Review of Industrial Organization
12: 767–780, 1997.
1997 Kluwer Academic Publishers. Printed in the Netherlands.
CEO Age and Outside Directors: A Hazard Analysis
R. RICHARD GEDDES
and HRISHIKESH D. VINOD
Department of Economics, Fordham University, Bronx, NY 10458, U.S.A.
Abstract. This paper examines the relationship between CEO tenure, CEO age, the ﬁrm’s industry
group, the proportion of directors from outside the ﬁrm, and the cost of ﬁring the CEO. A Cox
proportional hazard model of CEO survival is used to study the length of the CEO’s stay at the ﬁrm.
We ﬁnd that, contrary to previousstudies, a greater proportion of outsiders has a positiveeffectonCEO
tenure. The signiﬁcance of this result is however sensitive to the inclusion of age and performance
variables. We test for the effects of heterogeneity of industry, and ﬁnd that ﬁrms in homogeneous
industries exhibit lower durations. As the cost of ﬁring the CEO rises, tenure also rises.
Key words: CEO tenure, hazard analysis, outside directors.
Researchers have established that the board of directors in a large, public corpora-
tion plays an important role in monitoring the chief executive ofﬁcer (CEO) of the
ﬁrm and helps solve the “agency problem” between the owners of the corporation
and its managers. The board can monitor CEOs by linking their pay packages to
owners’ wealth or by removal of the CEO in the event of poor performance. Agen-
cy problems are reduced when the ﬁring or retention of the CEO by the board is
sensitive to changes in owners’ wealth. The threat of ﬁring during periods of poor
ﬁrm performance and retention during periods of good performance encourages
managersto act in the interests of owners.
Weisbach (1988) shows that members of
the board of directors who are chosen from outside the ﬁrm, or “outside directors”,
play an especially important role in monitoring CEOs through removal.
In this paper we explore the relationship between the board of directors, man-
agerial turnover, and a number of important variables. We use statistical survival
analysis to examine the interrelationship between CEO turnover, CEO age, the
composition of the board (“insiders” versus “outsiders”), the industry in which the
ﬁrm operates, and the cost of ﬁring the CEO. We develop a nonlinear function
of the CEO’s age to approximate the cost of ﬁring in terms of severance pay and
possible adverse impact on the reputation of the ﬁrm. While other researchers
have examined the relationship between CEO turnover and board composition,
We thank an anonymous referee for helpful comments and suggestions.
Numerous studies have focused on factors affecting CEO turnover, including Couglin and
Schmidt (1985), Warner et al. (1988), Weisbach (1988), Gibbons and Murphy (1990), and Jensen and