The authors empirically test three hypotheses with regard to financial services firms: (1) Corporate environmental responsibility (CER) investment improves a firm’s reputation and thereby improves financial performance, (2) CER investing is a trade-off and a negative synergy, and (3) CER investing is a positive synergy that leads to further CER investment. Only the first hypothesis is accepted as true based on empirical evidence.

The authors consider their findings to be consistent with the claims of stakeholder theory rather than the claims of shareholder wealth maximization theory. The evidence is greater in more-developed markets but still exists in less-developed markets. Generally, the benefit of CER investing takes one to two years to emerge. The authors posit that CER investment internally (reducing the firm’s negative environmental impact) or externally (providing capital for CER-type projects) creates a positive reputation that leads to lower borrowing costs and better firm performance.

Practitioners, whether they support stakeholder theory or not, will benefit by understanding how empirical evidence is produced to test stakeholder theory. Although the authors conclude that stakeholder theory is supported through the use of CER investment by financial firms, they also reveal that their variables are not sufficiently granular to determine whether CER investments are internal or external. Furthermore, there is no control variable to enable them to consider the effect of the firms’ regulatory environment.

Consequently, it could be argued that the hypothesis has not been directly tested and that the analysis is missing a potentially important control variable. Further research may correct such issues and still find support for stakeholder theory. In that case, firms with significant CER investment would become desirable investments.

The empirical research in this area is in the early stages, but it will be worth following as it progresses.

The authors use annual data (2002–2011) from financial services firms in 29 countries/territories (4,924 observations in total). Data that measure the direct and indirect annual environmental costs of a given firm are supplied by Trucost Plc. Other data, aside from annual return volatility based on monthly data, are primarily from financial statements.

Firm performance is measured by return on assets. As a robustness check, earnings before interest and taxes normalized by total assets is also used. The regression analysis includes environmental costs lagged by one year and two years and a number of control variables as independent variables. Different regression techniques are used to control for endogeneity and autocorrelation. The main result is a statistically significant negative relationship between firm performance and lagged environmental costs. The authors deduce that when environmental costs are reduced through CER investment, firm performance improves.

I think the authors overstate the implications of their results because there is no variable to directly measure CER investment by the firm. Also, CER investment might improve firm performance because of the regulatory consequence of not making such investments. No regulatory control variable was implemented in the regression analysis to determine whether this is the case. Further research and more granular data may address these issues in the future.

Another reason I am skeptical of the results is that a ratio with total assets in the denominator is used as a dependent variable in the regression and the natural logarithm of total assets is used as a control variable and in the denominator of the other control variables. Although the authors have attempted to control for endogeneity of variables in their analysis, the regression results may be compromised, given the structural links between the dependent and independent regression variables. Additionally, I am curious as to why a contemporaneous environmental cost variable is not used within the regression analysis. Perhaps a first difference between the contemporaneous variable and a lagged variable could provide an indirect measure of CER investment by the firm.

Despite my skepticism of the results, I am intrigued by the idea of testing the financial and nonfinancial benefits to the firm of CER investment and hope to see further research in the area.