No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
From taste-based to statistical discrimination
Abstract
Consider hiring managers who care not just about productivity but also some other, unrelated characteristic. If they treat that ascriptive characteristic differently across groups by, for example, valuing beauty more for women than men, then the hired women will be better looking but less productive, on average. This taste-based discrimination, focused entirely on an ascriptive characteristic, can lead to productivity-based statistical discrimination by the firm’s subsequent hiring managers who observe from their workforce that women tend to produce less. This identifies a new channel behind statistical discrimination that arises from the behavior of prior hiring managers.
... There are two main explanations for discrimination based on the beauty premium: taste-based and statistical discriminations (Neilson and Ying 2016). Taste-based discrimination refers to the tendency of employers to discriminate against certain groups of people regardless of their productivity (Becker 1957). ...
... On the other hand, in statistical discrimination employers are posited to assess productivity based on limited information, and therefore rely on physically attractiveness. Physically attractive people are often believed to be more confident, have more pleasing personalities and more likely to be successful in their careers (Dipboye, Arvey, and Terpstra 1977;Miller 1988;Mobius and Rosenblat 2006;Neilson and Ying 2016). ...
The presence of a beauty premium in the labor market has been convincingly demonstrated in developed countries, although studies examining this issue in other countries have been few. Additionally, two related issues remain unclear: (i) Does the beauty premium have the same effect on the preferences of male and female personnel in charge of hiring? (ii) Does the interaction of attractiveness with creativity influence their evaluation differently? This study investigates the issues above in the context of a non-western country. A field experiment was conducted and the data were analyzed using regressions to test the interaction of the three factors (attractiveness, creativity and gender). We found that in common with findings elsewhere beauty is still used to decide who to hire, when no additional information is available, regardless of the gender of the hiring personnel. However, when information on creativity becomes available, male hirers give priority to it while female hirers continue to emphasize good looks. The reasons for this difference and whether or not this finding is unique to Malaysia are worth investigating in future studies.
... The current paper contributes to the literature that theoretically models the beauty premium. A study that is closely related to ours is Neilson and Ying (2016). They provide a theoretical model with a hiring manager showing discrimination in regard to physical appearance differently between genders, and valuing beauty more for women than men. ...
... 1994 ;Harper, 2000;Borland and Leigh, 2014). One explanation can be that employers' marginal rate of substitution between ability and beauty, or their minimum beauty threshold, differs for the two groups (Neilson and Ying, 2016). Another explanation is women's self-selection on labor participation decisions: more attractive women may be more likely to work because they are usually more confident (Mocan and Tekin, 2010), and bad-looking women are more likely to avoid the labor market due to plainness penalty (Hamermesh, 2011). ...
Using a nationwide representative sample of the Chinese population, we examine the effect of physical appearance on the hourly wage. Our theoretical framework that connects physical appearance and the hourly wage has two predictions. First, the beauty premium is larger for high-skilled workers than for low-skilled workers. Second, this statistical-based discrimination arises from the hiring manager’s belief that beauty contributes to productivity through non-cognitive skills. Empirical results confirm both predictions. Good-looking individuals earn roughly 5.4% more than the rest, and bad-looking individuals earn roughly 3.3% less than the rest. Moreover, quantile estimates show that the effect of physical appearance at the 0.75 quantile is more than twice as the one at the 0.25 quantile. Finally, social skills, communication skills, and self-confidence are three transmission channels through which physical appearance can positively affect one’s hourly wage in our sample. Our findings imply that the beauty premium is widespread in the labor market and relevant anti-discrimination regulations are needed.
... It focuses on observable behaviour (Quillian 2006) which can be motivated by various factors, including prejudice, stereotypes, ideologies, or economic considerations, without presuming a singular underlying cause. Contemporary manifestations of discrimination often take subtle and covert forms due to significant shifts in social norms (Chang et al. 2019), posing challenges in conceptualizing and measuring it in the social sciences (Pager and Shepherd 2008). 2 In the economics literature, discrimination is categorized into taste-based and statistical discrimination (Neilson and Ying 2016). Taste-based discrimination emphasizes prejudice or racial animosity as driving forces (Rivera and Tilcsik 2016). ...
... The role of discrimination against women has been well-documented in the literature. Neilson and Ying (2016) wrote about 'taste discrimination'. This concept refers to people's preferences to work with others who are most similar to themselves. ...
This chapter asks how barriers to equal participation in adult education play out differently between (a) men and women and (b) women from different class and ethnic backgrounds. European Social Survey and Adult Education Survey data analyses confirm that low social origin is a significant barrier. Other socio-economic and socio-demographic characteristics can also act as barriers. The chapter draws attention to intersectionality: multiple simultaneous disadvantages, including gender (for women), social origin (for adults whose parents have basic or lower levels of education), and having an ethnic minority or migrant background. Women are less likely to receive financial support from their employer, though more likely to do so from public institutions. Ethnic minority status has a stronger negative effect for women than for men. Mediterranean and post-socialist welfare regimes reduce the effects of intersectionality less than social-democratic regimes.
... In statistical discrimination, employers can assess productivity based on limited information and rely on physical attractiveness. Physically attractiveness could be the determinant of resume evaluations and more likely to be successful in their careers (Mobius & Rosenblat, 2006;Neilson & Ying, 2016;Ch'ng & Narayamam, 2021). In the latter case, agents will place greater weight on information from the group that they are more familiar with or can associate with. ...
