Managerial risk aversion and the structure of executive compensation

• Published date: 01 July 2023
• Author: Karel Hrazdil,Jeong Bon Kim,Jiri Novak,Christopher Zatzick
• Date: 01 July 2023
• DOI: http://doi.org/10.1111/corg.12480

ORIGINAL ARTICLE

Managerial risk aversion and the structure of executive

compensation

Karel Hrazdil

1

| Jeong Bon Kim

2

| Jiri Novak

3

| Christopher Zatzick

1

1

Simon Fraser University, Burnaby, Canada

2

City University of Hong Kong, Hong Kong

3

Institute of Economic Studies, Faculty of

Social Sciences, Charles University, Prague,

Czech Republic

Correspondence

Karel Hrazdil, Simon Fraser University,

Burnaby, Canada.

Email: karel_hrazdil@sfu.ca

Funding information

European Union's Horizon 2020 research and

innovation programme under the Marie

Skłodowska-Curie, Grant/Award Number:

870245; Czech Science Foundation, Grant/

Award Number: 21-09231S; Social Sciences

and Humanities Research Council of Canada,

Grant/Award Number: 31-R640084

Abstract

Research question/issue: We examine how chief executive officers' (CEOs) innate

risk aversion influences the size and structure of their compensation contracts. In so

doing, we estimate managerial risk aversion based on the Big Five personality traits—

openness, conscientiousness, extraversion, agreeableness, and neuroticism—inferred

using IBM's Personality Insights service.

Research findings/insights: We provide evidence that executives' inherent risk aver-

sion is related to their compensation structure. Contrary to agency theory predic-

tions, we find that more risk-averse CEOs receive more cash-based and less equity-

based compensation but receive lower total compensation. This relationship is mod-

erated by differences in firms' resource advantages.

Theoretical/academic implications: Despite the theoretical prediction that manage-

rial risk aversion is a key factor determining the structure of executives' compensa-

tion contracts, there is limited empirical evidence on whether firms adjust the

components of compensation based on CEOs' risk preferences. Our results help us

better understand the interplay between CEO personality and executive

compensation.

Practitioner/policy implications: This study offers important implications for organi-

zations in that knowledge about executives' inherent risk aversion is important and

relevant for designing effective compensation contracts.

KEYWORDS

corporate governance, CEO personality, risk aversion, executive compensation, incentive pay

1|INTRODUCTION

Executive compensation has been extensively debated among the

public and academics in recent decades. Chief executive officers

(CEOs) are accused of making astonishing sums of money and taking

excessive risks to maximize short-term returns (Cable &

Vermeulen, 2016). Further, CEOs are portrayed as greedy and over-

confident, even to the detriment of shareholders' returns (Takacs

Haynes et al., 2017). While many people question the need for ridicu-

lously large executive pay, others suggest that these compensation

packages are vital to encourage innovation and risk-taking by CEOs.

Indeed, researchers have examined how CEOs' appetite for risk is

related to a wide range of economic outcomes ranging from principal-

agent contracting to asset pricing and emphasize the importance of

incentives to align executives' interests with those of shareholders

(Chen et al., 2006; Epstein, 1998; Hambrick & Mason, 1984;

Haubrich, 1994; Jensen & Murphy, 1990b; Page, 2018). Yet under-

standing how a leader's innate preference for risk is related to execu-

tive compensation has received relatively little attention in the

literature.

Two theoretical perspectives address how CEOs' risk preferences

are related to the size and composition of executive compensation.

First, a corporate governance perspective, mainly drawing on agency

theory (Jensen & Murphy, 1990b), suggests that firms use

Received: 12 August 2021 Revised: 27 June 2022 Accepted: 29 June 2022

DOI: 10.1111/corg.12480

Corp Govern Int Rev. 2023;31:563–581. wileyonlinelibrary.com/journal/corg © 2022 John Wiley & Sons Ltd. 563

compensation to encourage or discourage CEOs' risk-taking in order

to align owners' and executives' risk preferences. More specifically,

organizations need to increase variable pay to incentivize risk-taking

behavior from CEOs who are inherently more risk-averse than owners

(Dittmann et al., 2017; Dittmann & Maug, 2007; Hölmstrom, 1979).

This perspective views leaders as passive recipients of pay and pay

structure, as determined by owners (i.e., shareholders) and boards of

directors (including compensation committees). Despite the theoreti-

cal prediction that managerial risk aversion is a key factor determining

the structure of executives' compensation contracts, there is limited

empirical evidence on whether firms adjust the proportion of cash

and equity-based compensation based on CEOs' risk preferences

(Gray & Cannella, 1997). Indeed, granting stock options to risk-averse

executives may not necessarily increase their appetite for risk

(Armstrong & Vashishtha, 2012; Carpenter, 2000).

