Women On Boards In Finance And STEM Industries
Renee B. Adams
University of New South Wales, Department of Banking and Finance; ECGI; FIRN; ABFER
Tom Kirchmaier
London School of Economics – Financial Markets Group (FMG)
October 7, 2015
Abstract:
We document that women are less represented on corporate boards in Finance and more traditional STEM industry sectors. Even after controlling for differences in firm and country characteristics, average diversity in these sectors is 24% lower than the mean. Our findings suggest that well-documented gender differences in STEM university enrolments and occupations have long-term consequences for female business leadership. The leadership gap in Finance and STEM may be difficult to eliminate using blanket boardroom diversity policies. Diversity policies are also likely to have a different impact on firms in these sectors than in non-STEM sectors.
Women On Boards In Finance And STEM Industries – Introduction
Relative to their share in the population women are less likely to have a science or engineering degree and are less likely to be employed as scientists or engineers (CEOSE, 2013). For academia there is evidence that women are underrepresented at all academic levels. For example, even though women earn roughly half the doctorates in science and engineering in the United States, they comprise only 21% of full science professors and 5% of full engineering professors (Shen, 2013). There is also evidence that female academics in science are less likely to be invited to join corporate scientific advisory boards (McCook, 2013). But to our knowledge, there is as yet no evidence that the under-representation of women in STEM occupations persists at higher levels of the corporate hierarchy. We provide such evidence in this paper.
Using a comprehensive sample of board data for listed firms in 20 countries from 2001-2010, we show that the fraction of women on the board (Board Diversity) is lower for firms in the STEM and Finance sectors (STEM&F) than in the non-STEM sector. This finding is robust to controlling for firm and country characteristics and country and year fixed effects. On average STEM&F firms have 1.8% fewer women on boards than non-STEM firms. Relative to the sample mean of 7.56%, this represents an economically significant leadership gap in STEM&F fields. Women are most underrepresented on the boards in the “Natural Resources and Mining”, “Manufacturing” and “Financial Activities” sectors.
Our results connect two policy debates that are usually conducted separately: the debate about women’s under-representation in STEM fields and the debate about women’s under-representation on corporate boards. The fact that women are less represented on corporate boards in STEM&F fields suggests that women’s under-representation in STEM occupations may get worse at higher levels of the corporate hierarchy. Similar to the findings for academic positions (Shen, 2013), there may be biases or impediments to work-life balance that make it harder for women to achieve leadership positions in STEM&F sectors. This means that to solve the under-representation problem in STEM&F occupations, it may not be enough to simply encourage entry of women into these fields. More must be done to ensure they do not have reasons to exit the industry.
Recognizing that women’s under-representation on boards varies by sector is also important for the policy movement that aims to increase corporate board diversity. The EU recently approved a draft law that sets an objective of 40% female non-executive directors on boards of listed companies across the 28 member states of the EU (European Commission, 2012). Our results suggest that it will be more difficult for firms in STEM&F sectors to achieve these objectives. It is also plausible that diversity will have a greater impact (positive or negative) on firms in the STEM&F sectors than in other sectors. More generally, given that the under-representation of women on boards in STEM&F firms is likely due to the persistent under-representation of women in STEM&F fields, it is unlikely that board diversity targets can solve the problems leading to women’s under-representation on boards in these sectors. Policy makers interested in increasing board diversity may need to join forces with those who worry about the retention of women in STEM fields.
I. Data
Our data is from Adams and Kirchmaier (2015a). Their sample consists of data on boards of listed companies in 20 countries for the years 2001-2010. Countries in the sample are from Europe, the Commonwealth and the USA. Adams and Kirchmaier combine director data from Boardex with financial data from CapitalIQ and country-level data from Euromonitor, the OECD family database, the World Economic Forum and the World Value Survey amongst others. To ensure their county-level coverage is representative, they restrict their sample to county-years in which Boardex covers more than 70% of total market capitalization in that country and year and drop countries with low coverage and countries that are likely to be outliers with respect to women on boards. As a result, not all countries have 10 years of data in the sample.
Adams and Kirchmaier (2015a) drop financial firms, but we retain them. Our final sample consists of an unbalanced panel of 44,254 firm-year observations on more than 8,000 listed firms in 20 countries.
The dataset is complete with respect to gender and non-executive director (NED) and executive director (ED) classifications. In countries with a dual board system (Austria, Germany, Denmark, Netherlands), we classify supervisory board members as NEDs and management board members as EDs. Board size is the sum of the sizes of the supervisory and management boards.
We define STEM industries as industries in which a large share of employees is in STEM occupations. To determine which industries fall into this category, we first obtain a list of occupations that require education in science, technology, engineering, and mathematics disciplines from O*NET (2015). We match these STEM occupations to the 2012 Bureau of Labor Statistics’ National Employment Matrix by Industry. For each industry, the Employment Matrix indicates the percent of employees from each occupation. We sum the percentages for all STEM occupations to obtain the percent of employees in STEM occupations in each industry. We then average hese percentages across industries in each of 10 industry super sectors as defined by the Bureau of Labor Statistics (2015). We define the top 5 super sectors by share of STEM employees as STEM sectors. These are “Financial activities” with 7.16% STEM employees on average, “Natural Resources and Mining” (10.75% STEM employees), “Manufacturing” (15.1% STEM employees), “Professional and business services” (21.78% STEM employees) and “Information” (21.8% STEM employees). Non-STEM sectors are “Leisure and hospitality”, “Trade, transportation, and utilities”, “Educational and health services”, “Other services” and “Construction”.1
We match the super sector classification to our sample firms using NAICS codes from CapitalIQ. To highlight that the Finance sector is STEM-intensive even though it is not traditionally considered a STEM sector, we label firms in the top 5 STEM sectors as STEM&F firms.
See full PDF below.
