The debate over whether CEOs add value to firms has come under renewed scrutiny in the wake of the global financial crisis. Arguments have been made that other factors, such as core competence, the quality of a firm's products, other employees and stakeholders and even luck are more important determinants of firm performance. But there has been surprisingly little evidence, until now, to settle the question of CEO value.

The main stumbling block has been how to measure CEO impacts on a firm. Sudipto Dasgupta of HKUST and co-authors Yuk Ying Chang and Gilles Hillary fill that gap with a study of CEO departures so they can take a "before and after" snapshot. They find that yes, CEOs do matter.

The study assesses 298 voluntary and forced CEO departures from 1992-2002 on three factors: the stock market reaction to the departure news, the subsequent success of the CEO in the managerial labour market, and the performance of the firm after the CEO leaves. The authors examine how these factors relate to the past performance of the firm while under the CEO's management, and the CEO's pay relative to the other highest paid executives of the firm

"Our results are consistent with the view that cross-sectional differences in firm value and performance are related to ability differences across CEOs. We also find evidence against the view that CEOs are paid well simply because they are in a position to extract better compensation from their boards," they say.

In fact, stock market reactions to CEO departures link CEO ability with firm performance. "The market reacts more negatively when a more capable manager who generated higher stock returns leaves a company," they say.

Past firm performance and CEO pay also have an impact on a CEO's subsequent career. Those from stronger performing firms and with higher pay were more likely to progress to managerial positions in public or private companies. Moreover, the firm that a capable CEO departs from has a poorer stock and operating performance in the one to three years after he or she leaves. All of this adds up to evidence that managerial ability had been a factor in the firm's success.

"Collectively, the results reject the view that differences in firm performance stem entirely from non-CEO factors such as firm's assets, other employees, or luck, and offers evidence that CEO pay reflects the CEO's contribution to firm value," they say.

The authors' say an advantage of their multi-pronged approach is that it leaves little room for alternative explanations. For instance, it could be argued that CEOs "jump ship" or are forced to quit when the firm's future prospects look dim, but this would not explain why the manager's subsequent labour progress is positively related to past pay.