Association for Public Policy Analysis & Management

There is little research on the effectiveness of proactive climate action by corporations in reducing their carbon footprints. Existing studies on corporate participation in voluntary climate initiatives have by and large found that participation does not have an effect on firms’ carbon emissions, and in fact, one study finds that participation is associated with an increase in carbon emission intensity by firms. These prior studies model participation as a binary decision and do not account for the heterogeneity in firms’ intensive margins of participation. By contrast, this paper takes a different approach and hypothesizes that firms that participate more intensively (e.g., engage in higher levels of investments in renewable energy or higher levels of carbon disclosure) are more likely to reduce carbon emissions than those that do not. To test this hypothesis, this paper model the decision of firms to participate in the Carbon Disclosure Project (CDP)—a voluntary initiative that invites firms to disclose their carbon emissions—in a discrete-continuous choice model, whereby the binary choice of participation is linked with continuous choices about the 1) level of carbon disclosure and 2) intensity of carbon emissions. Dynamic panel model estimation serves as an alternative modeling approach. Empirical analysis is based on data from a panel of S&P 500 firms during 2011-2016. Firm-level characteristics include participation in the CDP, level of carbon disclosure, carbon emissions, carbon emission intensity, firm size, and data on firms’ carbon management practices, such as the existence of senior managers responsible for climate change, integration of climate risks in business operations, and adoption of quantifiable emissions targets. The empirical analysis also controls for political economy factors, such as U.S. state level policies and stakeholder pressures. For identification, difference-in-difference-in-differences specifications that are nested within discrete-continuous choice models are employed to exploit President Obama’s announcements of executive actions on climate change, which increased the likelihood of climate change regulation in the U.S. during the study period.