Executive Summary:

Ratings of corporations' environmental activities and capabilities influence billions of dollars of "socially responsible" investments as well as consumers, activists, and potential employees. But how well do these ratings predict socially responsible outcomes such as superior environmental performance? Companies can enhance their environmental image in one of two ways: by reducing or minimizing their impact on the environment, or by merely appearing to do so via marketing efforts or "greenwashing." This study evaluates the predictive validity of environmental ratings produced by Kinder, Lydenberg, Domini Research & Analytics (KLD), and tests whether companies that score high on KLD ratings generate superior environmental performance or whether highly rated firms are simply superior marketers of the factors that these rating agencies purport to measure. The data analysis examines all 588 large, publicly-owned companies in the United States that were both regulated by the U.S. Environmental Protection Agency and whose social performance was rated by KLD at least once during 1991-2003. This paper may be the first to examine the predictive validity of social or environmental ratings. Key concepts include:

• KLD ratings for environmental "concerns," such as hazardous waste and regulatory problems, have small but statistically significant effects in predicting future emissions and regulatory violations.
• KLD ratings for environmental "strengths," such as environmentally beneficial products or pollution prevention, do not predict future environmental outcomes.
• Most, but not all, of the predictive power of KLD ratings is due to the fact that lagged emissions and regulatory violations predict both lagged KLD ratings and future emissions and regulatory violations.
• KLD expends substantial resources attempting to measure the quality of companies' environmental management systems. The results suggest that this measurement is difficult to do well.