Social Justice issues

Genderpaygap2.pdf

Home >Psychology homework help >Social Justice issues

Previous Next

Journal of Business Research Volume 180, July 2024, 114710

The more the merrier? Exploring the effect of women on boards on the gender pay gap in top management teams Gurdeep Singh Raina , Arvin Sahaym , Leah D. Sheppard

Outline

https://doi-org.libproxy.aum.edu/10.1016/j.jbusres.2024.114710

Get rights and content

Full text access

Abstract

We examine the relationship between gender composition of boards of directors and the gender pay gap found in

top management teams (TMTs), as well as the extent to which this relationship is influenced by Chief Executive

Officer (CEO) gender. Using observations of 2,658 U.S. firms taken over 14 years, we find an inverted U-shaped

relationship between board gender composition and the TMT pay gap, such that increasing the proportion of

women on a board from zero to roughly 15 percent widens the pay gap whereby men are paid more than women.

However, as the proportion of women on boards increases beyond 15 percent, the pay gap narrows and continues

to do so. Moreover, the presence of a woman CEO strengthens the narrowing effect on the gender pay gap that an

increased proportion of women on the board exerts. Our findings are consistent with the integrated insights we

draw from queen bee, critical mass, and upper echelons theories. This paper discusses the implications of our work

for research and practice.

Keywords

Board gender composition; Gender pay gap; Top management teams; Queen bee syndrome; Critical mass theory;

Upper echelons theory

a b b

Share Cite

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 1 of 25

1. Introduction

The gender pay gap in the upper echelons of organizations persists despite the increased presence of women at all

levels, including the uppermost ranks (Kalogeraki and Georgakakis, 2021, Krishnan and Park, 2005, Kulich et al.,

2011). This is an important issue to investigate since along with promoting fairness and improving well-being

among employees, organizations need to enhance their ability to attract and maintain top performers (Blau and

Kahn, 2003, Ingersoll et al., 2023). Recent data show that women now constitute 32 % of boards and nearly 8 % of

executive officers in S&P 500 companies (Catalyst, 2023, Stuart, 2023), but earn approximately 8 % less than men in

such roles (Keller et al., 2023) after controlling for variables such as job title and organizational characteristics.

Bringing talented women leaders into the upper ranks, compensating them fairly, and retaining them will be

critical in resolving gender-based pay disparities throughout the organization, given the influence that women

leaders can have over the policies and practices that enhance diversity, equity, and inclusion (Cook et al., 2019, Shin,

2012).

Scholars have explored the under-representation of women at senior levels and have recommended greater

inclusion to reduce organizational disparities and facilitate the entry of women and minorities into leadership

roles. Some studies have confirmed the positive influence of women CEOs on organizational equity (e.g., Cook &

Glass, 2016) in a manner consistent with upper echelons (Hambrick & Mason, 1984) and social identity (Tajfel &

Turner, 1985) theories, but have overlooked the influence of boards of directors (Chin and Semadeni, 2017,

Kalogeraki and Georgakakis, 2021), which exert influence in consultation with CEOs and other top executives over

issues such as compensation (Conyon, 2014, Cook et al., 2019, Ho et al., 2022, Ozkazanc-Pan and Clark Muntean,

2018). Such entities are subject to group dynamics that can complicate the straightforward narrative that women in

key leadership roles will bring about positive changes to gender equity. For example, critical mass theory predicts

that simply placing a token woman in a critical leadership role is unlikely to be impactful, since people in such roles

are expected to conform to the views and behaviors of the dominant group found in c-suites and on boards of

directors, i.e., men (Carter et al., 2017, Cook and Glass, 2016, Kanter, 1977). Moreover, queen bee theory asserts that

senior women leaders, especially those who are tokens, may distance themselves from other women and issues

related to gender inequity to buffer the threat posed by other women who are perceived as competing for limited

roles for women leaders (Corwin et al., 2021, Derks et al., 2016, Duguid, 2011, Staines et al., 1974).

In the current work, we endeavor to integrate the insights of critical mass, queen bee, and upper echelons theories

to examine how the proportion of women on boards is related to the TMT gender pay gap. We propose that as the

proportion of women on boards increases from zero, the gender pay gap in TMTs (favoring men) will initially widen,

as token women board members endeavor to distance themselves from other women and matters related to gender

equity. However, as the proportion of women board members reaches and surpasses critical mass, the pay gap will

narrow when perceptions of threat are alleviated by the enhanced presence of women in elevated roles (Kalogeraki

& Georgakakis, 2021). Finally, we propose that the presence of a woman CEO should weaken the negative effects of

a sub-critical mass proportion of women on the board, but then matters less as the proportion of women on the

board reaches and surpasses critical mass. Our hypotheses were tested based on observation from 2658 U.S. firms

over 14 years.