The second perspective, based on executive power (e.g., Bebchuk

et al., 2002; Westphal & Zajac, 1995) and upper echelons

(Hambrick & Mason, 1984) research, focuses on how individual differ-

ences, such as CEOs' abilities and personal characteristics, influence

strategic actions (e.g., board composition, diversification, and acquisi-

tions), and organizational outcomes (e.g., innovation and performance)

(Wang et al., 2016). From this perspective, executives not only choose

jobs that offer compensation commensurate with their risk prefer-

ences (Graffin et al., 2020) but also influence various factors that

determine ongoing executive compensation decisions made by an

organization (O'Reilly et al., 2014; Tosi et al., 2004). For example,

CEOs meet with boards of directors and appoint compensation chairs,

giving them the potential to influence policies and strategies to their

advantage (O'Reilly et al., 2014). Hence, from this perspective, execu-

tive compensation is a result of CEOs' preferences and actions, rather

than a governance mechanism used by owners to align executives'

interests with their interests.

In this study, we seek to integrate these perspectives by viewing

executive compensation as jointly determined by CEO risk prefer-

ences and owners' desire to use equity-based compensation to incen-

tivize managers to make an effort and take risks, creating the need for

compensation bargaining between a risk-neutral owner and a risk-

averse manager (Bolton & Dewatripont, 2005; Hölmstrom, 1979;

Lambert, 2001). In the context of finance and economics, risk aversion

reflects an individual's preference for lower returns at a known level

of risk (i.e., certainty) over the potential for higher returns at a higher

level of risk (i.e., uncertainty). An individual exhibiting risk aversion is

said to be risk averse, where a concave shape of his/her utility func-

tion represents risk aversion. Consistent with recent empirical evi-

dence (e.g., Graffin et al., 2020; Graham et al., 2013), we expect that

risk-averse CEOs will prefer more cash compensation (salary and

bonus) and less equity compensation (options and stocks) compared

to more risk-tolerant executives. However, as part of a trade-off for

the reduced risk of cash compensation, risk-averse CEOs will receive

lower overall total compensation. Similarly, we propose that firms are

willing to accept such a trade-off because matching CEOs' risk prefer-

ences and compensation structure will not only help to attract execu-

tives but also lower overall compensation costs for the firm.

Finally, we examine a potential moderator of this relationship to

help tease apart the aforementioned competing perspectives. In par-

ticular, we explore whether and how the relations of CEO risk aver-

sion with the size and composition of executive pay are influenced

systematically by the efficiency of bargaining conditions (i.e., equal

distribution of power). Consistent with Pandher and Currie (2013), we

examine how the relative bargaining positions of CEOs and owners

are shaped in relation to a firm's growth position. Taken together, our

study examines the fundamental trade-off between executives' innate

risk preferences and owners' desire for growth and risk and the influ-

ence of efficient contracting conditions on both the total and struc-

ture of executive compensation.

To date, research linking CEO risk preferences to compensation

structure has been limited by the challenge of assessing CEO person-

ality traits in large, generalizable samples. One research stream uses

managerial fixed effects as a “catch-all”proxy for personality differ-

ences (including risk aversion) among executives (e.g., Dyreng

et al., 2010; Ge et al., 2011), while another infers executives' risk aver-

sion from demographic and personal characteristics, such as country

of origin, education, field of expertise, political orientation, and ethnic-

ity (e.g., Bamber et al., 2010; Ellahie et al., 2017; Graffin et al., 2020;

Hambrick & Mason, 1984) or infers executives' risk appetite based on

their actions such as exercising stock options, the amount of options

they are granted, or having risky hobbies (e.g., Hribar & Yang, 2016;

Kim et al., 2016; Malmendier & Tate, 2005,2008; Sunder

et al., 2017). While such indirect measures are useful, these

approaches do not allow researchers to capture specific differences

within broad personality categories. A more direct test linking CEOs'

risk preferences to compensation was provided by Graham et al.

(2013) and Holt and Laury (2002), who assessed CEOs' risk prefer-

ences using a survey of CEO responses to safe or risky alternatives in

various hypothetical scenarios. Yet, given the limitations of this type

of survey approach (Tourangeau & Yan, 2007)

1

and the general lack

of empirical evidence on how innate executive personality traits relate

to compensation structure, we are motivated to investigate our ques-

tions using a new, innovative method that allows objective assess-

ment of personality in a large sample setting.

This study utilizes a novel and recently validated approach based

on a machine learning technique that estimates the Big Five personal-

ity traits (openness, conscientiousness, extraversion, agreeableness,

and neuroticism [OCEAN]) from speeches made by CEOs. Specifically,

we use the Watson Personality Insights service (Watson PI, developed

by IBM) to process executives' answers to questions posed by ana-

lysts during the Q&A sessions of conference calls. Our study focuses

on these Big Five traits because they portray basic orthogonal, under-

lying trait dimensions of personality (Goldberg, 1990) and are recog-

nized as genetically based, relatively stable, and cross-culturally

generalizable (Cobb-Clark & Schurer, 2012; Costa & McCrae, 1997).

Based on prior research that provides relatively consistent guidance

on the relation between the Big Five personality traits and an individ-

ual's risk appetite (Borghans et al., 2009; Clarke & Robertson, 2005;

Judge & Bono, 2000), we combine the Big Five personality traits to

derive an inherent index of CEO risk aversion (RA).

2

This measure

564 HRAZDIL ET AL.