CO2 Emissions and Corporate Performance: Japan's Evidence with Double Machine Learning

This paper empirically examines the relationship between CO2 emissions and corporate performance in terms of long-term performance, short-term performance, and cost of capital, using available firm-level data in the First Section of the Tokyo Stock Exchange from FY2011 to FY2019. To address potential biases in previous empirical studies, we employ double machine learning, which is one of the semiparametric models introduced by Chernozhukov et al. [2018], for our empirical analysis. We find that corporations with lower CO2 emissions have (i) better long-term corporate performance and (ii) lower cost of equity. These results suggest that investors estimate that corporations with lower CO2 emissions have lower business risks, setting their risk premium to be low, which results in higher market value of such corporations. In addition, our analysis indicates that corporations with lower CO2 emissions have higher short-term performance and lower cost of debt, but also shows that the results of previous studies of these relationships may contain biases and should be evaluated with caution.