We offer significant contributions through this work. First, we synthesize the insights of three unique theoretical

lenses to propose a nuanced view of the phenomena under study, going beyond the dominant view which predicts

a linear relationship between the proportion of women on boards and the size of the gender pay gap among TMT

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 2 of 25

members (e.g., Carter et al., 2017). This is important because while previous work has acknowledged that token

women will have little impact, the results of our work suggest that a slight increase in women holding key

leadership roles may actually exacerbate existing gender inequities. Our work further distinguishes itself from

previous research by exploring the joint influence of women board members and a woman CEO on the TMT gender

pay gap, ultimately providing a more comprehensive picture of the forces that either exacerbate or mitigate

organizational gender inequities. Finally, we contribute to what is currently a scarcity of research examining factors

that influence the gender pay gap among TMT members specifically, who are responsible for guiding the strategy of

the organization (Benson-Armer et al., 2015, Chakravarthy and Doz, 1992, Singh et al., 2021). Given the significant

influence of top management on firm performance, understanding and resolving gender inequities among these

entities is important for retaining talented performers and realizing the benefit of enhanced diversity among top

leaders.

2. Theoretical foundation and hypotheses

2.1. Relationship between board gender composition and the gender gap among TMT members

Board gender diversity is a topic of great interest in organizational research, and numerous studies demonstrate

that greater gender diversity on boards is associated with positive organizational outcomes (Tonoyan & Olson-

Buchanan, 2023), including improved financial performance, enhanced corporate governance, and greater strategic

innovation (Post & Byron, 2015). Specifically, board gender diversity can be an important driver of firm innovation

(Makkonen, 2022, Tonoyan and Olson-Buchanan, 2023) and corporate social responsibility (Liao et al., 2018), and

produce higher returns on assets (ROA) and equity (ROE) (Bertrand et al., 2010, García-Meca et al., 2015).

Business leaders often assume that simply assigning a woman to a key leadership role will reduce gender inequity,

believing they will bring other women up through the ranks to identify and resolve issues related to gender bias

(Diehl et al., 2022). This assumption has some theoretical foundations. Indeed, social identity theory posits that

individuals derive a sense of identity from their social group memberships and prefer to affiliate with in-group

members over out-group members (Tajfel & Turner, 1985). Moreover, individuals tend to view members of their in-

group through rose-colored glasses, endowing them with favorable traits and demonstrating a willingness to

support them (Balliet et al., 2014, Halevy et al., 2008). Some studies have shown that women generally exhibit

stronger in-group favoritism than men (Rudman & Goodwin, 2004). Therefore, it seems reasonable that as women

break through the glass ceiling into leadership positions, they will support and advocate for other women.

Of course, various factors complicate this narrative. First, women leaders may feel their gender is highly visible

simply due to their token or marginalized status (Eagly & Carli, 2007) and, as a result, refrain from dealing with

matters of gender equity (Kanter, 1977). As such, women who comprise a very small minority of leaders are not

likely to have much impact on measures of gender equity throughout the organization. Moreover, women who

assume high-ranking roles may take on more masculine characteristics than feminine ones and distance

themselves from other women in their organizations, a phenomenon known as “queen bee syndrome” (Derks et al.,

2016, Duguid, 2011, Kalogeraki and Georgakakis, 2021, Staines et al., 1974).

A critical factor in determining whether women support or distance themselves from other women in an

organization is how many similar others occupy their ranks (Ely, 1995, Watkins et al., 2019). When women leaders

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 3 of 25

are considered tokens, women might perceive that there is a scarcity of high-ranking roles available to them (Ely,

1995). They might therefore perceive other women as posing a competitive threat, especially when they see other

women working their way up through the ranks and fear that they will lose their unique positions (Duguid, 2011;

Duguid, Lloyd, & Tolbert, 2012; Staines et al., 1974). Such women may distance themselves from other women and

from issues related to gender equity, and instead identify with and behave more like the dominant group, i.e., men

(Derks et al., 2016, Duguid et al., 2012, Kanter, 1977). Indeed, research has shown that token women on high-status

teams are more inclined to view another high-performing woman as a competitive threat, and are thus less

inclined to facilitate her addition to their teams (Duguid, 2011). Likewise, while it might be expected that women

CEOs elicit greater pay equity in TMTs than men CEOs, research suggests that conservative women CEOs are more

likely to perpetuate the gender pay gap than some of their men counterparts (Kalogeraki & Georgakakis, 2021).

