At the Table but Can't Break Through the Glass Ceiling: Board Leadership Positions Elude Diverse Directors
Laura Casares Field, Matthew Souther & Adam Yore
Journal of Financial Economics, forthcoming
We explore the labor market effects of gender and race by examining board leadership appointments. Prior studies are often limited by observing only hired candidates, whereas the boardroom provides a controlled setting where both hired and unhired candidates are observable. Although diverse (female and minority) board representation has increased, diverse directors are significantly less likely to serve in leadership positions despite possessing stronger qualifications than nondiverse directors. While specialized skills such as prior leadership or finance experience increase the likelihood of appointment, that likelihood is reduced for diverse directors. Additional tests provide no evidence that diverse directors are less effective.
The Symbolic Value of Ethnic Spaces
Teri Kirby et al.
Social Psychological and Personality Science, forthcoming
In four experiments, students read that their university was creating either an ethnic space (a space geared to people of particular ethnic groups) or a general space for students. In an internal meta-analysis, underrepresented students of color (N = 205), but not White students (N = 760), who read about the ethnic space reported greater belonging, value of underrepresented students by the university, support, and academic engagement compared to those who read about a general space. Ethnic spaces may hold broader psychological significance than that of mere gathering places, improving outcomes even for those who do not frequently use them. Creating ethnic spaces may be one strategy for making university environments more welcoming for underrepresented students of color.
An Economic Approach to Regulating Algorithms
Ashesh Rambachan et al.
NBER Working Paper, May 2020
There is growing concern about "algorithmic bias" - that predictive algorithms used in decision-making might bake in or exacerbate discrimination in society. When will these "biases" arise? What should be done about them? We argue that such questions are naturally answered using the tools of welfare economics: a social welfare function for the policymaker, a private objective function for the algorithm designer and a model of their information sets and interaction. We build such a model that allows the training data to exhibit a wide range of "biases." Prevailing wisdom is that biased data change how the algorithm is trained and whether an algorithm should be used at all. In contrast, we find two striking irrelevance results. First, when the social planner builds the algorithm, her equity preference has no effect on the training procedure. So long as the data, however biased, contain signal, they will be used and the algorithm built on top will be the same. Any characteristic that is predictive of the outcome of interest, including group membership, will be used. Second, we study how the social planner regulates private (possibly discriminatory) actors building algorithms. Optimal regulation depends crucially on the disclosure regime. Absent disclosure, algorithms are regulated much like human decision-makers: disparate impact and disparate treatment rules dictate what is allowed. In contrast, under stringent disclosure of all underlying algorithmic inputs (data, training procedure and decision rule), once again we find an irrelevance result: private actors can use any predictive characteristic. Additionally, now algorithms strictly reduce the extent of discrimination against protected groups relative to a world in which humans make all the decisions. As these results run counter to prevailing wisdom on algorithmic bias, at a minimum, they provide a baseline set of assumptions that must be altered to generate different conclusions.
Are Single Women Promoted into Leadership Positions in Business? Evidence of Incongruity Penalties of Single Women with Masculine Talents
Jennifer Merluzzi & Damon Phillips
Columbia University Working Paper, September 2019
We advance scholarship on gender inequality in leadership by drawing attention to a growing but less studied group: single women (non-mother) professionals. Our thesis is that single, non-mother status is perceived as incongruent with societal and role expectations of leadership for women, stunting their early career advancement in the workplace relative to other gender and marital status groups. In line with role expectations theory, we expect this incongruity with leadership to be penalizing under conditions that make singlehood status salient for women in the workplace, namely having gender inconsistent (stereotypically masculine) talents. Using a multi-method approach, we investigate the early career promotions of men and women MBAs and present evidence that supports our thesis. In Study 1, an experiment with MBA students, we observe a negative promotion bias against analytically-talented (masculine) single women MBAs whom participants assess as inferior leaders compared to otherwise identical analytically-talented single men, married men, and married women candidates. Study 2 provides external validity by analyzing the actual early career mobility outcomes of two cohorts of MBA graduates and shows consistent evidence of an incongruity penalty toward analytically-talented (masculine) single women. The combined studies unveil a discrimination-based penalty where singlehood status conflicts with the role of leadership ascribed to women when they excel in a masculine talent.
A Tale of Two Forums: Employment Discrimination Outcomes in Arbitration and Litigation
ILR Review, forthcoming
This article presents data from a novel survey of 1,256 employment plaintiff attorneys to test whether employee rights and remedies are affected by mandatory employment arbitration. By surveying attorneys directly about their most recent employment discrimination cases taken to verdict in arbitration and civil litigation, the author presents a systematic empirical comparison of outcomes between civil courts and arbitration with robust controls. The ability to control for the legal basis of the claim, defendant size, use of summary judgment, and attorney and plaintiff characteristics significantly improves on previous empirical research studies. Consistent with previous research, employee win rates in arbitration are lower than those found in state and federal court. In addition, monetary award amounts and percentage of claim amount awarded to employees who prevail in their cases are significantly lower in arbitration compared to outcomes in state and federal jury trials.
College Achievement and Attainment Gaps: Evidence from West Point Cadets
Dario Cestau et al.
NBER Working Paper, May 2020
Assessing the effectiveness of education by race and gender is as difficult as it is important. We investigate this question utilizing data for eleven cohorts at West Point, a distinguished military academy and highly ranked liberal arts college. Employing matching using entry scores on three comprehensive measures, we obtain exceptional matches of score distributions for black and matched white students. We find black students have lower graduating achievement scores than matched white students, but comparable rates of graduation, retention in the Army after graduation, and early promotion. Hispanic-white comparisons reveal no differences. Female-male comparisons reveal women have lower attainment and retention rates.
