VAT reforms solve pollution and creates incentives for companies to transition.

We next perform cross-sectional heterogeneity checks according to local environmental regulations and expect that there is a larger decline in pollution behavior after the VAT reform for firms located in areas with stricter environmental regulations. As noted earlier in our paper, the TCZ policy was implemented by the central government in 1998 with the objective of reducing SO2 emissions in cities with particularly high air pollution. Once a city was designated as a TCZ city, tougher regulatory policies were implemented. The three main policy measures embodied in the TCZ city were the closing of the biggest polluters, reducing the sulfur-content of coal, and promoting cleaner coal-burning (Cai et al., 2016; Hering & Poncet, 2014). We thus add an interaction term between VAT and the TCZ dummy to our regression and find that the coefficient on the interaction term is negative and statistically significant at 1%, which is consistent with our hypothesis. The detailed result is presented in column (4) of Table 8.

5.2.4 Water-polluting industries

We have verified that firms reduce their waste water emissions after the VAT reform as they are provided with more funds for environmentally friendly investments. If this is the case, firms in water-polluting industries are more prone to mitigate sewage discharge after the policy implementation. In this subsection, we thus examine whether the VAT reform exerts diversified impacts on firms in water-polluting and non-water-polluting industries. We follow Chen, Kahn, et al. (2018) in categorizing our sample into water-polluting industries and non-water-polluting industries. Specifically, we classify pulp and paper production, agricultural products and byproduct, chemical materials and products, textile, beverage production, and ferrous metal smelting and pressing as water-polluting industries according to each industry’s waste water discharge levels. We then construct a dummy variable, Water_polluting, which takes the value of one if an industry belongs to one of the above industries, and zero otherwise. We then include an interaction term between VAT and the Water_polluting dummy in our estimation. The result is shown in column (5) of Table 8. In line with our expectations, the coefficient on the interaction term is negative and statistically significant at 5%, suggesting that water-polluting industries experience a significantly larger reduction in sewage emission.

6 Conclusion

Using a panel of 380,387 firm-year observations in China from 1998 to 2012, we find robust evidence to confirm that the VAT reform decreases firm pollution, especially toxic emissions. These main results still hold under a series of endogeneity and robustness tests. The potential mechanisms of our findings are that the VAT reform stimulates investment in fixed assets for environmental protection and accelerates technological progress, which finally reduce firm pollution. We find that SOEs, large firms, firms with a high level of subsidies, firms located in stricter environmental-regulation areas, and firms in water-polluting industries reduce pollution to a larger extent after the VAT reform. Overall, our study provides timely policy implications for environmental protection and firm pollution behavior. We believe that this topic should be of particular interest to governments worldwide that are concerned with tax incentives and environmental protection.