Taken together, this research suggests that we should not expect to see that an initial increase in the proportion of

women on the board from zero will close the gender pay gap among TMT members, and that women as token

board members may actually exacerbate the TMT gender pay gap (Derks et al., 2016, Ely, 1995). However, once

women achieve a critical mass on the board (Kanter, 1977), they may be less inclined towards in-group distancing

and exhibit a greater willingness to enact changes that produce more gender equity (Graham et al., 2020, Karpowitz

et al., 2012, Pickett & Brewer, 2001, Torchia et al., 2011). The exact proportion of women needed to achieve a critical

mass in groups has been debated, but most scholars place it in the range of 15 to 30 percent (see Childs & Krook,

2008, for a review).

Accordingly, we predict that an inverted U-shaped relationship will exist between board gender composition and

the TMT gender pay gap (favoring men). That is, we predict that the pay gap favoring men will initially widen as the

proportion of women on a board increases from zero. As the proportion of women on the board reaches and

surpasses critical mass, however, we expect that the TMT gender pay gaps will begin to narrow and continue to do

so. We offer the following hypothesis:

Hypothesis 1 An inverted U-shaped relationship exists between the proportion of women on the board and the TMT gender pay

gap, such that as the proportion of women on the board grows up until critical mass, the TMT gender pay gap will

widen. As the number of women reaches and surpasses critical mass, the TMT gender pay gap will begin to narrow

and continue to do so.

2.2. The interactive effect of CEO gender and board gender composition on the TMT pay gap

CEOs occupy the most significant executive position in an organization and hold positional power based on

legitimate hierarchical status (Daily and Johnson, 1997, Ma et al., 2022). Upper echelons theory suggests that as the

head of the TMT, the CEO plays a key role in decision-making, and that the nature of their decisions is shaped by

their personal characteristics (e.g., identity, values, personalities, professional experiences, demography, etc.)

(Hambrick et al., 2015, Hambrick and Mason, 1984). Using their formal authority, CEOs may align organizational

practices with their own values, including resolving organizational inequities (Bertrand et al., 2010, Blevins et al.,

2019, Corwin et al., 2021, Finkelstein, 1992).

Given the positional power held by the CEO, we would expect that when the proportion of women on a board is

below critical mass, a woman CEO may buffer against the widening of the TMT gender pay gap that we predict in

Hypothesis 1. That is, we suggest that, together with tokenized women on their boards, women CEOs achieve the

effect of critical mass. However, as the proportion of women board members reaches and surpasses critical mass,

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 4 of 25

the presence of a woman CEO will matter less, thereby exhibiting less influence over the scope of the TMT gender

pay gap. Accordingly, we offer the following hypothesis:

Hypothesis 2 CEO gender will moderate the curvilinear relationship between board gender composition and the TMT gender pay

gap, such that in the presence of a woman CEO, the positive relationship between the proportion of women on the

board and the TMT gender pay gap that is predicted when the proportion of women on the board is below critical

mass will be weakened. Once critical mass is surpassed, the presence of a women CEO will not moderate the

relationship between the proportion of women on the board and the TMT gender pay gap.

3. Method

3.1. Sample

We tested our hypotheses using a sample of U.S.-based public firms active from 2005 to 2018. We initially collected

data from 2000 onward, but due to data missing from 2000 to 2005 for most organizations, we chose 2005 as the

starting year of our analyses. We selected companies whose data were publicly available. Some data were collected

from the Wharton Research Data Services (WRDS) database, which provides information on board gender

composition, board independence, and TMT compensation data. We also accessed BoardEx and Compustat-

CapitalIQ, Financial Ratio Suite for information not accessible on WRDS. We initially collected data from 3,520

organizations. After dropping firms with no data on our dependent variable (TMT compensation), a sample of 2,658

organizations resulted, yielding approximately 37,000 firm-year observations. Table 1 identifies the data sources for

variables and how they were operationalized.

Table 1. Variables, source, and measurement.

ROA (Return on assets) WRDS Ratio (net profit/ total assets)

Log of revenues WRDS Log of total revenues

Net margin WRDS Net profit margin (net profit/ total revenues)

Board independence BoardEx Percentage of outside directors on the board

Board size BoardEx Total number of directors on the board

Proportion of non-US board members BoardEx Percentage of directors of nationality other than US on the board

Independent Variables

Proportion of women on the board BoardEx Percentage of women directors on the board

CEO gender BoardEx Gender of the CEO, 1 = Female, 0 = Male

3.2. Dependent, Independent, and moderator Variables

Control Variables Source Measurement

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 5 of 25

Our dependent variable was the TMT gender pay gap. We operationalized this as the difference between average

pay of men and women TMT members. Positive scores indicated higher average pay for men than for women. In the

few cases where women received greater compensation, the difference between compensation for men and women

members was negative in sign. This precluded us from log transforming the data to reduce skewness (Gupta et al.,

2018). A similar variable, horizontal pay gap, was used by Devers et al. (2007), although its context differed from

ours.

