Statistical Discrimination Theory Research Articles

Establishment Closures in Germany: The Motherhood Penalty at Job Search Durations

AbstractThis study contributes to the research on motherhood penalties by investigating based on German longitudinal register data whether mothers take longer to search for a job than childless women. We develop a unique research design to test the theory of statistical discrimination as well as the theory of status-based discrimination. We use establishment closures as a starting point that creates equal conditions for all previously employed women. Following a subsequent coarsened exact matching approach, we investigate the job search length of almost 3,000 comparable full-time working childless women and mothers from almost 700 establishments by applying event history techniques. Even after extensive robustness checks that back up the main findings, we can show that mothers have lower transition rates to re-employment than childless women. When including additional information, such as higher age of the youngest child or shorter parental leave durations, both of which serve as indicators of higher labour market attachment, we find reduced differences and buffering effects on the motherhood penalty.

Getting a Lawyer While Black: A Field Experiment

In this Article, I present new evidence that African-Americans face unique impediments in obtaining access to counsel. Using a randomized audit design, I show that those with black-sounding names receive only half the responses of those with white-sounding names regarding requests for legal representation. I design a larger, follow-up experiment to evaluate variations on the theory of “statistical discrimination,” that lawyers are merely responding to economically-relevant signals correlated with race. I find no evidence supporting the expectations of the statistical discrimination theory, but some evidence that racial preferences matter. I conclude by presenting a more nuanced theory of racialized service rationing that is consistent with the body of experimental evidence presented and is supported by observational data. I discuss the implications of these theories for potential policy responses, including debates about affirmative action and the size of the legal profession.

Gender gap in entrepreneurship

Using data on the entire population of businesses registered in the states of California and Massachusetts between 1995 and 2011, we decompose the well-established gender gap in entrepreneurship. We show that female-led ventures are 63 percentage points less likely than male-led ventures to obtain external funding (i.e., venture capital). The most significant portion of the gap (65 percent) stems from gender differences in initial startup orientation, with women being less likely to found ventures that signal growth potential to external investors. However, the residual gap is as much as 35 percent and much of this disparity likely reflects investors’ gendered preferences. Consistent with theories of statistical discrimination, the residual gap diminishes significantly when stronger signals of growth are available to investors for comparable female- and male-led ventures or when focal investors appear to be more sophisticated. Finally, conditional on the reception of external funds (i.e., venture capital), women and men are equally likely to achieve exit outcomes, through IPOs or acquisitions.

Gender Gap in Entrepreneurship

Using data on the entire population of businesses registered in the states of California and Massachusetts between 1995 and 2011, we decompose the well-established gender gap in entrepreneurship. We show that female-led ventures are 63 percentage points less likely than male-led ventures to obtain external funding (i.e., venture capital). However, investors’ gendered preferences can, at most, explain about 35 percent of this differential (or 22 percentage points). The most significant portion of the gap (65 percent) stems from gender differences in initial startup orientation, with women being less likely to found ventures that signal growth potential. Moreover, consistent with theories of statistical discrimination, the residual gap diminishes significantly when stronger signals of growth are available to investors for comparable female- and male-led ventures or when focal investors are more sophisticated. Finally, conditional on the reception of external funds (i.e., venture capital), women and men are equally likely to achieve exit outcomes, through IPOs or acquisitions.

The Relationship Between Male and Female Youth Volunteering and Extrinsic Career Success: A Growth Curve Modeling Approach

Drawing from investment and statistical discrimination theories, we test a model to examine the income growth trajectories of male and female youth volunteers and non-volunteers. Using growth curve modeling for four waves of longitudinal data for the reference period 2001-2007 ( n = 7,447), we find that male and female youth volunteers face an initial earnings disadvantage vis-à-vis youth non-volunteers; this penalty is smaller for females compared with males. However, over time, the income growth of volunteers is higher than that of non-volunteers. Male volunteers experience faster earnings growth than female volunteers. Furthermore, we find that, given the more rapid earnings growth of male volunteers relative to female volunteers, volunteering serves to widen, rather than narrow, the gender wage gap. The implications for future research and the relevance of the findings for career counselors, youth, and voluntary organizations are discussed.

Rental housing discrimination across protected classes: Evidence from a randomized experiment

This paper provides the first evidence of how landlord treatment of rental housing applicants varies across race, gender, religion, sexuality, and family status. We adopt a randomized correspondence audit design by sending inquiry emails to randomly selected property owners or agents from Craigslist. The signals are conveyed by names, body text, and signature-line quotations. The findings suggest landlords have preferences about tenants and make decisions based on signals communicated in inquiry emails from potential applicants. Blacks, Arab males, Muslims, and single parents are treated unfavorably while gay couples are preferred. These discrimination patterns are generally consistent with the theory of agent-based statistical discrimination. The results provide evidence that many groups protected under U.S. fair housing law continue to face obstacles when searching for housing despite decades of fair housing law.

