Is Gender Diversity Profitable? Evidence from a Global Survey

Is Gender Diversity Profitable? Evidence from a Global Survey

Marcus Noland, Tyler Moran, and Barbara Kotschwar

Abstract

Analysis of a global survey of 21,980 firms from 91 countries suggests that the presence of women in corporate leadership positions may improve firm performance. This correlation could reflect either the payoff to nondiscrimination or the fact that women increase a firm’s skill diversity. Women’s presence in corporate leadership is positively correlated with firm characteristics such as size as well as national characteristics such as girls’ math scores, the absence of discriminatory attitudes toward female executives, and the availability of paternal leave. The results find no impact of board gender quotas on firm performance, but they suggest that the payoffs of policies that facilitate women rising through the corporate ranks more broadly could be significant.

JEL codes: G3, J16, M14
Keywords: Women, gender, diversity, boards of directors, CEOs

Marcus Noland is executive vice president and director of studies at the Peterson Institute for International Economics and a nonresident senior fellow at the East-West Center. Tyler Moran is a research analyst at the Peterson Institute. Barbara Kotschwar is adjunct professor of Latin American studies and economics at Georgetown University. She was research fellow at the Peterson Institute for International Economics.

Note: This paper was made possible through generous support from EY. The authors thank Caroline Freund, Adam Posen, Kevin Stahler, and Nicolas Veron for helpful comments on an earlier draft.

Copyright © 2016 by the Peterson Institute for International Economics. All rights reserved. No part of this working paper may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or by information storage or retrieval system, without permission from the Institute. This publication has been subjected to a prepublication peer review intended to ensure analytical quality. The views expressed are those of the authors. This publication is part of the overall program of the Peterson Institute for International Economics, as endorsed by its Board of Directors, but it does not necessarily reflect the views of individual members of the Board or of the Institute’s staff or management. The Peterson Institute for International Economics is a private nonpartisan, nonprofit institution for rigorous, intellectually open, and indepth study and discussion of international economic policy. Its purpose is to identify and analyze important issues to make globalization beneficial and sustainable for the people of the United States and the world, and then to develop and communicate practical new approaches for dealing with them. Its work is funded by a highly diverse group of philanthropic foundations, private corporations, and interested individuals, as well as income on its capital fund. About 35 percent of the Institute’s resources in its latest fiscal year were provided by contributors from outside the United States. A list of all financial supporters for the preceding four years is posted at http://www.piie.com/institute/supporters.pdf.

Women do not participate in the global economy to the same extent as men do. Implicit is a normative question of why this is the case and a positive question of what would be the economic impact if women were to participate more fully in economic life. The McKinsey Global Institute (2015) estimates that a scenario in which women achieved complete gender parity with men could increase global output by more than one-quarter relative to a business-as-usual scenario.

The role of women is particularly salient for countries with rapidly aging populations. Emblematic of the exigency of making fuller use of women in the economy has been the Japanese Diet’s passage of the Act Concerning the Promotion of Women’s Career Activities and the entry of the term womenomics into Japanese economic discourse. In South Korea President Park Geun-hye has pledged to boost women’s workforce participation by making public funds available to encourage companies to offer more flexible schedules and subsidize childcare.

The relative dearth of women in corporate leadership positions is an emerging political issue. A number of countries, including France, Finland, Iceland, Norway, and Spain, have gone so far as to mandate female representation on corporate boards, and other countries are considering following suit. Several Asian countries are adopting similar policy responses, setting workforce gender balance targets, quotas on corporate boards and political leadership roles, or expanding family-focused policies.

This paper addresses the relative absence of women on corporate executive boards and at the upper levels of management globally. It is based on a 2014 sample of 21,980 firms headquartered in 91 countries. Nearly 60 percent of these firms have no female board members, just over half have no female “C-suite” executives (a firm’s most senior executives and members of corporate boards), and less than 5 percent have a female chief executive officer (CEO).

The presence of women in corporate leadership is positively correlated with some firm characteristics, such as size, as well as some national characteristics, such as girls’ performance on math assessments, the relative absence of discriminatory attitudes toward female executives, and the availability of paternal leave, among others. The data reveal considerable variation in female representation across regions and countries as well as sectors of the economy.