Can Better Selection Tools Help Us Achieve Our Diversity Goals in Postgraduate Medical Education? Comparing Use of USMLE Step 1 Scores and Situational Judgment Tests at 7 Surgical Residencies
Aimee Gardner et al.
Academic Medicine, May 2020, Pages 751-757
Purpose: Use of the United States Medical Licensing Examination (USMLE) for residency selection has been criticized for its inability to predict clinical performance and potential bias against underrepresented minorities (URMs). This study explored the impact of altering traditional USMLE cutoffs and adopting more evidence-based applicant screening tools on inclusion of URMs in the surgical residency selection process.
Method: Multimethod job analyses were conducted at 7 U.S. general surgical residency programs during the 2018-2019 application cycle to gather validity evidence for developing selection assessments. Unique situational judgment tests (SJTs) and scoring algorithms were created to assess applicant competencies and fit. Programs lowered their traditional USMLE Step 1 cutoffs and invited candidates to take their unique SJT. URM status (woman, racial/ethnic minority) of candidates who would have been considered for interview using traditional USMLE Step 1 cutoffs was compared with the candidate pool considered based on SJT performance.
Results: A total of 2,742 general surgery applicants were invited to take an online SJT by at least 1 of the 7 programs. Approximately 35% of applicants who were invited to take the SJT would not have met traditional USMLE Step 1 cutoffs. Comparison of USMLE-driven versus SJT-driven assessment results demonstrated statistically different percentages of URMs recommended, and including the SJT allowed an average of 8% more URMs offered an interview invitation (P < .01).
The Gender Gap in Tech & Competitive Work Environments? Field Experimental Evidence from an Internet-of-Things Product Development Platform
Kevin Boudreau & Nilam Kaushik
NBER Working Paper, May 2020
Many technology companies struggle to fill all their positions and to achieve gender parity in their ranks. One explanation for gender disparities is the possibility that men and women differ in their willingness to work under competitive organizational environments of tech firms. To investigate this question, this paper reports on a large platform-based field experiment in which 97,696 U.S. university-educated individuals were given the opportunity to join a tech-related product development activity. Individuals were randomly assigned to treatments emphasizing either competitive or collaborative interactions with other participants. We find that (1) in non-STEM fields, the competition treatment leads to a 27% drop in participation for females in comparison to males. However, in our main finding, (2) in STEM fields, we find no statistical differences in men and women's responses to competition. The patterns are consistent with (3) men in non-STEM fields exhibiting overconfidence in their likelihood of succeeding under competition. We also find that, while participation in highest in STEM fields, (4) the ratio of female to male participation in a field is better predicted by whether the field is male- or female-dominated, than it is by whether it is a STEM field or not. We discuss theoretical interpretations and implications for organizations.
Why are US Corporate Boards Under-Diversified among Genders and Races?
Zawadi Lemayian, Grace Pownall & Justin Short
Emory University Working Paper, April 2020
The popular and academic literatures report that women, Blacks, Hispanics, and Native Americans are under-represented on US corporate boards of directors, relative to their incidence in the US labor force. We confirm this observation by tabulating the presence of female and under-represented racial minority (URM) directors on the boards of the S&P 1500 firms. We examine two hypotheses that may explain this empirical regularity. The first is that qualified female and URM candidates are scarce. Firms must compete for their services, putting price pressure on their compensation and leading to higher compensation for female and non-URM directors. The second hypothesis is a perception that women and URM directors in general are less valuable to firms, leading to lower compensation for them relative to their white male peers. Our evidence is mixed: we find that women (URM) make significantly higher (insignificantly lower) compensation than males and non-URM directors, controlling for firm and director variables (like age, education, achievements, and committee appointments). However, when we compare the compensation of female and URM directors against male and non-URM directors' compensation on the same boards, that significant difference reverses, even though they have more education and more achievements than those peers. This result is consistent with the "myopic hypothesis" that predicts female and URM are paid less because of a biased view of their value to the board. However, when we compare the average compensation of female and URM directors who hold board leadership roles (dual CEO-Chairs and independent lead directors), we find very low participation of female and URM directors, but they are paid substantially more than their male and non-URM counterparts. We conclude that females and URM directors are chosen by larger firms with larger and more active boards (and higher base compensation), but are very rarely chosen to lead those boards. When they are chosen for board leadership positions, they are paid significantly more than their non-URM male peers. This result is more consistent with the "glass ceiling" hypothesis that appropriate candidates who add to the diversity of the board are scarce and command higher compensation.
Who Can Access the "Good" Jobs? Racial Disparities in Employment among Young Men Who Work in Paid Care
ANNALS of the American Academy of Political and Social Science, March/April 2020, Pages 55-76
Men have slowly increased their presence in paid care jobs that have long been considered as "women's jobs." But job growth in the paid care sector is polarized between "good" jobs and "bad" jobs in terms of pay and job security, and racial minority men are more likely to enter low-paying care-work jobs. Using work history data from the National Longitudinal Survey of Youth 1979 and 1997, this study examines the patterns and mechanisms of racial disparity in young men's access to jobs of varying pay levels in the care-work sector and how such patterns have changed as the labor market has become more precarious and unequal. Findings suggest that young black men - especially those without a college education - have been increasingly excluded from accessing "good" jobs in the paid care sector. Moreover, this black-white disparity cannot be fully explained by racial differences in individual-level characteristics.