We included the following roles as TMT members: Chief Finance Officer (CFO), Chief Operating Officer (COO), Chief

Marketing Officer (CMO), Chief Human Resource Officer (CHRO), Chief Information Officer (CIO), Chief Technology

Officer (CTO), Chief Administrative Officer (CAO), president, senior executive vice president (SEVP), senior vice

president (SVP), executive vice president (EVP), vice president (VP), general manager (GM), company secretary (CS).

Table 2 provides the list of roles determined to comprise the TMT in our study.

Table 2. TMT Roles.

CAO Chief Administrative Officer

CFO Chief Financial Officer

CHRO Chief Human Resource Officer

CIO Chief Information Officer

CMO Chief Marketing Officer

COO Chief Operating Officer

CS (Secy.) Company Secretary

CTO Chief Technology Officer

EVP Executive Vice President

GC General Counsel

GM General Manager

President President

SEVP Senior Executive Vice President

SVP Senior Vice President

VP Vice President

Our independent variable was the proportion of women on the board. We operationalized this variable by taking

the number of women directors on boards and dividing them by the total number of board members (Guldiken et

al., 2019).

Our moderator was CEO gender, which was coded 1 = woman and 0 = man. The WRDS database identified the

gender of each CEO.

TMT Title Description

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 6 of 25

3.3. Control Variables

A review of relevant literatures guided our selection of control variables, which included company revenue (log of

revenues in USD), organization size (number of employees), return on assets, net margin (net income as a

percentage of revenue for sample organizations), board size (total number of board members), board independence

(the proportion of outside board members), nationality mix on the board (proportion of non-U.S. board members),

industry (the four-digit industry SIC classification; dummy-coded), and functional TMT member title (e.g., COO).

4. Analyses

We used linear panel regression to analyze the data since the dependent variable in this study was a continuous

variable. Data were unbalanced as the number of observations varied across our independent, dependent, and

moderator variables. We tested our hypotheses using panel regression using the xtreg command in Stata 16.0, which

fits a regression model to panel data (StataCorp, 2019). The Hausman test (1978) indicated that a fixed effects

model was suited to our data. We included margin plots to depict significance of our effects.

We followed a hierarchical regression approach in which we introduced control variables first, followed by

independent variables and the moderator variable in subsequent models. We used the results of the full model,

Model 4, to test our hypotheses.

5. Results

Table 3 presents the correlations and descriptive statistics for the variables included in our analyses. We found that

men TMT members were compensated an average of USD $468,990 (SD = $275,110) more than women members

annually. This is consistent with previous findings (Bertrand and Hallock, 2001, Kalogeraki and Georgakakis, 2021,

Kulich et al., 2011, Shin, 2012).

Table 3. Descriptive Statistics.

TMT gender

pay gap

468.99 275.11 1.00

ROA 0.12 0.12 0.04*** 1.00

Log of

revenues

3.10 0.69 0.29*** 0.25*** 1.00

Net margin −0.22 9.40 0.00 0.18*** 0.16*** 1.00

Board

independence

8.00 2.68 0.11*** −0.04*** 0.39*** 0.03* 1.00

Board size 9.08 2.18 0.16*** −0.08*** 0.44*** 0.03*** 0.76*** 1.00

Nationality

mix on the

board

0.07 0.15 0.07′ −0.02′ 0.13*** 0.01 0.12*** 0.12*** 1.00

Variable M SD 1 2 3 4 5 6 7 8 9 10

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 7 of 25

***p < .001, ** p < .01, * p < .05, 'p < .1.

Table 4 reports the results of the hierarchical regression assessing the impact of the proportion of women board

members on the TMT gender pay gap. Model 1 includes only control variables. Model 2 adds our independent

variable, Model 3 adds our moderator, and Model 4 adds the interaction between our independent variable and

moderator.

Table 4. Hierarchical Regression Output with robust standard errors.