Past evidence on the impact of female leadership on corporate performance, typically derived from research undertaken in a single country, has been mixed. Examining the economic performance of large US firms, Erhardt, Werbel, and Shrader (2003) and Carter et al. (2007) find that greater gender balance among corporate leaders is associated with higher stock values and greater profitability. Other research on US firms finds that mixed-gender boards outperform all-male boards (McKinsey 2012b) and that the Fortune 500 companies with the highest proportion of women on their boards performed significantly better than firms with the lowest proportion (Catalyst 2011). Accounting firm Rothstein Kass (2012) finds that hedge funds headed by women outperform hedge funds headed by men. More diverse boards have also been found to contribute positively to firm performance in Latin America (McKinsey 2013) and
Spain (Campbell and Mínguez-Vera 2008). Some studies find greater gender balance gains in particular sectors and circumstances.

However, even Carter et al. (2007), in their generally positive assessment of the impact of diversity on corporate performance, observe that the process through which diversity affects board performance is complex and that while some board functions may benefit from greater gender or racial diversity, others may not.2 Perhaps not surprisingly, some studies conclude that greater balance has a neutral or even negative impact. In a study of German companies, Lindstädt, Wolff, and Fehre (2011) find no overall relationship between female board membership and stock performance. In their study of 2,000 firms, O’Reilly and Main (2008) find no evidence that adding women to boards enhances corporate performance and conclude that such appointments are generally undertaken for normative rather than profit-seeking motives.

This paper uses a global dataset of nearly 22,000 firms to examine the impact of gender diversity on corporate performance. It then analyzes the correlates with diversity. Caution should be exercised in interpreting the statistical results, which are the product of a single snapshot. The dearth of cross-national evidence on these issues justifies taking this first cut.

The results suggest that the presence of women in corporate leadership positions may improve firm performance and that the magnitudes of the correlations are not small. The largest gains are for the proportion of female executives, followed by the proportion of female board members; the presence of female CEOs has no noticeable effect on firm performance. This pattern underscores the importance of creating a pipeline of female managers and not simply getting lone women to the top.

The positive correlation between the proportion of women in corporate leadership and firm profitability could reflect the existence of discrimination against women executives (which gives nondiscriminating firms an edge) or the fact that the presence of women contributes to skill diversity (to the benefit of the firm). There is no evidence that the female board quotas enacted by some countries have had an impact, for good or ill, though the statistical analysis may be too crude to detect such effects.

1. Dezso and Gaddis Ross (2011) find that adding women leaders improves performance in innovation-oriented firms. Lindstädt, Wolff, and Fehre (2011) find positive results from increasing female leadership in consumer-oriented companies. They also find that companies that have a large female workforce benefit from having female leaders. Jurkus, Park, and Woodard (2011) find that increased gender equity can be beneficial in firms with weak external governance. The Credit Suisse Research Institute (2012) finds that companies with women on their boards perform better than companies with all-male boards in challenging markets. Following the 2008 global economic crisis, for example, net income growth for companies with women on their boards averaged 14 percent, compared with 10 percent for companies with all-male boards.
2. One channel for these gains is a more supportive work environment brought about by greater gender balance. Dezso and Gaddis Ross (2011) find that female representation in top management brings informational and social diversity benefits, improves the performance of other managers, and helps motivate women in middle management. Adding women may also lead to better board behavior. Adams and Ferreira (2009) find that female directors have a positive impact on board inputs (attendance at meetings, participation on committees) and firm outcomes. They find that more gender-balanced firms devote more time to monitoring and tend to hold CEOs more accountable for poor stock performance.

GENDER BALANCE IN CORPORATE LEADERSHIP

We compiled a global dataset from Reuters profiles of publicly traded firms in late 2014. The Reuters data did not explicitly provide identifying information about a firm’s home country, top leadership, or the gender identity of corporate leaders (defined here as members of corporate boards and firms’ most senior executives). Appendix A describes the techniques used to obtain these data.

This dataset differs from other firm-level studies on gender balance among corporate leaders in its size and scope, spanning 21,980 firms headquartered in 91 countries.3 Its main shortcoming is that it is limited to a single year (2014). Constructing a panel dataset is a task for future research.