Regional ethnic diversity and the employment prospects of immigrants

ABSTRACTThis paper develops a model based on statistical discrimination theories which predicts that higher educated immigrants have better and less educated immigrants have worse employment prospects in ethnically more diverse regions. Evidence consistent with this hypothesis is found in an empirical analysis. These results, however, also show that even the least educated recent immigrants have higher employment probabilities in ethnically more diverse regions. This suggests that the positive effects of ethnic diversity on overall labour demand compensate for any negative effects on less educated recent immigrants.

Discrimination and Administrative Burden in Public Service Markets: Does a Public–Private Difference Exist?

Public services are often provided in markets where both public and private providers operate. Irrespective of ownership status, public services ought to be accessible regardless of clients’ race, gender, ethnicity, or age. However, as theories of statistical discrimination and cream skimming suggest, market-based incentives may lead service providers to focus on non-minority clients because they perceive them as easier-to-serve and therefore less costly. This may lead to discrimination and hence jeopardizes equal access. In this study, we ask whether private, for-profit providers are more likely to discriminate on ethnic grounds compared to publicly owned providers. We implement a field experiment within the Flemish elderly care market by sending out email requests with either a Flemish or a Maghrebian name to all public and privately owned nursing homes. For overall response rates, no statistically significant differences between senders were found. However, we do find that privately owned facilities are about 20 percentage points less likely to provide information on how to enroll when the request is send from a Maghrebian name alias, and blinded coders perceive the information sent to the Maghrebian alias as less comprehensive. In publicly owned facilities, no such differences exist. We conclude that a public-private difference does exist, but that the mechanism of discrimination is subtler than expected. Rather than directly refusing to respond, nursing homes increase administrative burdens and learning costs for minority applicants.

Gender-oriented statistical discrimination theory: Empirical evidence from the Hong Kong labor market

This paper proposes a simple search model to shed light on the role of aggregate fertility as a form of statistical discrimination against young working women in the labor market. Our proposed theory assumes that workers generate identical production value that does not differ by gender. When matched with a firm, the worker and the firm bargain over wage rates. Since female pregnant workers receive wages during maternity leave and generate no production value, a matched pair with a female worker generates lower expected profit to the firm. Rent-sharing ensures a male worker to be paid more than a female worker even with the absence of overt discrimination, thus resulting in statistical discrimination against female workers. Using the 5% random subsamples of census and by-census data in Hong Kong in 1996, 2001, and 2006, the study provides unequivocal evidence that age-specific fertility rates exert negative impact on female wages, after controlling for other observable characteristics. More importantly, the negative effects of fertility on women's earnings are notably greater in female dominated occupations, whereas similarly situated men are unaffected. Since fertility rates tend to be increasing with age during the early stage of women's work careers, our theory partially explains why the gender gap rises with age.

L’économie des discriminations : conquête impérialiste ou contribution à la théorie des prix?

Becker’s Ph.D. thesis on discrimination is commonly viewed as an expression of economics imperialism. This paper questionned the imperialist nature of the study of discrimination in economics, using Becker’s work as a turning point in history of economics. Debates on wage inequalities and discrimination occured in economics before Becker’s work and are central to the discipline (section 1). Becker’s model is more a contribution to the development of “analytical economics” than an explanation of discrimination – essentially because it aims at studying the economic consequences of discrimination (section 2). Statistical discrimination theories – built to oppose Becker’s model – are more representative of economics imperialism because their objective is to explain causes of the discrimination phenomenon. (This abstract was borrowed from another version of this item.)

Part-Time Work and Employer-Provided Training: Boon to Women and Bane to Men?

Previous studies on employer-provided training have consistently shown a gap in training participation between part-time and full-time workers. This study examines whether the training disadvantage for part-time workers differs by gender. To capture the uncertainty in the firm’s training decision and to factor in heterogeneity among part-time workers, our analysis draws not only on human capital but also on statistical discrimination theory. Our empirical results indicate that gender plays a role in determining part-time/full-time training differences. Whereas for women working part-time or full-time makes only a minor difference, for men working part-time constitutes a serious disadvantage in access to employer-provided training. The results remain consistent among different subsamples.

What's in a Picture?: Evidence of Discrimination from Prosper.com

We analyze discrimination in a new type of credit market known as peer-to-peer lending. Specifically, we examine how lenders in this online market respond to signals of characteristics such as race, age, and gender that are conveyed via pictures and text. We find evidence of significant racial disparities; loan listings with blacks in the attached picture are 25 to 35 percent less likely to receive funding than those of whites with similar credit profiles. Conditional on receiving a loan, the interest rate paid by blacks is 60 to 80 basis points higher than that paid by comparable whites. Though less significant than the effects for race, we find that the market also discriminates somewhat against the elderly and the overweight, but in favor of women and those that signal military involvement. Despite the higher average interest rates charged to blacks, lenders making such loans earn a lower net return compared to loans made to whites with similar credit profiles because blacks have higher relative default rates. This pattern of net returns is inconsistent with theories of accurate statistical discrimination (equal net returns) or costly taste-based preferences against loaning money to black borrowers (higher net returns for blacks). It is instead consistent with partial taste-based preferences by lenders in favor of blacks over whites or with systematic underestimation by lenders of relative default rates between blacks and whites.

