International differences in the CEO gender pay gap

• Published date: 01 September 2022
• Author: Xiaoqi Chen,Wouter Torsin,Albert Tsang
• Date: 01 September 2022
• DOI: http://doi.org/10.1111/corg.12421

ORIGINAL ARTICLE

International differences in the CEO gender pay gap

Xiaoqi Chen

1

| Wouter Torsin

2

| Albert Tsang

3

1

Institute for Financial and Accounting Studies,

Xiamen University, Xiamen, China

2

HEC Management School, University of

Liège, Liège, Belgium

3

The Hong Kong Polytechnic University,

Kowloon, Hong Kong

Correspondence

Xiaoqi Chen, Institute for Financial and

Accounting Studies, Xiamen University,

5rd Floor, Jiageng 2 Building,

422 South Siming Road, Siming District,

Xiamen, Fujian Province, China.

Email: chen.xiaoqi001@gmail.com

Abstract

Research question/issue: This study examines (i) whether and to what extent differ-

ences in the level of compensation paid to men and women can be observed at the

corporate executive (i.e., CEOs) level in an international setting and (ii) whether and

how country-level attitudes toward gender equality help explain the heterogeneity in

the CEO gender pay gap observed across countries.

Research findings/insights: Using executive compensation data from 27 countries

over the 2001–2016 period, this study documents a 3.3% average difference in the

level of executive compensation between male and female CEOs in countries around

the world. The results further indicate that firms that transition from a male to a

female CEO tend to have a significantly lower level of compensation. These results

are robust when controlling for an array of key CEO attributes including education,

age, and work experience and when US firms are excluded from the sample. Our

findings also suggest that attitudes related to gender equality across countries can be

an important factor explaining the CEO gender pay gap observed in our study.

Theoretical/academic implications: The results of our study suggest the existence of

a CEO gender pay gap at the international level. Specifically, our study suggests that

the difference in pay between men and women, which is well documented in the lit-

erature of many countries, not only applies to regular employees in an economy but

can also be generalized to those who work at the chief executive level. More impor-

tantly, our study suggests that a country's institutional environment vis-à-vis gender

equality can contribute to the gender pay differences observed in various countries

around the world.

Practitioner/policy implications: We provide evidence supporting the existence of a

gender pay gap using executive compensation data at the CEO level. We further

show that country-level institutional characteristics related to gender equality partly

contribute to gender pay differences between male and female CEOs across coun-

tries. Thus, the findings of our study provide new insights into the potential determi-

nants of executive compensation (i.e., CEO gender and country-level gender equality)

and are valuable to policymakers in proposing legislative changes to reduce the

gender pay gap.

KEYWORDS

corporate governance, executive compensation, international, equality, institutions

Received: 4 March 2020 Revised: 18 October 2021 Accepted: 19 November 2021

DOI: 10.1111/corg.12421

516 © 2021 John Wiley & Sons Ltd Corp Govern Int Rev. 2022;30:516–541.wileyonlinelibrary.com/journal/corg

1|INTRODUCTION

Common wisdom surrounding the gender pay gap dictates that male

employees are remunerated more than female employees for the

same work, and this practice is echoed by a host of statistics publi-

shed by governments and government agencies around the world.

1

According to the World Economic Forum, it will take 257 years to

close the gender pay gap at the rate observed between 2006 and

2020. However, it is not clear whether there are gender pay differ-

ences in firms' upper echelons of management. Although female chief

executive officers (CEOs) are a minority in both developed and devel-

oping economies (see Fang & Huang, 2017; Oakley, 2000), the find-

ings on CEO pay across gender are mixed at best and are country-

specific (see Adams et al., 2007; Lam et al., 2013). This study conducts

a cross-country examination to document whether there are global

differences in CEO gender pay and why such variation occurs.

Grounded in role congruity theory (Eagly, 1987) and the existence

of stereotypes that dissociate female characteristics with leadership

qualities (see Schein et al., 1996), male CEOs are expected to receive

a premium over female CEOs. This view is reinforced by contempo-

rary media sources (see, e.g., Equilar, 2016,2017) that often

univariately compare average compensation without considering

appropriate control variables. Studies using multivariate analyses are

much less unequivocal. Research on CEO gender pay disparities is

largely skewed toward the US (see, among others, Bertrand &

Hallock, 2016; Adams et al., 2007; Bugeja et al., 2012; Gupta

et al., 2018), in which virtually no significant differences are observed.