Low levels of female participation are evident on both corporate boards and in executive ranks. Nearly 60 percent of the sample (13,017 firms) recorded no female board members. Just over 50 percent (11,802 firms) have no female executives. Of the remaining half, 57 percent have only one female executive. Only 945 firms—less than 5 percent of the sample—have a female CEO. Appendix table A.1 reveals the gender distribution of the 21,954 firms surveyed (all firms less the 26 that could not be linked to a specific country).

The vertical axis of figure 1 sorts firms by the share of women on their boards; the horizontal axis shows the share of women on the executive committee.4 At firms directly above the horizontal axis, women hold 0–5 percent of board positions. At firms in the leftmost column, women hold 0–5 percent of executive positions. The intersection of this row and this column is a single cell (the cell at the bottom left), which covers firms with less than 5 percent female executives and less than 5 percent female board members. This cell is by far the densest, containing 7,859 firms, or roughly one-third of all observations.

The opposite case, in which all executives and all board members are women, counts 11 firms.5 The vast majority of firms (about 17,000) fill less than 30 percent of executive positions and less than 30 percent of board seats with women.

Only 11 percent of the nearly 130,000 corporate board members in the database and 659 of more than 17,000 board chairs (3.8 percent) are women. Women represent 14 percent of the 144,000 executives in the sample and 4.5 percent of the CEOs of the roughly 22,000 firms. If one accepts the

1. The 91st “country” comprises 26 firms that could not be linked to a particular country and were therefore not included in the analysis. Most of the Korean data could not be used, because of difficulties identifying the gender of corporate leaders (Korean names are notoriously gender-neutral, a well-documented challenge for research of this type; see Yoon et al. 2008).
2. The number of firms is about 600 short of the total because a firm must have had both a nonzero number of executives and a nonzero number of board members identified as male or female to be sorted. About 600 firms did not.
3. For all but the very largest firms, having 0–5 percent of executives and board members be women is equivalent to employing no such women. For a firm to have women make up less than 5 percent of executives while still employing a nonzero number of female executives, there would have to be 21 or more executives in total. Such firms represent far less than 1 percent of the total.

premise that equally educated women in a workforce are as capable as men of moving into management, this low number is a puzzle worth pursuing.

The broad averages obscure considerable cross-country and cross-sectoral variation. Figures 2 and 3, which illustrate women’s representation on corporate boards and C-level positions for all countries in which 10 or more firms in the dataset are headquartered, highlight the cross-country differences (see also appendix tables A.1 and A.2). As these maps make clear, although there is some relationship between gender balance and per capita income, the correlation is far from perfect. In East Asia, for example, where per capita income is relatively high, women hold only 6 percent of board positions and just 3 percent of board chairs. Intraregional variation is significant, however, with the share of female executives ranging from 2.5 percent in Japan to 13.5 percent in China.

A number of countries have implemented quotas for women on corporate boards (table 1). Norway is the best known, with a 40 percent gender quota for state-owned and, as of 2008, public limited companies. Denmark and Finland impose quotas on female representation for boards of majority stateowned enterprises.6 To date only Norway and Iceland have applied their complete quota instrument to publicly listed companies.7 The two countries have the highest female board representation, at 40 and 51 percent, respectively. France implemented a 20 percent quota in 2014, half of the 40 percent quota that will become binding in 2017.

Similar, if less dramatic, variation in outcomes is evident across sectors, with the financial, healthcare, utilities, and telecommunications sectors exhibiting the largest shares of female executive and board representation and basic materials, technology, energy, and industrials exhibiting the smallest (figure 4). Research by McKinsey (2012a) has shown that different sectors create different trajectories for women. In finance, men and women take entry-level positions in roughly equal numbers, but the number of women shrinks by about half by the middle-management level, leaving fewer female candidates to select for leadership positions. In contrast, in sectors that traditionally hire fewer women, such as transport, logistics, and energy, women who are hired have a better chance of promotion to middle management and beyond. It may also be the case that relatively low scores in certain natural resource–based sectors may reflect the location of those assets in countries predisposed toward low representation of women in leadership ranks.

1. The first country to set quotas was Norway, whose Gender Equality Act of 1981 stipulated a requirement of at least 40 percent of each gender on publicly appointed boards, councils, and committees. The law was extended to boards of publicly owned enterprises in 2004 and to larger joint stock companies in 2006. Quotas for state-owned enterprises have been in force in Denmark since 2000 and in Finland since 2005.
2. State-owned companies may be listed publicly, with the government as a major shareholder, but not all publicly listed companies have the government as a shareholder.