ROA 24.30 24.62 27.96 29.31

Log of revenues 149.14*** 151.03*** 147.86*** 145.38***

Net margin −0.55** −0.56** −0.55** −0.54**

Board independence −3.51 −3.46 −3.37 −3.56

Board size 0.39 −0.41 −0.13 −0.18

Proportion of non-US board members 26.48 26.18 28.30 27.49

Constant 134.28 130.71 147.92 149.75

Independent Variables

Proportion of women on the board 178.29* 119.18

Proportion of women on the board squared −674.80** −289.38

CEO gender −325.77*** −65.89

Proportion of

women on

the board

0.13 0.10 −0.03*** 0.04*** 0.32*** 0.03* 0.28*** 0.29*** 0.06*** 1.00

CEO gender 0.02 0.13 −0.26 0.01 −0.04*** 0.00 −0.04*** −0.04*** 0.00 0.16*** 1.00

Proportion of

women on

compensation

committee

0.13 0.22 0.00 0.00 0.12*** 0.01 0.09*** 0.11*** 0.03* 0.42*** −0.01 1.00

Number of

women board

members

1.20 1.06 0.01 0.02′ 0.39*** 0.03* 0.44*** 0.52*** 0.09*** 0.94*** 0.12*** 0.39***

Avg. time on

the board

4.93 6.27 0.03*** 0.09*** 0.02* 0.01 0.16*** −0.10*** −0.09*** −0.10*** −0.03*** −0.06***

Control Variables Model 1 Model 2 Model 3 Model 4

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 8 of 25

CEO gender × Proportion of women on the board −1469.56*

CEO gender × Proportion of women on the board squared 1695.37

***p < .001, ** p < .01, * p < .05, 'p < .1.

Hypothesis 1 predicted an inverted U-shaped relationship between the proportion of women on the board and the

TMT gender pay gap. Our results confirm that the squared term for proportion of women on the board was related

to the gender pay gap in the expected direction (b = -674.80, p < .01), thereby providing support for Hypothesis 1.

We conducted a U test analysis to estimate the threshold level, which was 14.6 %. This suggests that as the

proportion of women on a board increases from zero up until this threshold, the TMT gender pay gap (favoring

men) widens. However, when the proportion of women on the board surpasses this threshold, the TMT gender pay

gap narrows. Using this threshold and given an average board size of 9.03 members, adding a token woman to a

board with no women previously is likely to result in a widening of the TMT gender pay gap. However, adding two

or more women to a board with no women previously will likely narrow the TMT gender pay gap. This effect is

illustrated in Fig. 1.

Download: Download high-res image (118KB)

Download: Download full-size image

Fig. 1. Main effect of proportion of women on the board on the TMT gender pay gap FGR = Proportion of women on

the board.

Hypothesis 2 predicted that CEO gender moderates the curvilinear relationship between proportion of women on

the board and the TMT gender pay gap. As seen in Table 4, CEO gender was not a significant moderator of the

curvilinear effect. As such, we did not find support for Hypothesis 2. Notably, the interaction between CEO gender

and the linear relationship between the proportion of women on the board and the TMT gender pay gap was

significant (b = -1469.56, p < .05), which simply suggests that, overall, the presence of a woman CEO strengthens the

narrowing effect on the TMT gender pay gap that an increasing proportion of women on the board exerts.

5.1. Addressing endogeneity

Endogeneity is a significant concern in strategic management research (Wolfolds & Siegel, 2019). We took two

approaches to address potential endogeneity for our analysis. First, we followed Bednar et al. (2013) to include a

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 9 of 25

broad set of controls in our models. To rule out concerns of endogeneity, we also used a two-stage least squares

method (2SLS) with instrumental variable analysis techniques (Berry & Kaul, 2016). We separately checked the

degree of correlation between the variable being estimated, and the dependent and independent variables. We

used one instrument to conduct the first stage of our 2SLS model, which was the average time on the board of

board members (Guldiken et al., 2019), as this could reasonably be associated with board gender composition.

Indeed, boards with longer average tenure might be more likely to increase the proportion of women on their

boards. Moreover, the average time on the board is exogenous to the error term. This means it is not correlated with

other factors that might influence the gender composition of a board. For example, average time on the board was

not correlated with net margin.

We accessed information about directors’ tenure on their boards from BoardEx. As shown in Table 5, the average

time on the board was not a significant predictor of the TMT gender pay gap, though it did predict the proportion of

women on the board. Table 6 presents the results of our retested models when including the average time on the

board as a control. Our results did not meaningfully change when the average time on the board was included as a

control variable in our main analyses.

Table 5. Test of Instruments.

0.11*** −0.33

Control Variables

−0.03** 27.55

0.09*** 148.63***

−0.0003*** −0.55**

0.007*** −3.16

−0.004*** −0.36

0.001 27.99

0.11*** −325.78***

−0.15*** 149.63**

***p < .001, ** p < .01, * p < .05, 'p < .1.

Table 6. Instrumental Variable Models.

