Article Journal of Accounting,

Auditing & Finance 2016, Vol. 31(4) 415–427

�The Author(s) 2016 Reprints and permissions:

Gender Diversity in Compensation Committees

sagepub.com/journalsPermissions.nav DOI: 10.1177/0148558X16655704

jaf.sagepub.com

Sascha Strobl1, Dasaratha V. Rama2, and Suchismita Mishra2

Abstract

Gender diversity in corporate boards and the functioning of compensation committees are two issues that have caught the attention of legislators and regulators in many countries. Using data from 5,630 observations from public companies in the United States, we find that firm and board characteristics are associated with the presence of females on the com- pensation committee. However, female presence on the compensation committee is not significantly associated with CEO pay (as a proportion of shareholder wealth increase). We then discuss issues related to research on gender diversity in corporate boards, and offer some suggestions for future research on diversity in corporate boards.

Keywords

compensation committee, gender, diversity, executive compensation, board committees

Introduction

In this article, we examine the gender diversity of compensation committees, and whether

such diversity is associated with the relative magnitude of CEO compensation. We then

derive an agenda for future research about diversity in corporate boards, with a particular

focus on compensation and audit committees. We are motivated by recent efforts of legisla-

tors in many countries seeking to increase the proportion of women in corporate boards,

through persuasion and/or legislative quotas.

The non-profit group Catalyst, which seeks to promote women in business and leader-

ship, notes that less than 20% of board seats in the S&P 500 companies are held by women

(Catalyst, 2015); the proportion is even lower among smaller firms, and the lack of female

representation in corporate boardrooms is a global phenomenon (Deloitte, 2013). This has

led to legislative gender quotas on corporate boards in some countries. Norway led the way

in 2003 with a requirement that at least 40% of the directors of public companies must be

female. Subsequently, a diverse array of countries such as France, Spain, Italy, Malaysia,

India, and the UAE have implemented gender quotas for boards of public companies

(Deloitte, 2013; ‘‘The Spread of Gender,’’ 2014). Similar legislation has been proposed

1University of Vaasa, Finland 2Florida International University, Miami, FL, USA

Corresponding Author: Dasaratha V. Rama, School of Accounting, Florida International University, Miami, FL 33199, USA. Email: [email protected]

Downloaded from jaf.sagepub.com by guest on September 19, 2016

416 Journal of Accounting, Auditing & Finance

elsewhere, including Brazil, Canada, Germany, Israel, and the Philippines, while govern-

ments in Australia, Sweden, and the United Kingdom have encouraged companies to volun-

tarily increase female representation on corporate boards or risk having quotas imposed at

a later date (Ernst & Young, 2014; ‘‘The Spread of Gender,’’ 2014). In the United States,

the Securities and Exchange Commission (SEC; 2009) has issued a rule requiring compa-

nies to disclose the role of diversity in considering candidates for director nominations.

Evidence related to any relationship between board diversity and performance has both

public policy and governance consequences. If there is consistent evidence that there is a

positive association between board diversity and firm performance, then there is a business

case for diversity. In the absence of such evidence, arguments for diversity must be made

on other grounds. And, if diversity leads to lower performance then the costs of diversity

must be considered in any discussion about making boards more diverse.

In theory, gender diversity on boards can have positive implications for firms. Diversity in

the boardroom can lead to consideration of alternative viewpoints (Zahra & Pearce, 1989); board

diversity also increases the external legitimacy, and can make the firm more attractive to talented

employees (Hambrick, Werder, & Zajac, 2008). Conversely, diversity can also increase conflict,

worsen communication, and reduce trust (‘‘The Downside of Diversity,’’ 2014). Empirical evi-

dence about the performance effects of gender diversity on boards is mixed. Adams, de Haan,

Terjesen, and vanEes (2015, p.78) note that the mixed findings related to diversity and firm per-

formance can be attributed to ‘‘differences across studies in measures of performance, methodol-

ogies, time horizons, omitted variable biases and other contextual issues.’’

In evaluations of board performance, it may be more appropriate to focus on the compo-

sition of the sub-committees of the board because most board decisions are made within

smaller groups or committees (Kesner, 1988). In the context of accounting, two board com-

mittees are particularly relevant: audit and compensation committees. The former is an

integral part of the financial reporting process, while the latter often uses accounting num-

bers in setting targets related to executive compensation contracts (and, thus, influences

managerial judgments related to accounting). Both of these committees have a crucial role

in the corporate governance process.