Do expert patients get better treatment than others? Agency discrimination and statistical discrimination in obstetrics

We address models that can explain why expert patients (obstetricians, midwives and doctors) are treated better than non-experts (mainly non-medical training). Models of statistical discrimination show that benevolent doctors treat expert patients better, since experts are better at communicating with the doctor. Agency theory suggests that doctors have an incentive to limit hospital costs by distorting information to non-expert patients, but not to expert patients. The hypotheses were tested on a large set of data, which contained information about the highest education of the parents, and detailed medical information about all births in Norway during the period 1967–2005 (Medical Birth Registry). The empirical analyses show that expert parents have a higher rate of Caesarean section than non-expert parents. The educational disparities were considerable 40 years ago, but have become markedly less over time. The analyses provide support for statistical discrimination theory, though agency theory cannot be totally excluded.

Discrimination, Complexity, and the Public/Private Sector Question

Does employment discrimination vary in degree or character across public and private labor market sectors? Prior research cannot fully address this question because it typically relies on one dimension of discrimination—estimates of wage gaps. This study extends the literature by analyzing 11,528 legally verified cases of race and sex discrimination from the Ohio Civil Rights Commission (1986-2003). Quantitative analyses demonstrate that aggregate rates of verified discrimination vary little by sector, yet there are elevated rates of public sector promotion discrimination and elevated rates of private sector firing discrimination. In-depth qualitative analyses show that specific sectoral processes contribute to these aggregate patterns. In the public sector, limited accountability for promotion decisions allows managers to devalue seniority, augment “soft skills,” and sabotage multiple stages of formalized proceedings. Moreover, the very devices intended to curb discriminatory promotion may inadvertently multiply the stages for bias to enter decisions. In the private sector, managers exploit the latitude afforded by the employment-at-will doctrine to differentially terminate workers, sometimes justifying their actions as cost saving in a competitive market. The author argues that these processes are in line with statistical discrimination and social closure theories and concludes by discussing their implications for understandings of workplace inequality.

The Dynamics of Collective Reputation

I present a stochastic version of Tirole's (1996) collective reputation model. In equilibrium, group behavior is persistent due to complementarity between the current incentives of group members and the group's reputation, which depends on its history. A group can maintain a strong reputation even as conditions become unfavorable, but an improvement in the environment may not help a group with a poor reputation. I also connect the model to the theory of statistical discrimination and show that the same mechanism can explain why discrimination might persist over time.

Analysing the Gender Wage Gap (GWG) Using Personnel Records

AbstractWe use monthly personnel records of a large German company for the years 1999–2005 to analyse the gender wage gap (GWG). The unconditional GWG is 15 per cent for blue‐collar and 26 per cent for white‐collar workers. Different returns to entry age explain a substantial part of the GWG as well as segregation of men and women in different hierarchical levels. The relative GWG increases with increasing tenure for blue‐collar but declines for white‐collar workers. Taking into account the different impact of general and firm‐specific human capital on white‐collar and blue‐collar occupation, this is consistent with theories of statistical discrimination.

Disparity and diversity in the contemporary city: social (dis)order revisited1

Disparity and diversity in the contemporary city: social (dis)order revisited¹

Racial disparities in training, pay-raise attainment, and income

Data from the 2001 to 2002 California Workforce Survey are used to assess statistical discrimination and social-closure theories in explaining racial disparities in crucial labor outcomes. The results are mixed. Racial differences between minority groups and whites appear in three crucial labor outcomes: training attainment, pay raises, and income. In particular, compared with white workers, Hispanics are less likely to obtain employer-paid job training, blacks are less likely to receive pay raises, and receive lower income. In-depth statistical analyses are conducted to identify the roots of those inequalities. The conclusion summarizes major findings and discusses their implications for future studies in ascriptive inequalities.

Coaching and Athletic Career Investments: Using organisational theories to explore women's boxing

The following paper explores the extent to which women choose to invest in the sport of boxing, the ways coaches perceive women in boxing, and whether these perceptions influence coaches’ decisions in training female boxers. Human capital theory and statistical discrimination theory are applied to the sport of female boxing to better understand women's participation in the sport, and the gendered expectations coaches have for women in the sport. In‐depth interview data with boxers and coaches were analysed in relation to these theories. Findings suggest that while women are willing to invest in their personal athletic careers, coaches are hesitant to invest time and training in female boxers because of the boxers’ current or expected family obligations. This research has important implications for women's participation in male‐dominated leisure activities, like boxing, and the conflict that arises between family and leisure obligations.

Of Mommy Tracks and Glass Ceilings : A Case Study of Men's and Women's Careers in Management

Data from a 1989 survey ofover 600 middle-level managers in a large Canadian corporation were analyzed to examine the characteristics of jobs held by career-family and career-primary men and women. Hypotheses were developed based on human capital theory, statistical discrimination theory, and gender role congruence theory. Examining career outcomes suggested that participation in household labor had a significantly more negative association with men's hierarchical level than with women's. Implications for theory and suggestions for research are discussed.