ROA 24.29 27.95 23.91 27.55 24.62 28.93

Log of revenues 149.14*** 147.86*** 149.86*** 148.63*** 151.03*** 146.15***

Proportion of women on the board Gender pay gap

Avg. time on the board

ROA

Log of revenues

Net margin

Board independence

Board size

Proportion of non-US board members

CEO gender

Constant

Control Variables Model 5 Model 6 Model 7 Model 8 Model 9 Model 10

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 10 of 25

Net margin −0.55** −0.55** −0.55** −0.55** −0.56** −0.54**

Board independence −3.51 −3.36 −3.32 −3.16 −3.45 −3.37

Board size 0.38 −0.12 0.17 −0.36 −0.41 −0.39

Proportion of non-US board members 26.47 28.30 26.17 27.99 26.18 27.22

Constant 134.28** 147.91** 135.83** 149.63** 130.71** 151.03**

Independent Variables

Avg. time on the board −0.30 −0.33 −0.28

Proportion of women on the board 178.29* 118.22

Proportion of women on the board squared −674.80** −290.09

CEO gender −325.77*** −325.78*** −66.34

CEO gender × Proportion of women on the board −1464.38*

CEO gender × Proportion of women on the board squared 1686.37

***p < .001, ** p < .01, * p < .05, 'p < .1.

5.2. Supplementary analyses: proportion of women on the compensation committee and number of women on the board

Table 7 shows the results of a supplementary analysis in which the proportion of women on the compensation

committee was used as our independent variable. Our reasoning for providing this analysis was that the women on

a board’s compensation committee may have more direct influence over decisions related to executive

compensation than women on the board more generally. Some cases were dropped because many of the TMT

members in our sample were also members of the compensation committee.

Table 7. Proportion of women on the compensation committee as a predictor of the TMT gender pay gap.

ROA 24.30 84.01 27.96 83.71

Log of revenues 149.14*** 110.96*** 147.86*** 113.52***

Net margin −0.55** −0.26 −0.55** −0.28

Board independence −3.51 −3.99 −3.37 −3.93

Board size 0.39 −0.86 −0.13 −0.93

Proportion of non-US board members 26.48 44.93 28.30 40.84

Constant 134.28 279.03 147.92 276.71

Control Variables Model 1 Model 2 Model 3 Model 4

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 11 of 25

Independent Variables

Proportion of women on the compensation committee −28.23 −12.03

Proportion of women on the compensation committee squared −1.91 −27.58

CEO gender −325.77*** −278.67***

CEO gender × Proportion of women on the compensation committee −1358.60***

CEO gender × Proportion of women on the compensation committee squared 1587.52***

***p < .001, ** p < .01, * p < .05, 'p < .1.

Our results showed that the proportion of women on the compensation committee was not a significant predictor

of the TMT gender pay gap. However, CEO gender was a significant moderator for both the linear (b = -1358.60,

p < .001) and curvilinear (b = 1587.52, p < .001) relationships between the proportion of women on the

compensation committee and the TMT gender pay gap. In terms of moderating the linear effect, these results

suggest that the presence of a woman CEO may strengthen the narrowing effect on the TMT gender pay gap that an

increasing proportion of women on the compensation committee exerts. Regarding the curvilinear effect, our

results suggest that the presence of a woman CEO will narrow the TMT gender pay gap when there is sub-critical

mass of women on the compensation committee. However, as the proportion of women on the compensation

committee increases, the presence of a woman CEO widens the pay gap, but this time favoring women over men

TMT members.

Table 8 displays the results of another supplementary analysis in which the absolute number of women on a board

was tested as a predictor of the TMT gender pay gap. Although it was not a significant predictor, it did interact

significantly with CEO gender to affect the TMT gender pay gap (b = −68.06, p < .01). The nature of this interaction

suggested that the presence of a woman CEO strengthens the narrowing effect on the TMT gender pay gap that

having a higher number of women on the board exerts.

Table 8. Number of women board members as a predictor of the TMT gender pay gap.