Some studies have examined the association between gender diversity and the functioning

of audit committees.1 Srinidhi, Gul, and Tsui (2011), using data from 2001 to 2007 and after

controlling for self-selection, find that firms with female directors on the audit committee

exhibit higher earnings quality (as measured by lower discretionary accruals and lower pro-

pensity to manage earnings and beat benchmarks by a small amount). Similarly, Thiruvadi

and Huang (2011) show, using data from 320 S&P SmallCap 600 firms, that the presence of

a female director on the audit committee is associated with lower discretionary accruals. In

contrast, Sun, Liu, and Lan (2011) find, using a sample of 525 observations over the period

2003 to 2005, that there is no association between the proportion of female directors on audit

committees and performance-matched discretionary accruals. In terms of audit committee

processes, Thiruvadi (2012) shows that audit committees with at least one female director are

likely to meet more often than all-male audit committees while Ittonen, Miettinen, and

Vähämaa (2010) find that firms with female audit committee chairs have lower fees.

However, empirical research about diversity in compensation committees is sparse. Two

prior studies find that female directors are less likely to be appointed to the compensation

committee (Adams & Ferreira, 2009; Bilimoria & Piderit, 1994). We extend this line of

research by examining (i) the factors associated with a gender-diverse compensation commit-

tee and (ii) the association between the presence of female directors on the compensation

committee and (a) executive pay, and (b) subsequent restatement of financial statements.

Downloaded from jaf.sagepub.com by guest on September 19, 2016

Strobl et al. 417

Compensation Committees and Gender Diversity

The performance of compensation committees of public companies has come under

increased scrutiny from the public and regulators in recent years. Some common criticisms

of compensation committees are that top executives are overpaid and that such compensation

often has little association with performance (Landy, 2008; Morgenson, 2013).2 A significant

part of the blame for the global financial crisis is attributed by some to the outsized compen-

sation and skewed incentive structures of corporate executives (Bebchuk, 2012; Ritholtz,

2009). Such concerns have also been reflected in recent actions by legislators and regulators.

Sections 951 to 954 of Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010 includes multiple provisions related to executive compensation of public companies,

including the independence of compensation committees and their use of compensation con-

sultants, as well as more detailed disclosures about pay-for-performance and the ratio of

CEO pay relative to other employees. Subsequently, the SEC (2012) promulgated new rules

related to the stock exchange listing standards for compensation committees.

Much of the blame for outsized executive compensation is attributed to docile and ineffec-

tive compensation committees. In theory, a strong and independent nominations committee

should select effective and independent directors, a subset of whom would become members

of the compensation committee; this should ensure that the executives are adequately, but not

excessively, compensated. In practice, however, the CEO is dominant in many public compa-

nies and has a strong influence on who is selected for the board and also who is appointed

on the compensation committee (Bear, Rahman, & Post, 2010). Compensation committees

use compensation consultants to set the pay of the executive; however, a former Chancellor

of the Exchequer noted in testimony to the U.K. Parliament that ‘‘remuneration consultants

. . . are a profession that makes prostitution seem thoroughly respectable’’ (Lawson, 2013).

Westphal and Zajac (1995) find that CEOs are likely to select directors who are demo-

graphically similar to them, which in turn can lead to a board that is more likely to support

the CEO (including, executive compensation related matters). In contrast, greater diversity

on the board can lead to more effective monitoring (Bear et al., 2010).

Prior studies show that gender diversity on the board is associated with higher levels of

monitoring by the directors. Nielsen and Huse (2010) find that higher proportions of

female directors are positively related to board strategic control. Adams and Ferreira

(2009) show that gender-diverse boards are associated with better attendance records, and

that CEO turnover is more sensitive to stock performance in such firms.3 Thus, based on

prior empirical evidence that gender-diverse boards are associated with greater monitoring,

we conjecture that female directors will tend to emphasize more on pay-for-performance;

hence, gender diversity on the compensation committee makes it less likely that there will

be excessive executive pay.4 This leads to our hypothesis (in the alternative form):

Hypothesis: Companies that have gender-diverse compensation committees will be

less likely to have excessive executive pay.

Method and Data

Our primary data source is The Corporate Library’s directorship database for the years

2006 to 2008. We obtain our financial and stock price data from the Compustat and CRSP

databases. To minimize the influence of outliers, we winsorize all continuous variables at

the first and 99th percentiles.

Downloaded from jaf.sagepub.com by guest on September 19, 2016

�

418 Journal of Accounting, Auditing & Finance

We use the following model to examine the determinants of factors associated with the

presence of a female on the compensation committee:

FEM = a + b13LMV + b23PROUT + b33BDNUM + b43CCNUM ð1Þ

+ b53LTPERF + b63INSTOWN + b73NYSE + b83OM + e:

The dependent variable, FEM, is measured in two different ways. First, we use a dichot-

omous variable, FEMD, which takes the value of 1 if there is at least one female on the

compensation committee and 0 otherwise; we use a logistic regression model for this analy-

sis. Second, we use a continuous measure, FEMP, which is the proportion of females on

the compensation committee; we use an ordinary least squares (OLS) regression for this

analysis. In our analyses, we control for clustering by firms.

