Ideological homophily in board composition and interlock networks: Do liberal directors inhibit viewpoint diversity?

• Published date: 01 May 2022
• Author: Kerry Hudson,Robert E. Morgan
• Date: 01 May 2022
• DOI: http://doi.org/10.1111/corg.12406

ORIGINAL ARTICLE

Ideological homophily in board composition and interlock

networks: Do liberal directors inhibit viewpoint diversity?

Kerry Hudson

1

| Robert E. Morgan

1,2

1

Department of Marketing and Strategy,

Cardiff Business School, Cardiff University,

Cardiff, UK

2

Department of Marketing, Copenhagen

Business School, Frederiksberg, Denmark

Correspondence

Kerry Hudson, Department of Marketing and

Strategy, Cardiff Business School, Cardiff

University, Colum Drive, Cardiff CF10 3EU,

UK.

Email: hudsonkl@cardiff.ac.uk

Abstract

Research issue: A consistent feature of social networks is homophily: the tendency

for people to interact with similar others. Psychological and sociological research sug-

gests that homophily is most pronounced along ideological lines, with conflicting evi-

dence as to whether this tendency is higher among individuals who hold liberal or

conservative beliefs. Based on this literature, we conduct the first study of ideological

homophily in two key organizational networks: the intrafirm connections among

directors on the board and the interfirm connections created by board interlocks.

Research insights: In a panel of 408 U.S. firms between 2000 and 2020, we find that

liberalism increases homophily both within and between boards. Furthermore, we

find that homophily has decreased over time but that this has been driven by conser-

vative boards while the effect of liberalism has strengthened in recent years. These

findings provide the first evidence for an ideological component in the composition

of intra- and interorganizational networks.

Academic implications: Most research on director selection and interlock formation

has focused on situational or demographic antecedents. Our findings contribute to

the development of a broader theoretical framework that accounts for individual dis-

positional factors in these processes.

Practitioner implications: Our findings bring attention to the issue of ideological

homogeneity in firms. Given the growth of homophilic tendencies among liberal

directors in recent years, we suggest that it may be increasingly important for direc-

tors to become aware, and mitigate the effects, of their ideological biases in order to

maintain cognitive diversity in information networks and decision making.

KEYWORDS

corporate governance, board interlocks, boards of directors, homophily, social networks

1|INTRODUCTION

The board of directors is the “apex of decision control”(Fama &

Jensen, 1983, p. 311), setting the strategic direction and objectives of

the firm (Bailey & Peck, 2013). Board interlocks—formed when a

director serves on the board of two firms (Mizruchi, 1996)—are a key

conduit of information for boards' decision making, providing access

to market intelligence (Yoshikawa et al., 2019), aiding in the diffusion

of new and best practices (Beckman & Haunschild, 2002), and open-

ing access to critical resources (Withers et al., 2012). Consequently,

Received: 22 May 2021 Revised: 15 September 2021 Accepted: 17 September 2021

DOI: 10.1111/corg.12406

272 © 2021 John Wiley & Sons Ltd Corp Govern Int Rev. 2022;30:272–289.wileyonlinelibrary.com/journal/corg

the composition of the board and the firm's position within interlock

networks are pertinent topics in organizational research for two rea-

sons: (1) Interlocks affect the volume and content of interfirm infor-

mation flows (Li, 2019; Yoshikawa et al., 2019), and (2) the cognitive

frameworks of directors influence how this information is used in

decisions (Bailey & Peck, 2013; van Ees et al., 2009).

Both board composition (Withers et al., 2012) and interlock for-

mation (Bazerman & Schoorman, 1983) are consequences of the

social embeddedness of corporate boards, being substantially

influenced by social and individual factors beyond the economic con-

siderations of the firm and its shareholders (van Ees et al., 2009;

Westphal & Zajac, 1995). Specifically, the appointment of new

directors—and thus the formation of board interlocks—is necessarily

limited by extant social connections and dependent upon interper-

sonal political factors and individual biases (Bazerman &

Schoorman, 1983; Withers et al., 2020). Antecedents to the composi-

tion of boards and interlock networks are therefore both situational,

pertaining to the operating environment of firms or social context of

interpersonal interactions, and dispositional, related to the cognitive

and affective biases of individuals (c.f. Kelley, 1973).