ROA 24.30 22.89 27.96 27.72

Log of revenues 149.14*** 154.27*** 147.86*** 147.87***

Net margin −0.55** −0.57** −0.55** −0.55**

Board independence −3.51 −3.16 −3.37 −3.49

Board size 0.39 1.06 −0.13 0.07

Proportion of non-US board members 26.48 26.48 28.30 27.81

Constant 134.28 117.83 147.92 144.89

Control Variables Model 1 Model 2 Model 3 Model 4

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 12 of 25

Independent Variables

Number of women board members −6.69 1.47

CEO gender −325.77*** −132.71

CEO gender × Number of women board members −68.06**

***p < .001, ** p < .01, * p < .05, 'p < .1.

Overall, the results of the supplementary analyses were somewhat consistent with the findings in our primary

analysis showing that the presence of a woman CEO strengthens the narrowing effect on the TMT gender pay gap of

having a larger number of women on the board.

6. Discussion

In the current work, we examined the relationship between board gender composition and the gender pay gap in

top management teams (TMTs), as well as the extent to which this relationship is moderated by CEO gender. Based

on a sample of 2658 U.S. firms over 14 years, we found an inverted U-shaped relationship between the proportion

of women on the board and the TMT gender pay gap. Specifically, we found that increasing the proportion of

women on the board from zero to about 15 percent widens the TMT gender pay gap favoring men. As the

proportion of women on the board increases above 15 percent, however, the TMT gender pay gap narrows and

continues to do so. Moreover, the presence of a woman CEO appears to strengthen the effect that a higher

proportion of women on the board has on the TMT gender pay gap. Our findings are consistent with the insights

that we drew and integrated from queen bee, critical mass, and upper echelons theories.

Important contributions of this work to both research and practice include challenging the popular narrative that

increasing the number of women in key leadership roles will improve the metrics of gender equity. First, our study

shows that initial increases in the proportion of women from zero could actually exacerbate gender inequities, at

least until the proportion of women meets and surpasses critical mass. As such, our findings suggest that recruiting

women board members to become tokens could do more harm than good (Ingersoll et al., 2023, Schwab et al.,

2016), at least in terms of the TMT gender pay gap. This finding highlights the tension between social identity

theory, which predicts support and cooperation (Balliet et al., 2014, Ellemers et al., 1997, Pickett & Brewer, 2001,

Spears, 2021, Tajfel and Turner, 1985), and the in-group threat and distancing predicted by the queen bee theory

(Duguid, 2011, Duguid et al., 2012, Duffy et al., 2012, Garcia et al., 2013, Reh et al., 2018).

Second, our research also enhances our overall understanding of how the gender composition of boards interacts

with CEO gender to impact metrics of gender equity on TMTs. Our moderating results suggest that the presence of a

woman CEO may strengthen the overall effect that increasing the proportion of women on boards will have on the

TMT gender pay gap. As such, a woman CEO in conjunction with more women on the board may speed up the

process by which organizational inequities in top management are addressed. This is consistent with the notion

that as women ascend to key leadership roles in greater numbers, their collective capacity to bring about positive

outcomes in organizational equity will be enhanced. It also speaks to the critical positional power of the CEO and

suggests that in the presence of men CEOs, increases in the proportion of women on boards may be less effective at

reducing gender disparities found in top executive pay. As such, our results suggest that placing a critical mass of

women in leadership roles that exert influence within the organization will be most effective at enhancing their

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 13 of 25

capacity to effect meaningful change.

Third, we contribute to the current dearth of research exploring factors that influence the gender pay gap among

TMT members specifically. This group is ultimately responsible for guiding the strategy of an organization

alongside the CEO and is therefore of critical importance to organizational performance and survival. Given the

significant influence that TMT members can have over the direction and future of the organization, developing an

enhanced understanding of gender inequities in TMTs is of particular importance. Indeed, organizations with TMT

gender inequities may struggle to attract, motivate, and retain their most talented performers, thereby failing to

realize the benefits of enhanced diversity among leaders and ultimately seeing organizational performance decline.

Therefore, understanding the factors that contribute to TMT inequities is a significant challenge that this work helps

to address.

Finally, for theory, practice, and policymaking, the implications of our findings go beyond the relationship between

the gender composition of BODs and the TMT gender pay gap. Specifically, our findings show the importance of

considering the interaction between various important entities operating alongside and in conjunction with one

another within the upper echelon of organizations (e.g., CEO, BOD, TMT), rather than considering these forces in

isolation, as much of the previous research has done. In the current work, we show that BOD gender composition

does not have a straightforward effect on the TMT gender pay gap, but rather interacts with CEO gender to

influence executive pay. More research that takes the same approach is needed to better understand the full scope

and nuances of the influence that various groups and individuals in the upper echelon of organizations exert over

organizational strategy, as well as firm- and individual-level outcomes.

7. Limitations and future research

The present study has a number of strengths, including a large data set spanning 14 years and multiple archival

sources, and the inclusion of numerous control variables, which offers more assurance of the robustness of our

results. However, the current work is not without limitations, some of which may be addressed by future research.