Our independent variables are based on results from prior studies. As larger firms are

more likely to have a diverse board (Carter, D’Souza, Simpkins, & Simpson, 2003;

Hillman, Shropshire, & Cannella, 2007), we expect the coefficient on LMV (log of market

value) to be positive. Carter et al. (2003) find that the proportion of women and minorities

on boards increases as the number of outsiders on the board increases; extending this logic,

we expect a positive coefficient for PROUT (proportion of outside directors on the compen-

sation committee). Higher the number of directors on the board, the greater the likelihood

that there will be females on the board; this in turn increases the likelihood of female repre-

sentation on the compensation committee. Similarly, the greater the number of directors on

the compensation committee, the higher is the likelihood of a female director being present

on the committee. Hence, we include BDNUM and CCNUM in our model.

Coffey and Fryxell (1991) argue that the influence of institutional investors may make it

more likely that companies will adopt diversity practices; hence, we expect a positive coef-

ficient on INSTOWN. Pressure from governance activists is more likely when a firm has

not been performing well in the recent past. Such arguments suggest that female directors

on the compensation committee would be less likely in firms with better long-term perfor-

mance (LTPERF). We include an indicator variable for NYSE listing because there are sig-

nificant differences between NYSE- and non-NYSE-listed firms along many governance

dimensions (Wintoki, 2007). Finally, we expect that women are more likely to be appointed

to the board and committees of the board in firms that are incorporated in more liberal

states. We measure this using the proportion of votes for Barack Obama minus the propor-

tion of votes for John McCain during the U.S. presidential election of 2008 (OM).

Next, we use the following model to examine the relation between gender-diverse com-

pensation committees and executive compensation:

PAYPER = a + b13LMV + b23PROUT + b33BDNUM + b43CCNUM

+ b53INSTOWN + b63NYSE + b73OM + b83AGE ð2Þ

+ b93TENURE + b103Book Market + b113Volatility

+ b123FEMCC + b133Industry Dummies + e:

The dependent variable, PAYPER, is the ratio of annual change in market value of the

company over the total compensation of the CEO. This ratio measures the ‘‘bang for the

buck’’ for shareholders from a given dollar of CEO compensation; higher values are

‘‘better’’ from the perspective of shareholders (who are represented by the compensation

committee). As in most corporate governance research, we control for firm size. We also

Downloaded from jaf.sagepub.com by guest on September 19, 2016

Strobl et al. 419

control for compensation committee composition (PROUT) and board size (BDNUM), as

these two factors can drive the extent to which the compensation committee and the board

are proactive and look out for shareholder interests. We control for institutional ownership

(INSTOWN) as institutions are more likely to exert pressure on the board. As in the earlier

regression for compensation committee membership, we include controls for the stock

exchange (NYSE) and the political climate of the state of location of the company (OM).

AGE and TENURE measure the average age and tenure of the compensation committee

directors. We also include Book-to-Market ratio and return volatility as additional controls,

apart from industry dummies based on the Fama-French classification. FEMCC is our vari-

able of interest, and we use two different measures for this variable: FEMD and FEMP,

which are defined as before.

Results

Table 1 provides descriptive evidence about the sample. Overall, 34% of the observations

have at least one female director on the compensation committee. In terms of individual

directors, just over 10% of the compensation committee directors are female. The average

number of directors on the full board and compensation committee are 15 and 3.76, respec-

tively. Outside directors account, on average, for 93% of compensation committees. The

average age of compensation committee directors is 61.31, and the average tenure of such

directors is 8.47 years. 5

In untabulated results, we find that women account for about 11% of the total number of

directors and the overall number of compensation committee directors for the firms in our

sample.6 Thus, the dearth of female directors, documented in many prior studies, continues

to be the case in our sample. However, the data also indicate that given a board member-

ship, a female is at least as likely to be on the compensation committees as her male coun-

terparts. In addition, we also find that, compared with their male colleagues, female

directors on compensation committees in our sample are younger (60.1 vs. 61.6 years old),

have shorter tenure as directors (8.0 vs. 8.5 years), and are more likely to be outsiders

(94.6% vs. 93.4%). These findings reinforce results from prior research that female direc-

tors are different from their male counterparts in many other demographic characteristics.

