The authors propose two alternative hypotheses regarding the relationship between the level of CEO confidence and resources spent on corporate social responsibility (CSR). The first hypothesis is that overconfident CEOs are less likely to invest in CSR initiatives because they undervalue the hedging benefits that CSR may provide in difficult times. The second hypothesis is that overconfident CEOs tend to exhibit narcissistic behavior and thus revel in the additional media attention and admiration that an extensive investment in CSR may deliver to them as their firms’ key representatives.

Using regression analysis, the authors find evidence that overconfident CEOs are less likely to invest in CSR. This finding reduces the likelihood that narcissistic behavioral tendencies drive investment in CSR programs.

Institutional investors are under increasing pressure from their own investors to place more emphasis on socially responsible principles as part of their investment process. The authors show that the relationship between CSR investment and CEO confidence should be monitored to ensure that overconfident CEOs do not neglect this aspect of their firms. Because CEOs, given their status, are likely to project an aura of confidence, it is difficult in practice to ascertain extreme confidence without significant additional research.

Although the calculations are extremely high level, the authors attempt to estimate the financial impact on a firm of higher-than-average CEO confidence levels.

The authors perform a series of regressions to determine whether CEO confidence levels affect organizations’ CSR scores. The CSR scores are based on ratings from the Kinder, Lydenberg, and Domini database. The CSR scores are broken down into seven subcategories: community, corporate governance, workforce diversity, employee relations, environment, human rights, and product quality. To test for robustness, the authors group the scores for these subcategories into two categories: institutional (community, workforce diversity, environment, and human rights) and technical (corporate governance, employee relations, and product quality).

The authors’ measure of CEO confidence is based on CEO option exercise behavior. Prior research has suggested that rational CEOs with undiversified portfolios exercise options that are deep in the money. Overconfident CEOs, however, tend to overestimate their firms’ future prospects and undervalue current stock prices and fail to exercise their vested company-issued options. The authors obtained data on CEO option holdings from Execucomp.

They also add a number of control variables to the regressions to control for such factors as the CEO’s gender, age, and tenure; return on assets; levels of assets and debt; levels of advertising and R&D; capital expenditure; and levels of external ownership. A summary of the data suggests that older, longer-serving, male CEOs are likely to be more confident than other CEOs.

The authors’ regression analysis shows that CEO overconfidence is negatively related to CSR scores. To confirm that their results are robust, they test various definitions of CEO confidence, although these definitions are still all based on quantitative measures derived from actions concerning the exercise of in-the-money options.

The authors further decompose CSR scores into institutional and technical components to test whether CEO confidence has different impacts on different aspects of CSR. The results show a more negative relationship between CEO confidence and the institutional components (community, workforce diversity, environment, and human rights) compared with the technical components. The authors test the individual components of CSR as the independent variables in the regressions, finding a positive connection between the product quality aspect of CSR and CEO confidence, which suggests that overconfident CEOs are better innovators.

The ubiquitous and anonymous nature of social media means that operational events are widely publicized and difficult for firms to manage without awarding high levels of compensation to affected parties. More general disapproval of corporate behavior—including executive compensation, rising shareholder profits, and deteriorating real wages—adds to the anger felt by some aggrieved stakeholders.

Although a firm can act as a good corporate citizen regardless of any perceived secondary benefits, its well-intentioned principles and actions may be overlooked when things go awry. The silver lining is that the principles ingrained in strong corporate governance programs might reduce the frequency of these adverse events, regardless of the CEO’s nature.