We offer suggestions in this regard.

First, our reliance on archival firm-level data means we could not directly test the mechanisms foundational to our

hypotheses and the interpretation of our results. For example, we could not assess the thought processes and

behaviors of TMT members and confirm that feelings of competitive threat and in-group distancing were exerting

influence over the results that we found, for example (Duguid, 2011, Ely, 1995). Researchers may wish to directly

test our theoretical assumptions through observational studies, surveys, interviews, and even experimental work.

Indeed, research on decision-making processes among board members may significantly advance our

understanding of the dynamics by which organizational inequities emerge, are maintained, and are resolved. Such

research would respond to calls for investigation of the antecedents and mechanisms that underlie the collective

behavior of individuals in strategic decision-making positions, particularly in terms of how decisions affect equity

(e.g., pay, upward mobility and other growth opportunities) (Carnahan and Greenwood, 2018, Chin and Semadeni,

2017, Cook et al., 2019, Corwin et al., 2021, Elsaid and Ursel, 2018, Kalogeraki and Georgakakis, 2021, Ryan et al.,

2016).

One strength of the current work is the exploration of the interactive effects of CEO gender and board composition.

More research is needed to provide a better understanding of the dynamics by which CEOs, boards, and top

management teams work in conjunction to resolve organizational inequity. While CEOs maintain significant

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 14 of 25

positional power, they may find that their capacity to enact change is hindered or empowered depending on who

serves on their boards and the rest of the top management team. Likewise, board members who share a vision of

resolving organizational inequities can only do so much in the presence of a CEO who does not share their

perspectives or motivations. Future research should explore these internal dynamics in order to understand how

organizational inequities are elicited, maintained, and resolved by those in leadership roles.

Future research may also go beyond the scope of the current work to investigate the downstream effects of gender

inequities on top management teams, including the retention of talent, especially in terms of women leaders.

Furthermore, it would be prudent to explore how disparities along gender lines shape the nature of gender

inequities in the broader organization. Given the critical role of top management in shaping organizational

performance and survival, understanding the consequences of gender inequities, as well as those tied to race and

other demographic categories, will be essential for organizational scholars to translate their research into

meaningful practice.

8. Conclusion

To our knowledge, the current research is the first large-sample study that explores the curvilinear relationship

between board gender composition and the TMT gender pay gap, while also considering the moderating effect of

CEO gender. The current study uncovers an inverted U-shaped relationship between the proportion of women on

the board and the TMT gender pay gap, such that as the proportion of women on the board increases from zero and

up to 15 percent, the TMT gender pay gap favoring men widens. However, when the proportion of women on the

board increases beyond 15 percent, the TMT gender pay gap narrows. Furthermore, we found that CEO gender

moderates the linear relationship between the proportion of women on the board and the TMT gender pay gap,

such that the presence of a woman CEO strengthens the overall narrowing effect that a higher proportion of women

board members has on the TMT gender pay gap. The present study highlights the need for more research into

women's representation on boards and understanding the dynamic between boards and CEOs as it applies to

resolving organizational inequities. We hope these findings will lead to a more comprehensive understanding of

the importance of women’s representation in top leadership roles.

CRediT authorship contribution statement

Gurdeep Singh Raina: Writing – review & editing, Writing – original draft, Project administration, Methodology,

Investigation, Data curation, Conceptualization. Arvin Sahaym: Writing – review & editing, Writing – original draft,

Supervision, Project administration, Conceptualization. Leah D. Sheppard: Writing – review & editing, Writing –

original draft.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could

have appeared to influence the work reported in this paper.

Acknowledgement

The authors would like to thank Dr. Benjamin Warnick, Dr. Alexander Kier, Dr. Thomas Rotolo, Dr. Chandresh Baid,

https://www-sciencedirect-com.libproxy.aum.edu/science/article/pii/S0148296324002145 10/26/25, 8:59 AM Page 15 of 25

Balliet et al., 2014

Bednar et al., 2013

Berry and Kaul, 2016

Bertrand and Hallock, 2001

Bertrand et al., 2010

Benson-Armer et al., 2015

Blau and Kahn, 2003

Blevins et al., 2019

Dr. Smita Srivastava, Dr. Anup. M. Nandialath, Ms. Soni Jha and Ms. Deepti Kalra for their valuable feedback and

suggestions. The authors would also like to thank the seminar (brown bag) participants at the Carson College of

Business, colleagues, the 2020 SMS Annual Conference and the 2022, 2023 AOM Annual Conference attendees for

their valuable feedback and suggestions.