BOARD DIVERSITY AND CEO SELECTION

• Date: 01 December 2014
• Author: Atul Gupta,Kartik Raman
• Published date: 01 December 2014
• DOI: http://doi.org/10.1111/jfir.12044

BOARD DIVERSITY AND CEO SELECTION

Atul Gupta and Kartik Raman

Bentley University

Abstract

We find that the likelihood of a female CEO appointment increases with the proportion of

female directors on the board. However, this positive relation is significant only in the

subsample of firms where one of the directors is appointed CEO, and is insignificant in

firms where the new CEO is not from the board. The results are consistent with the

explanation that female directors constitute a supply of viable candidates for the CEO

position, rather than female directors affecting the likelihood of female CEO appoint-

ments by interpreting noisy signals about the abilities of female candidates.

JEL Classification: G30, G34, J17

I think there is a glass ceiling. But it’s glass, and glass means you can see through it and you

can break through it. But it’s not easy. And the reason it’s not easy is because the people who

are going to help you break through that glass ceiling, at least in my life, have all been men.

I think the glass ceiling will go away when women help other women break through that

glass ceiling. That’s what is really going to make a difference.

1

—Indra K. Nooyi, Chairman and CEO, PepsiCo

I. Introduction

Less than 2% of CEOs at S&P 1500 firms are women.

2

The underrepresentation of

women as CEOs is presumably due to the unavailability of qualified candidates and/or

because boards are reluctant to appoint females to this position. Developing an

understanding of why female CEO appointments are infrequent is important, given that

We are grateful for helpful comments from the referee, Susan Adams, Sandra Betton, Harjeet Bhabra, Dhaval

Dave, Rachel Gordon, Rani Hoitash, Karthik Krishnan, Ashwin Malshe, Miriam Schwartz‐Ziv, Emery Trahan,

Anand Venkateswaran, Toni Wolfman, and seminar participants at Bentley University, Concordia University,

Northeastern University, the 2013 Southwest Finance Association conference, and the 2013 Southern Finance

Association conference. We are solely responsible for any errors.

1

http://2001‐2009.state.gov/secretary/rm/2008/10/111186.htm; The Women’s Conference, October 22, 2008,

Long Beach, CA.

2

Based on data in the S&P ExecuComp database. Bertrand (2009) and Bertrand and Hallock (2001) document

that the percentage of women among CEOs was even lower during earlier periods.

The Journal of Financial Research Vol. XXXVII, No. 4 Pages 495–517 Winter 2014

495

© 2014 The Southern Finance Association and the Southwestern Finance Association

RAWLS COLLEGE OF BUSINESS, TEXAS TECH UNIVERSITY

PUBLISHED FOR THE SOUTHERN AND SOUTHWESTERN

FINANCE ASSOCIATIONS BY WILEY-BLACKWELL PUBLISHING

female directors and executives play a value‐relevant role in shaping investment,

financing, and governance decisions at corporations (Adams and Ferreira 2009;

Huang and Kisgen 2013; Levi, Li, and Zhang 2012; Schwartz‐Ziv 2012). In this

article we examine how board diversity influences the firm’s choice of a male or female

CEO.

Our work is motivated by Bilimoria (2006) and Matsa and Miller (2011), who

find that greater female representation on the board increases the likelihood of the firm

having female executives, including a female CEO. These authors, however, do not

explore the underlying drivers of this “gender spillover,”where the presence of females

on the boards increases the likelihood of females among the top executive ranks. As Matsa

and Miller (2011) point out, the results may be driven by gender‐based discrimination,

where “one or both sexes discriminate in favor of members of their own sex,”or could

reflect the proficiency of individuals to “interpret noisy signals about ability for members

of their own sex”(p. 639). A more diverse board may also signal a female‐friendly

environment, which could in turn increase the supply of female candidates for positions at

the firm. The contribution of our article is to disentangle the different explanations to

understand why and under what circumstances boards with female representation are

more likely to choose women CEOs.

Given the endogenous nature of CEO selection decisions, empirically we control

for the potentially common influence of the firm’s culture on both the diversity of the

board and the decision to appoint a female CEO. Using a sample of female CEO

appointments and a matched sample of male CEO appointments, we first establish a

positive relation between the likelihood of a female CEO appointment and the proportion

of female directors in the year preceding the appointment. Next, using exogenous

variation in the percentage of female legislators in the state where the firm is

headquartered as a proxy for variation in the female‐friendly environment or culture in

which firms operate, we find that the positive relation is robust to removing the influence

of culture on board diversity.

We distinguish between two distinct mechanisms by which gender diversity on

the board might influence the selection of a female CEO. The first, which we term the

supply hypothesis, posits that because female directors are potential CEO candidates, a

higher proportion of female directors increases the likelihood that one of them will be

selected for the position. We note that female directors include both independent

directors and top executives serving on the board. For instance, in our overall sample, on

average, 12.7% of the board is composed of female directors, with 8.6% made up of

independent female directors and the remaining 4.1% female top executives. In contrast to

the supply explanation, the information hypothesis posits that female candidates may

have valuable intangible traits that differ from those possessed by males. Female directors

may be better able to pick up on these traits, enhance the information set being used

to evaluate female candidates, and consequently improve the odds of a female being

appointed CEO.

Empirically, we expect that in environments where male directors are likely to be

better informed about a female candidate (e.g., when the candidate is a fellow director) the

informational role of female directors will be less important in influencing the likelihood

of female CEO appointments. Thus, the information hypothesis predicts that the positive

496 The Journal of Financial Research