In contrast, the limited studies outside the US lean in favor of a male

remuneration premium. Studies with supporting evidence include Lam

et al. (2013) for China and Keloharju et al. (2016) for Sweden. Mean-

while, Geiler and Renneboog (2015) and Ellwood et al. (2020) fail to

find evidence for the UK. These conflicting findings across the globe

call for a broader cross-country examination. Therefore, the first con-

tribution of this study is to empirically test the presence of an interna-

tional CEO gender pay gap, across firms from both developed and

developing economies. Specifically, using executive compensation

data from 1,352 female CEOs and 17,656 male CEOs from 27 differ-

ent countries over the 2001–2016 period, we find evidence

suggesting the existence of a global female penalty in remuneration,

in that the compensation of female CEOs is about 3.3% lower than

that of their male counterparts.

The second purpose of this study is to understand why the litera-

ture finds differences in CEO gender pay in some countries but not in

others. We argue that social norms and institutions differ across coun-

tries, which can then affect women's remuneration and opportunities

(see Guthrie & Roth, 1999; Lam & Dreher, 2004). Social norms are

particularly important in explaining gender wage disparities at the

aggregate level. For example, Algan and Cahuc (2006) report an

increase in the overall gender pay gap in countries with higher levels

of religiosity. Fortin (2005) finds that anti-egalitarian views provide a

strong explanation for gender pay differences. Triventi (2013) exam-

ines gender pay differences among EU graduates and reports that the

“wage determination is deeply rooted in the institutional contexts”

(p. 575) and concludes that more egalitarian policies hinder gender

pay discrimination. Lending support to this view, we empirically dem-

onstrate that the female penalty in CEO remuneration is significantly

lower in countries that are more socially and institutionally egalitarian.

The results of this study have implications for policymakers and

market participants. Given the ongoing debate on gender pay dispar-

ities in general, our paper provides evidence of this phenomenon at

the CEO level. This finding is of particular importance to policymakers

because it seems that even in positions of power, societal norms and

prejudices have a significant impact on the wage setting mechanism.

Moreover, our finding suggests that the extent to which there is a

CEO gender pay gap varies with the norms and regulations protecting

equal treatment at the country level. Beyond its direct relevance to

female CEOs, the findings of this study should also be useful to inves-

tors and firms. Given investors' growing demand for more corporate

social responsibility and gender equality in their portfolio strategies

(see Lublin & Krouse, 2017), progressive firms will benefit from under-

standing that their remuneration policy may be biased against women

(perhaps unconsciously).

The remainder of this paper is organized as follows. Section 2

provides a review of the literature and develops the hypotheses.

Section 3describes the construction of the sample, defines the vari-

ables, and discusses the research design. Section 4presents the main

empirical results and the results of cross-sectional tests. Section 5dis-

cusses the findings, while Section 6concludes the paper.

2|THEORY AND HYPOTHESES

Theoretically, gender-based discrimination can be explained by the

existence of stereotypes that favor men over women (see

Heilman, 1983; Heilman & Okimoto, 2007) which are particularly

salient with regard to leadership positions (Eagly & Carli, 2003;

Stuhlmacher & Poitras, 2010). More precisely, role congruity theory

(Eagly, 1987) states that there is an incongruence between the female

gender role and the expectations of leadership qualities: Women are

seen as nurturing and caring, and such traits are not favorable for

management positions (Koenig et al., 2011). Using survey evidence,

Schein (1973) finds that being a manager is strongly associated with

being a man, which is often referred to as the “think manager–think

male”stereotype (Schein et al., 1996). Meanwhile, Powell and

Butterfield (1979,1989) and Powell et al. (2002) perform a three-

wave study among students regarding managerial stereotypes and

valuable traits, in which the respondents consistently considered

female traits to be irrelevant for good leadership. This is often

referred to as the “think manager–think masculine”stereotype.

2

Not surprisingly, gender stereotypes have been linked to gender

pay disparities in general. For example, Moss-Racusin et al. (2012) find

that male and female applicants for lab manager positions at research-

intensive universities in the US receive differential treatment and

female applicants are offered significantly lower starting salaries.

Roth (2006) studies gender inequality on Wall Street and finds that

women earn only 60.5% of the salary of their male peers. In particular,

CHEN ET AL.517