Can Directors Impact Performance? A case‐based test of three theories of corporate governance

• Author: Gavin J. Nicholson,Geoffrey C. Kiel
• Published date: 01 July 2007
• DOI: http://doi.org/10.1111/j.1467-8683.2007.00590.x
• Date: 01 July 2007

Can Directors Impact Performance?

A case-based test of three theories

of corporate governance

Gavin J. Nicholson* and Geoffrey C. Kiel

We examine hypothesised links between the board of directors and f‌irm performance as

predicted by the three predominant theories in corporate governance research, namely agency

theory, stewardship theory and resource dependence theory. By employing a pattern matching

analysis of seven cases, we are able to examine the hypothesised link between board demog-

raphy and f‌irm performance expected under each theory. We f‌ind that while each theory can

explain a particular case, no single theory explains the general pattern of results. We conclude

by endorsing recent calls for a more process-orientated approach to both theory and empirical

analysis if we are to understand how boards add value.

Keywords: Boards of directors, agency theory, stewardship theory, resource dependence

theory, organisational performance

Introduction

Do boards of directors really have any

impact on corporate performance? This

question is central to the normative assump-

tion that boards should both contribute to,and

be held accountable for, f‌irm performance

(Drucker, 1999; NACD, 2000). The belief that

directors do have an impact on f‌irm perfor-

mance is ref‌lected in survey research, which

indicates institutional investors are willing

to pay a premium for “good governance”

(Felton et al., 1996, p. 170; Investor Relations

Business, 2000, p. 1). This assumption is

ref‌lected at virtually all levels of the global

business system. Institutional investors world-

wide expect boards to contribute to f‌irm per-

formance (Black, 1992; Useem, 1993), there are

repeated calls to overhaul national systems of

corporate governance and make boards more

accountable, particularly in developing nations

(Johnson et al., 2000), and there is widespread

public criticism of particular boards (Lavelle,

2002) and even of individual directors

(Chernoff, 2000).

There has also been an escalation of research

interest in corporate governance and the re-

lationship between the board and f‌irm perfor-

mance over the past 15 years (e.g. Zahra and

Pearce, 1989; Pettigrew, 1992; Johnson et al.,

1996; Bhagat and Black, 1999). Given the

importance of the subject and the level of

research activity, it would seem reasonable to

expect that a clear and demonstrable link

between the board and corporate performance

has been established. Despite a sustained ef-

fort, however, researchers have so far failed

to identify this link.

The majority of academic research into

the board–performance nexus has adopted

Pfeffer’s (1983) argument that demographic

variables provide parsimonious and objective

representations of constructs that are other-

wise diff‌icult to collect and validate. As a

result, the research agenda has concentrated

on large-sample, quantitative studies directly

examining the relationship between corpor-

ate performance and various board attributes

such as board independence (Bhagat and

Black, 1999), leadership structure (Fosberg and

Nelson, 1999), board size (Eisenberg et al.,

1998), and the role of the CEO (Finkelstein and

Boyd, 1998; Sanders, 2001). In general, these

studies report either small (but conf‌licting)

*Address for correspondence:

School of Accountancy &

Centre of Philanthropy and

Nonprof‌it Studies, Queensland

University of Technology,GPO

Box 2434, Brisbane, Qld 4001,

Australia. Tel: +61 7 3138 9299;

Fax: +61 7 3138 9131; E-mail:

g.nicholson@qut.edu.au

CAN DIRECTORS IMPACT PERFORMANCE? 585

Volume 15 Number 4 July 2007

© 2007 TheAuthors

Journal compilation © 2007 BlackwellPublishing Ltd, 9600 Garsington Road,

Oxford, OX4 2DQ, UK and 350 Main St,Malden, MA, 02148, USA

results or no demonstrable link. Lawrence and

Stapledon (1999), for example, found onlyscat-

tered non-robust correlations between various

performance measures and the proportion of

independent directors, while Hermalin and

Weisbach (1991) found no correlation between

board composition and f‌irm performance.

Recent summary meta-analytic studies have

not aided in clarifying these relationships,

with Daltonet al. (1998) f‌inding no relationship

between board composition and f‌inancial per-

formance, while Rhoades et al. (2000) found a

small positive relationship.

