The impact of board size on board demographic faultlines

• DOI: https://doi.org/10.1108/CG-03-2020-0100
• Published date: 14 August 2020
• Pages: 1205-1222
• Date: 14 August 2020
• Author: Muhammad Ali,Oluremi B. Ayoko
• Subject Matter: Strategy,Corporate governance

The impact of board size on board

demographic faultlines

Muhammad Ali and Oluremi B. Ayoko

Abstract

Purpose –Demographicfaultlines are associated with negative group processes and low performance.

Little is known aboutthe formation of faultlines in boards and how they can be weakenedto capitalize on

the positiveeffects of diversity.

Design/methodology/approach –This study draws on social identity theory and faultlines theory to

provide insightsinto how gender and age faultlinesare formed in a board. Subsequently,it proposes and

tests a U-shaped board size–faultlines strength relationship. Archival data were collected on 288

organizationslisted on the Australian Securities Exchange.

Findings –Hierarchical regression analyses indicate that small- and large-sized boards experience

strongerfaultlines than medium-sized boards.

Originality/value –This study providespioneering evidence for a U-shapedrelationship between board

size and demographicfaultlines strength. These findings informpractice by suggesting an optimal board

size.

Keywords Board size, Faultlines, Gender, Age, Social identity theory, Faultlines theory

Paper type Research paper

Introduction

Faultlines are alignments of diversity attributes in a diverse group creating homogeneous

subgroups (Lau and Murnighan, 1998). For example, a group comprising three women and

three men will form faultlines if, for example, the women are younger than the men. In this

case, group members’ genders and ages align, creating two homogeneous subgroups of

young women and aged men. Faultlines strength refers to the extent to which diversity

attributes align (Thatcher et al., 2003). For instance, a group comprising three young

women and three aged men demonstrates stronger faultlines strength compared to a group

comprising two young women, one agedwoman and three aged men.

Faultlines are generally associated with negative group processes, such as low

productive energy (Kunze and Bruch, 2010) and high emotional conflict ( Pearsall

et al.,2008

). These negative group processes can be stronger than those

experienced by a group where the subgroups are based on a single diversity attribute

(e.g. women vs men and no alignment of gender and age) (Lau and Murnighan, 1998;

Price et al., 2007;Tuggle et al., 2010). Therefore, faultlines can explain performance,

above and beyond what can be explained by demographic diversity (Bezrukova

et al., 2016;Bhat et al., 2019;Martinez-Jimenez et al., 2020;Mazzotta et al.,2017;

Wellalage and Locke, 2013). These possibilities drove several studies on the impact

of faultlines on group processes (Flache and Ma

¨s, 2008;Lau and Murnighan, 2005;

Meyer et al., 2016;Molleman, 2005;Stanciu, 2017), group outcomes (Ren et al.,2014;

Rico et al.,2012;van Knippenberg et al.,2011;Zhang et al., 2017)andmoderating

effects on the faultlines–processes/outcomes relationship (Adair et al.,2017;

Scho

¨lmerich et al., 2016;Spoelma and Ellis, 2017). The moderating effect of faultlines

Muhammad Ali is based at

the QUT Business School,

Queensland University of

Technology, Brisbane,

Australia. Oluremi B. Ayoko

is based at the UQ

Business School, University

of Queensland, Brisbane,

Australia.

Received 16 March 2020

Revised 22 May 2020

16 June 2020

Accepted 20 July 2020

DOI 10.1108/CG-03-2020-0100 VOL. 20 NO. 7 2020, pp. 1205-1222, ©Emerald Publishing Limited, ISSN 1472-0701 jCORPORATE GOVERNANCE jPAGE 1205

on the predictor–outcome relationship (Hutzschenreuter and Horst kotte, 2013;Mo

et al., 2019;Price et al., 2007;Spell et al., 2011;Vora and Mark

oczy, 2012)and

measurement of faultlines have also been an area of interest (Shaw, 2004;Trezzini,

2008;Lawrence and Zyphur, 2011;Meyer and Glenz, 2013). A very small body of

research investigated the formation or antecedents of faultlines (Kaczmarek et al.,

2012a), dormant and activated faultlines (Jehn and Bezrukova, 2010;Veltrop et al.,

2015) and crisscrossing group members (Ma

¨set al., 2013).

Following the conceptualization of faultlines at the group level (Lau and Murnighan, 1998),

most empirical studies focused on this phenomenon at this level (Bezrukova et al., 2009;

Choi and Sy, 2010). A body of research also studied faultlines at the top management team

level (Cooper et al.,2014;Georgakakis et al.,2017;Li and Jones, 2019;Ou et al., 2017;

Richard et al.,2019;Xie et al., 2015). Few board level studies mainly investigated the

faultlines–performance relationship (Kaczmarek et al., 2012b;Van Peteghem et al.,2018;

Vecchiarini et al.,2014;Veltrop et al., 2015), with little emphasis on the formation or

antecedents of faultlines (e.g. presence of chief executive officer on the nomination

committee as an antecedent; Kaczmarek et al., 2012a). Studying formation or antecedents

of faultlines in corporate boards is especially important (Veltrop et al., 2015). Stereotypical

incongruent subgroups (e.g. young female directors vs aged male directors) may be

formed, having a strong impact on board processes (Steger, 2014;Thatcher, 2013). These

negative board processes can have adverse effects on organizational outcomes. Thus,

demographic faultlines may partly explain why board demographic diversity can lead to

negative outcomes (Hafsi and Turgut, 2013;Haslam et al.,2010) or non-signifcant

outcomes (Fern

andez-Temprano and Tejerina-Gaite,2020;Horak and Cui, 2017).

The current study advances the field of faultlines in three ways. First, we extend the

faultlines field by examining the formationof demographic faultlines in corporate boards in a

real-world setting. Most group level studies used field/laboratory experiments with artificial

allocation of individuals to groups to form various faultlines strengths (Homan et al., 2007;

Lau and Murnighan, 2005;Sawyer et al.,2006). The artificial field/laboratory experimental

settings fail to fully capture the faultlines formation possibilities present in the real world

(Chrobot-Mason et al., 2009). Second, we pioneer a new direction in faultlines literature on

the group size–faultlines relationship by examining a curvilinear impact of board size on

faultlines. Research so far has mainly controlled for the effect of groupsize [1]. Specifically,

we present a positive a U-shaped board size–faultlines strength relationship. We test this

relationship using archival data on 288 organizations listed on the Australian Securities

Exchange (ASX).

Theoretical underpinnings and hypotheses development

Formation of gender and age faultlines

Diversity includes any attributes (e.g. gender, age and ethnicity) people use to tell

themselves that another person is different (Williams and O’Reilly, 1998). The formation of

subgroups based on diversity attributes and their alignment can be derived by social

identity theory (Tajfel, 1978;Tajfel and Turner, 1986) and faultlines theory (Lau and

Murnighan, 1998;Thatcher,2013).

The members of a group may start categorizing themselves into various subgroups based

on multiple diversity attributes (Turner et al.,1987). Social identity theory postulates that

following the categorization, members of those subgroups establish a positive social identity

with their subgroup (ingroup) (Tajfel, 1978;Tajfel and Turner, 1986). The members of each

subgroup may start showing favoritism to ingroup members over the members of the other

subgroup (outgroup) (Billig and Tajfel, 1973). The extent that an ingroup m embership is

salient, one’s perceived similarity to others in the ingroup is increased (Hogg, 2006). The

social identity is especially salient when it is linkedwith minority status (e.g. female or young

PAGE 1206 jCORPORATE GOVERNANCE jVOL. 20 NO. 7 2020