Most research to date has examined situational factors, leading to

a theoretical understanding of board composition and network forma-

tion that may understate the role of directors' cognitive and affective

frames (Gupta & Wowak, 2017; Shropshire, 2010), despite

longstanding recognition that the values, beliefs, and attitudes of deci-

sion makers affect firm-level outcomes (Chin et al., 2013). Research

on top management teams (TMTs) (Chin et al., 2013; Chin &

Semadeni, 2017; Christensen et al., 2015; Gupta et al., 2020; Hutton

et al., 2014; Kashmiri & Mahajan, 2017), and some notable exceptions

to the situational focus in board research (di Giuli &

Kostovetsky, 2014; Gupta & Wowak, 2017; Park et al., 2020), high-

light a key dispositional factor: ideology. This refers to an individual's

internally consistent belief system, comprising the attitudes and

values that underlie thought and behavior (Jost, 2006; Tedin, 1987),

and is observable and measurable by political orientations (Chin

et al., 2013; Erikson & Tedin, 2003; Jost et al., 2009). The liberal–

conservative spectrum is most commonly applied, as the distinction

has remained stable over time (Jost, 2006); predictably correlates with

personality traits (Gerber et al., 2011), cognitive biases (Fatke, 2017),

and values (Carney et al., 2008); and provides a framework for action

across a range of domains (Jost et al., 2009).

Accordingly, there is evidence for effects of decision makers' ide-

ologies on a range of firm outcomes. Liberal chief executive officers

(CEOs) are more likely to engage in corporate social responsibility

(CSR) (Chin et al., 2013) and appoint CSR executives (Gupta

et al., 2020) and have a higher rate of new product introductions

(Kashmiri & Mahajan, 2017). Firms with conservative managers have

lower levels of debt, higher profitability, and less risky investments

(Hutton et al., 2014), greater pay dispersion within the TMT (Chin &

Semadeni, 2017), and lower rates of tax avoidance (Christensen

et al., 2015). At the board level, conservativism is associated with

higher CEO compensation and a stronger correlation between com-

pensation and performance (Gupta & Wowak, 2017), higher rates of

CEO dismissal following financial misconduct (Park et al., 2020), and

lower adoption of CSR policies (di Giuli & Kostovetsky, 2014). These

outcomes are predictability aligned with the personality characteris-

tics typically associated with each pole of the political spectrum, such

as differences in risk tolerance and perceptions of fairness (c.f. Gerber

et al., 2010; Haidt, 2001). To date, however, there have been no stud-

ies of the influence of ideology on the structure of board interlock

networks and the position of the firm within these (Gupta &

Wowak, 2017), despite evidence that the ideologies of peer firms are

salient to decision makers at the TMT level (Gupta et al., 2020). Simi-

larly, the relationship between ideology and board composition has

only been studied tangentially to the monitoring effectiveness of

inside and outside directors (Kim et al., 2013).

We posit that director ideology is an overlooked dispositional

antecedent to board composition and interlocks. We base this asser-

tion in the psychological literature on homophily,a“remarkably consis-

tent structural feature”of social connections whereby individuals

demonstrate a preference for forming ties to similar others

(McPherson et al., 2001, p. 429). Thus, director appointment may

preferentially select for ideologically similarity to incumbent board

members. As this process also determines the structure of the board

interlock network, we examine two outcomes: board ideological

homophily, the degree of homogeneity in political orientations among

directors, and network ideological homophily, the degree of homogene-

ity in political orientations of the boards to which a focal firm

connects.

We make an ostensibly counterintuitive prediction: liberalism will

increase ideological homophily, such that boards with more liberal

directors will exhibit less viewpoint diversity within the board and

establish fewer ideologically incongruent interlocks. This conflicts with

the stereotype of the “open-minded liberal”(Jost et al., 2003), but

aligns with studies of social and professional networks (Inbar &

Lammers, 2012: e.g., Yoo et al., 2018), and psychological evidence

(e.g., Brandt et al., 2014; Crawford et al., 2017) that indicates greater

ideological intolerance among liberals. We examine why these dis-

crepancies have emerged and posit that the social context of the

board is likely to induce the latter effect, with liberals' beliefs about

the social purpose of business encouraging the maintenance of ideo-

logical homogeneity.

Analysis of data on board composition and interlocks from

408 large U.S. firms between 2000 and 2020 demonstrates that board

liberalism increases homophily at both the intra- and inter-

organizational level. Furthermore, despite overall levels of ideological

homophily decreasing over the 20 years of our sample, the effect of

liberalism on board and network homophily has increased. This sug-

gests that increases in the ideological diversity of boards and interlock

networks have been primarily driven by conservative directors.

This study provides the first theoretical rationale and empirical

evidence for an ideological component in the composition of boards

and the structure of interlock networks, with implications for under-

standing the dispositional antecedents to director selection. Notably,

our findings run counter to the long-held assumption in political psy-

chology of the “rigidity of the right,”that is, the attribution of

HUDSON AND MORGAN 273