In a related research stream, academics

have examined the relationship between

board attributes (such as independence) and

various corporate activities thought to im-

pact on shareholder wealth. Results are

similar to those examining the direct board–

performance relationship, producing equivo-

cal f‌indings (Westphal, 1999). For example,

studies analysing the relationship between

board structure and various activities such as

corporate diversif‌ication (Hill and Snell, 1988;

Baysinger and Hoskisson, 1990), CEO com-

pensation (Fosberg, 1999), the use of long-term

incentive plans (Zajac and Westphal, 1994), the

adoption of takeover defences such as poison

pills (Brickley et al., 1994; Coles and Hesterly,

2000) or paying of green mail (Kosnik, 1987),

and the commission of illegal acts (Kesner et

al., 1986) have produced negative f‌indings, or

been unable to identify any correlation at all.

In short, there is a long line of research that

provides little consensus as to the effect of the

board of directors on the performance of the

corporation both directly or through corporate

activities thought to affect shareholder wealth

(Johnson et al., 1996; Coles et al., 2001).

More recently, research efforts aimed at

examining the processes by which boards

carry out their roles, rather than impacts on

corporate behaviour or performance directly,

have met with more promising results. For

instance, Westphal (1999) reported that social

ties between the board and CEO typically

enhanced the likelihood of independent direc-

tors providing advice and counsel to the CEO.

In studies with colleagues he also reported

that a board’s engagement in the strategic

decision-making process encourages inter-

locking directorates (Gulati and Westphal,

1999) and that the strategic context of social

network ties between directors, rather than

number of interlocks, is an important inf‌lu-

ence on corporate governance (Carpenter and

Westphal, 2001).

In studies investigating the board’s involve-

ment in strategy, Golden and Zajac (2001)

found that, in the governance of hospitals,

board processes and demography signif‌icantly

affect strategic change. Similarly, Westphal

and Fredrickson (2001) found that, while the

prior experience of new CEOs predicts corpo-

rate strategic change, this might mask the

process by which an experienced board can

inf‌luence strategy development.

A third board role relates to a director

providing access to resources such as informa-

tion (Baysinger and Zardkoohi, 1986). When

investigating a board’s access to information,

Haunschild and Beckman (1998) found the

process by which boards gain information

about acquisitions variesaccording to whether

the information is derived from a personal or

impersonal source.

While these studies contribute signif‌icantly

to our understanding of how board attributes

contribute to board roles, none of them has as

yet attempted to link board attributes with

corporate performance. By reviewing both the

traditional board–performance and more re-

cent board-behaviour studies it becomes ap-

parent that it is necessary to understand the

processes that link the board of directors to

corporate performance, rather than looking for

a parsimonious relationship (such as simple

correlation) between the two (Pettigrew, 1992;

Forbes and Milliken 1999). Our objective,

therefore, is to build upon the recent literature

and attempt to unravel the processes that link

board attributes to f‌irm performance and in so

doing make two contributions to the research

agenda. First, we aim to examine the entire

process predicted to link boards to corporate

performance by investigating the three theo-

retical paradigms that dominate corporate

governance research, namely agency theory

(Jensen and Meckling, 1976; Fama and Jensen,

1983; Eisenhardt, 1989a), stewardship theory

(Donaldson, 1990; Donaldson and Davis, 1991,

1994) and resource dependence theory (Zald,

1969; Pfeffer, 1972, 1973; Pfeffer and Salancik,

1978). Second, our methodology allows us to

move beyond traditional samples that have

concentrated on the top tiers of the for-prof‌it

business community and respond to Forbes

and Milliken’s (1999) call for a greater under-

standing of the differences between the boards

of for-prof‌it companies and boards that work

under different ownership structures. In short,

we aim to employ a qualitative methodology

to shed new light onto the entire board–

performance nexus across a variety of corpo-

rate structures.

Theories of corporate governance

and pattern development

Agency theory, stewardship and resource

dependence theories have undoubtedly as-

586 CORPORATE GOVERNANCE

Volume 15 Number 4 July 2007 © 2007 TheAuthors

Journal compilation © BlackwellPublishing Ltd. 2007