Table 2 presents the results from the regressions to examine factors associated with the

presence of female directors on the compensation committee. The first (logistic) regression

has FEMD as the dependent variable, while the second (OLS) regression has FEMP as the

dependent variable. All of our control variables, except institutional ownership, are signifi-

cant at conventional levels in both regressions. The results indicate that females are more

likely to be present on the compensation committee of firms when the firm (a) is larger,

(b) has a higher proportion of outsiders on the board, (c) has more directors on the board, (d)

has more members on the compensation committee, (e) has worse long-term performance, (f)

is listed in the NYSE, and (h) is incorporated in a state that is politically more liberal.

In Table 3, we present the results related to our hypothesis. Both of the regressions in

Table 3 are statistically significant. Considering the governance-related variables, we find

that (a) the coefficient of BDNUM is negative and significant, indicating that larger boards

are less likely to be ‘‘efficient’’ in terms of protecting shareholder interests; and (b) the

coefficient of TENURE is negative and significant, indicating that compensation commit-

tees that have long director tenure are more likely to be associated with higher executive

compensation.7 However, the gender variable is not significant in either regression.

Downloaded from jaf.sagepub.com by guest on September 19, 2016

420 Journal of Accounting, Auditing & Finance

Table 1. Descriptive Statistics. This table presents the mean; standard deviation; first, second and third quartile of our variables. PAYPER is the ratio of the change in market equity to total compensation of the CEO for each year in our sample (times 1,000). FEMD is the presence of at least one female on the compensation committee; FEMPR is the proportion of female directors on the compensation committee; LMV is the natural log of the market value of the equity (in millions of dollars); PROUT is the proportion of outside members on the board; BDNUM is the number of directors on the board; CCNUM is the number of directors on the compensation committee; LTPERF is the average stock return of the past 5 years; INSTOWN is the proportion of shares outstanding owned by institutions; NYSE is 1 if a stock is listed at the NYSE and 0 otherwise; OM is the proportion of votes for President Obama minus the proportion of votes for Senator John McCain in 2008 in the state in which the company is headquartered; AGE and TENURE are the average values of the age and tenure of the compensation committee directors; BOOK-MARKET and VOLATILITY are the Book to Market Ratio and idiosyncratic return volatility for each firm in our sample, respectively. The sample includes 5,630 firm- years from 2006-2008 of firms included in The Corporate Library’s directorship database.

Variable M SD 25th percentile Median 75th percentile

PAYPER 0.73 8.96 21.21 0.08 2.00 FEMD 0.34 — — — — FEMPR 0.10 0.16 0.00 0.00 0.25 LMV 7.43 1.48 6.30 7.25 8.35 PROUT 0.93 0.15 1.00 1.00 1.00 BDNUM 15.00 4.04 12.00 14.00 17.00 CCNUM 3.76 1.28 3.00 3.00 4.00 LTPERF 0.20 0.15 0.09 0.17 0.27 INSTOWN 0.52 0.31 0.29 0.48 0.70 NYSE 0.56 — — — — OM 0.03 0.20 20.15 0.06 0.24 AGE 61.31 5.25 58.25 61.50 64.50 TENURE 8.47 4.29 5.33 7.67 10.67 BOOK-MARKET 0.43 0.31 0.20 0.39 0.60 VOLATILITY 0.01 0.01 0.00 0.01 0.02

Further Tests

As an additional test, we use the CEO efficiency ranking from Forbes magazine in 2009

and use the reverse of this ranking (i.e., higher values represent a better compensation com-

mittee) as the dependent variable. When we combine the Forbes data with our dataset, we

are able to obtain all relevant data for 172 firms. In this analysis, we use the average values

(over the 3 years) of the independent variables in regression model (Equation 2) above.

Once again, we find that the gender variables are not significant in the model.

While the audit committee is primarily responsible for the quality of financial reporting,

the actions of the compensation committee can have an indirect effect on financial report-

ing quality. If a compensation committee is effective, then the magnitude of the possible

executive compensation would not be excessive and executives should be less likely to

manipulate financial statements (to achieve compensation related goals) which then have to

be restated.8 Hence, we examine subsequent restatements of financial statements. We find

that the subsequent restatement rates are 12.4% and 12.9% for the observations with and

without a female director on the compensation committee; the difference is not statistically

significant.

Downloaded from jaf.sagepub.com by guest on September 19, 2016

Strobl et al. 421

Table 2. Presence of Female Directors on Compensation Committees

Our website has a team of professional writers who can help you write any of your homework. They will write your papers from scratch. We also have a team of editors just to make sure all papers are of HIGH QUALITY & PLAGIARISM FREE. To make an Order you only need to click Ask A Question and we will direct you to our Order Page at WriteDemy. Then fill Our Order Form with all your assignment instructions. Select your deadline and pay for your paper. You will get it few hours before your set deadline.