DOES GENDER BALANCE AFFECT FIRM PERFORMANCE?

There is no unambiguous theoretical presumption with respect to the impact of gender diversity on corporate performance. The standard argument is that as firms are value maximizing, if anything could be gained by adding women to corporate leadership positions, they would do so; attempts to force change (by mandating female representation on boards, for example) could be counterproductive (Demsetz and Lehn 1985). This negative outcome might be particularly likely if the new board members (or executives) were less experienced or of lower quality than men or if, because of the scarcity of qualified women, the small pool was stretched across too many boards (the “golden skirt” phenomenon). In the case of the Norwegian board quota, there is empirical research on this point, as discussed below.

The opposite view is that the relative scarcity of women in corporate leadership reflects discrimination and that firms that do not discriminate will exhibit superior performance, particularly if women possess unique or different skill mixes. Indeed, it has been argued (by Iriyama 2015, for example) that although skill diversity generally contributes to corporate performance, demographic or gender diversity per se does not. In their examination of US publicly traded firms, Kim and Starks (2015) find that functional diversity contributes to firm performance and that women bring specific functional expertise to boards of US firms, thereby enhancing performance.

Finally, the functioning of boards and senior management teams reflects complex small group dynamics. Depending on circumstances, introducing women into the mix could either boost or detract from leadership and, by extension, firm performance.

Given such ambiguity, it is perhaps not surprising that empirical evidence on these issues is mixed. Erhardt, Werbal, and Shrader (2003) find a positive relationship between the diversity of executive boards and returns on assets and on investments among 112 Fortune-listed US companies. Carter, Simkins, and Simpson (2003) find that Fortune-listed US firms with at least two women on their boards exhibit higher Tobin’s Q ratios (a measure of firm assets in relation to a firm’s market value) than firms with lower female representation.

In contrast, after examining nearly 2,000 firms appearing in various S&P indices, Adams and Ferreira (2009) find that gender-diverse boards allocate more effort to monitoring management performance and that as a consequence, CEO turnover is more sensitive to stock performance. Directors also receive more equity-based compensation in more gender-diverse boards. They conclude that gender diversity has a negative impact on firm performance on average (firms with more gender-diverse boards have fewer takeover defenses), though for a subset of firms with weak governance, diversity adds value. Smith, Smith, and Verner (2006) and Rose (2007) obtain mixed evidence on the impact of women on boards on firm performance in Denmark.

The data examined here both are more extensive and cover a vastly more diverse set of economies than the data used in previous studies. Tables 2 and 3 report multivariate regressions on female leadership and gross and net margins. (The gross margin is revenue less cost of goods as a share of revenue; the net margin is gross profit less overhead and other expenses as a share of revenue.) Because these figures are expressed as deviations from sectoral averages, they take positive and negative values dispersed around zero, and simple ordinary least square (OLS) estimation is adequate.

The two tables share a common format, with the included regressors consisting of the share of female board members; the share of female C-level executives; the share of both female C-level executives and female board members (all ranging from 0 to 1); the number of individuals on the board; the log of total revenue (as a scale variable); and the product of a dummy variable taking the value 1 if the firm is headquartered in a country with a quota for female board representation and the firm’s share of female board members.8 We also include country and industry dummies. In addition to the results reported, we estimated alternate regressions on a binary variable indicating whether a firm had a female CEO.

This analysis revealed that the CEO’s gender does not have a significant impact on firm profitability, when controlling for gender balance elsewhere in the firm, and those results are not reported in the interest of brevity. The results thus do not suggest that female CEOs tend to outperform their male counterparts. Instead, the benefits of female leadership participation appear to be driven by the fact that, for the reasons noted above, a more diverse leadership team tends to deliver better outcomes on average.

Five alternative specifications are reported in each table. In the first the shares of female board and executives are entered separately. In the second the board quota variable is added. In the third the total female share in corporate leadership replaces the separate female board and executive shares. The fourth and fifth specifications reproduce the first and third specifications, respectively, restricting the sample to profitable firms.

The evidence on the impact of female board membership on firm performance is not robust. The female board share is significantly correlated with gross revenue in the full sample, but the correlation becomes statistically insignificant when the sample is limited to profitable firms or net margins are used as the dependent variable.