Being the CEO's boss: An examination of board chair orientations

• Author: Ryan Krause
• DOI: http://doi.org/10.1002/smj.2500
• Published date: 01 March 2017
• Date: 01 March 2017

Strategic Management Journal

Strat. Mgmt. J.,38: 697–713 (2017)

Published online EarlyView 8 March 2016 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/smj.2500

Received 8 February 2014;Finalrevision received 8 December 2015

BEING THE CEO’S BOSS: AN EXAMINATION

OF BOARD CHAIR ORIENTATIONS

RYAN KRAUSE*

Department of Management, Entrepreneurship, and Leadership, Neeley School of

Business, Texas Christian University, Fort Worth, Texas, U.S.A.

Research summary: Scholars have traditionally conceptualized board leadership as a dichoto-

mous construct. A combined CEO and board chair position is interpreted as reecting a more

collaborative approach to corporate governance, whereas separate positions are interpreted as

ensuring greater board control.I challenge this conceptualization and posit that a separate board

chair can be oriented toward collaboration as well as—or in place of —control. I analyze newly

available data from corporate proxystatements to identify these two board chair orientations and

test competing perspectives on how they impact protabilitygrowth in a sample of S&P 500 rms.

The results indicate that board leadership is a more nuanced phenomenon than the extantlitera-

ture would suggest.

Managerial summary: What is the role of the board chair when not the CEO? Corporate gover-

nance experts assert the board chair’s role is to monitor and control the CEO. Yet, board chairs

often play another,more collaborative role. Board chairs frequently provide advice and guidance

to CEOs and relieve CEOs of board leadership burdens, enabling the CEOs to focus on their

primary responsibilities. In this study, I examine the effect of board chair orientations on nan-

cial performance and nd that, as with separating or joining the CEO and board chair positions,

the protability implications of the selected orientation are far fromuniversal. Board chairs must

consider their rm’s performance context in order to get the most out of a particular approach to

being the CEO’s boss. Copyright © 2016 John Wiley & Sons, Ltd.

INTRODUCTION

On November 2, 2015, technology giant

Hewlett-Packard split into two separate rms,

with Margaret (Meg) Whitman transitioning from

CEO and board chair of the parent rm to CEO

of Hewlett-Packard, Inc., and board chair of

Hewlett-Packard Enterprise. In an interview on

CNBC the day of the split, Whitman (2015) was

asked about her approach to performing the CEO

and board chair roles at separate rms, to which she

responded: “I know the role of the chairman, and I

Keywords: board chair orientations; board leadership;

control; collaboration; content analysis

*Correspondence to: Ryan Krause. TCU Box 298530, Fort Worth,

TX 76129. E-mail: r.krause@tcu.edu

Copyright © 2016 John Wiley & Sons, Ltd.

know how it is different than the role of the CEO.

The chairman is not there to run the company. The

chairman [role] is to help the board be productive,

help the CEO be successful.”

Though common in practice, Whitman’s con-

cept of the role a board chair performs when sep-

arate from the CEO contrasts sharply with the

traditional scholarly conceptualization. For nearly

30 years, researchers have theorized that boards

face a fundamental trade-off: pursue greater con-

trol of management by separating the CEO and

board chair positions, or pursue greater collabora-

tion by combining the CEO and board chair posi-

tions (e.g., Boyd, 1995; Finkelstein and D’Aveni,

1994). Despite sparse evidence supporting a perfor-

mance benet to either the combined or separate

leadership structure (Dalton et al., 1998, 2007), few

have questioned this trade-off. The dichotomous

698 R. Krause

nature of the board leadership phenomenon seems

to have entrenched the assumption that the sepa-

rate board chair structure reects a control approach

to governance, with collaboration a benet solely

linked to the combined CEO-board chair struc-

ture (Gove and Junkunc, 2013; Sundaramurthy and

Lewis, 2003).

While this assumption has enabled the devel-

opment of the board leadership literature (Krause

et al., 2014), it nonetheless misaligns theory with

reality. While boards can use the separate leader-

ship structure to control the CEO more effectively,

control is not the only reason to have a separate

board chair. As the quote from Meg Whitman sug-

gests, the separate structure can actually enhance

CEO-board collaboration by reducing the demands

on the CEO’s time, allowing the CEO to specialize

in managing the rm’s strategy and operations, and

providing the CEO with a close source of advice

and guidance. Consider how Walgreen Co. (2010:

17) explains its choice to have a board chair that is

separate from its CEO:

We believe this structure is optimal for

Walgreens at this time because it allows Mr.

Wasson [the CEO] to focus on leading the

Company’s business and operations. At

the same time, the Chairman can focus

on leadership of the Board of Directors,

including …serving as a liaison and supple-

mental channel of communication between

independent directors and the Chief Execu-

tive Ofcer, and serving as a sounding board

and advisor to the Chief Executive Ofcer.

(Emphasis added)

This rationale reects a philosophy about the role

of the separate board chair that is virtually unrec-

ognized in corporate governance scholarship (For

an exception, see Lorsch and Zelleke, 2005). It

represents a collaborative approach to board lead-

ership that is grounded more in organization the-

ory and stewardship theory than in agency the-

ory (Donaldson, 1990). While prior research has

acknowledged that boards can employ different

approaches to corporate governance— a control

approach or a collaboration approach (Sundara-

murthy and Lewis, 2003)— the use of a separate

board chair remains conceptually linked with the

control approach (Krause et al., 2014). With this

research, I introduce the construct of board chair

orientation, dened as the corporate governance

approach a separate board chair applies to his or her

working relationship with the CEO, and I maintain

that board chairs can adopt a collaboration orien-

tation as well as— or instead of— a control orienta-

tion. As the share of large U.S. rms with a separate

board chair continues to climb toward 50 percent

(Spencer Stuart, 2013), understanding the orien-

tation of the board chair will only become more

important.

To ascertain board chair orientations, I rely on

newly available passages from corporate proxy

statements, in which boards provide the rationale

behind their choice of board chair. Since 2010,

the Securities and Exchange Commission (SEC)

has required publicly traded rms to disclose this

rationale. I analyzed the disclosures from all S&P

500 rms with a separate board chair, and the

analysis yielded novel insights about the divergent

effects of board chair orientations. Drawing on

the work of Finkelstein and D’Aveni (1994), and

Krause and Semadeni (2013), I develop competing

predictions as to how board chair orientations will

interact with past performance at separate-structure

rms to impact protability growth. The results

provide clear support for the logic of the latter

article, and also show that different board chair

orientations can have distinct consequences, even

when the board leadership structure is the same.

The present research contributes to the literature

on board leadership in a number of ways. First

and foremost, this is the rst theory-driven study

to examine alternative orientations toward the role

of the separate board chair. Second, the present

research extends prior contingency-based scholar-

ship by illuminating the nature of the moderating

effect that past performance has on the relationship

between board leadership and protability growth.

The results of this study help to reconcile discrep-

ancies in the literature, particularly between the

competing predictions of Finkelstein and D’Aveni

(1994), and Krause and Semadeni (2013) regarding

the effects of control and collaboration in a board

leadership context. Third, the present research

extends Krause and Semadeni’s (2013) initial

parsing of the board leadership construct. Whereas

Krause and Semadeni (2013) maintained that

“demotion” separations, in which the CEO loses

the board chair position, are the most consistent

with the control approach to corporate governance

and will therefore exhibit the greatest effects, I sub-

mit that a separate board chair’s orientation toward

Copyright © 2016 John Wiley & Sons, Ltd. Strat. Mgmt. J.,38: 697–713 (2017)

DOI: 10.1002/smj