 4.3% for the one-year lead and reduced PM2.5 concentration by 5% for the current year of the policy implementation. The year before implementation of the EVPB significantly increases PM2.5 concentration in medium city, large city and megacity and the EVPB implemented for the current year reduces PM2.5 concentration by an additional 3.2% if air quality worsens by 25% in the city. Driving and purchasing restrictions have significant impact on the causal relationship of the EVPB on air quality. Our results remain robust through a series of robustness tests. By developing a theoretical model, we analyze the effect of subsidy policy on air quality as a supplement for empirical results and compare the optimal subsidy intensities for NEVs from the perspectives of consumer surplus, emissions and social welfare. The results indicate that the cost of subsidy policies is lower at the initial stage of the promotion of NEVs, while when consumers begin to accept NEVs widely, the cost of the subsidy policies will increase accordingly.