This study aims to estimate gender discrimination in the labour market by using fictitious job applicants, carrying gender identifiable names at the top of the curriculum vitae. Fictitious job applications were sent to genuine vacancies in Malaysia. An audit experiment was conducted on new graduate employment in a labour market with no legislation against discrimination. We recorded if there was a significant difference in call-back interview rates. We estimated the extra amount of applications females would need to make to obtain the same number of interview calls as males. Results show a female bias in call-backs was evident only in broader markets that traditionally offer female employment.
... Differential assumptions about how profitable or difficult prospective patients will be to treat may result in statistical discrimination (Arrow, 1973) if physicians rely on group stereotypes to form expectations about individuals. These types of discrimination can be difficult to separate, as taste-based discrimination may lead to statistical discrimination, and the two can occur simultaneously (Neilson and Ying, 2016). ...
We examine how differences in questions asked and information provided by physicians’ offices contribute to differences in new-patient appointment offers. Data is from a 2013–16 field experiment involving calls to a random sample of US primary care physicians on behalf of simulated new patients differentiated by race/ethnicity (Black, Hispanic, White), sex, and insurance. We find that the rates and stated reasons for denial of appointment offers differ substantially across patient groups.
... To former for 'pre-entry' diskrimination forekommer. (1) Smags-eller praeferencediskrimination refererer til, at arbejdsgiveren frasorterer seniorer, fordi arbejdsgiveren har en personlig aversion eller uvilje mod at ansaette denne kategori af lønmodtagere (Becker, 1971;Neilson & Ying, 2016). (2) Statistisk diskrimination beskriver en situation, hvor kategorier af jobansøgere bliver fravalgt på basis af antagelser om gruppens gennemsnitlige produktivitet; dvs. ...
I de senere år er der sket en markant ændring i seniorernes vilkår på arbejdsmarkedet. Antallet af beskæftigede seniorer er nærmest eksploderet. Beskæftigelsesfrekvensen blandt de 55-64-årige er steget fra 56 til 71 i perioden 2000 til 2018. Imidlertid ved vi meget lidt om, hvor på arbejdsmarkedet seniorerne har fundet beskæftigelse, hvorfor formålet med denne artikel er at analysere, hvad det er for type virksomheder (og brancher), der især er disponerede for at ansætte nye 55-64-årige medarbejdere. Det væsentligste fund er, at seniorer rekrutteres på virksomheder, hvor gennemsnitsalderen på arbejdspladsens ledelseslag er høj, men især på virksomheder hvor der i forvejen er mange seniorer ansatte. Ligeledes har virksomhedens størrelsen en betydning for, om der rekrutteres seniorer. Især store virksomheder er tilbøjelige til at trække seniorer ind i virksomheden.
... Hiring managers set different group-based thresholds of jobirrelevant characteristics (like ethnic accent), leading to lower ability estimations of minority members (i.e., taste-based discrimination). Statistical discrimination occurs when observed group differentials in productivity are mistakenly attributed to group identity (Neilson & Ying, 2016) Statistical discrimination theory Majority workers/employers disavow minority individuals because they consider minority groups as a whole less productive from an economic point of view. ...
Resume screening is the first hurdle applicants typically face when they apply for a job. Despite the many empirical studies showing bias at the resume‐screening stage, fairness at this funnelling stage has not been reviewed systematically. In this paper, a three‐stage model of biased resume screening is presented. We first discuss relevant theoretical perspectives (e.g., job market signalling and impression formation theories) to explain why resume screening is vulnerable to biased decision‐making and ethnic discrimination in particular. On the basis of the best available evidence, we consider contingencies of ethnic discrimination in the applicant, the decision‐maker, and the broader context (e.g., organisation), as well as the effectiveness of interventions that might counter ethnic bias in resume screening. The paper ends with a critical agenda for further research and practice.
... Tastebased discrimination occurs when people consciously discriminate based on prior beliefs of the victim's group membership. Research has found that taste-based discrimination often leads to statistical discrimination where a manager may have an added standard of beauty for women that is not present for men which then creates an environment where men perform better than women (Neilson and Ying, 2016). Statistical discrimination can then be justified under this artificially created environment. ...
Does our perception of corruption affect our interactions? This question was answered through an experimental survey of 90 participants and the Corruption Perception Index (CPI) from Transparency International’s 2017 Report. The survey revealed that those who were categorized as low-corrupt with an above average CPI exhibited altruism towards individuals from perceived high-corrupt regions but discriminated negatively against them when restrictions were placed on the interaction. In the experiment, participants were allowed to give any amount between zero and six dollars in both games. Altruism was measured through the dictator game and low-corrupt participants were found to give one dollar more, $3.60, to those perceived to be high-corrupt than they gave to participants who were perceived to be low-corrupt, $2.60. The ultimatum game was the measure for a restricted environment and low-corrupt participants gave sixty cents less, $2.80, to those perceived to be high-corrupt than they gave to participants perceived to be low-corrupt, $3.40. Other findings from the survey exhibited discriminatory behavior by categorized high-corrupt participants but were ultimately not statistically significant.
... His theory of statistical discrimination is based on the scarcity of information about workers and jobs, and the idea that group membership provides information about relevant perceived characteristics. Neilson and Ying (2016) explore how some practices of taste-based discrimination in the work place can ended up in statistical discrimination practices. For example, when employers hire women on taste-based characteristics, they can end up distorting information about the entire group. ...
Purpose
The purpose of this paper is to explore the inhibitors preventing Spanish women from attaining positions of leadership
Design/methodology/approach
Using a qualitative approach based on focus groups and structured personal interviews on 42 Spanish male and female executives, the study tests the main gender normative expectations applicable to specific work and personal roles.
Findings
Findings suggest that there is significant consensus about the existence of traditional and emerging models. Research also suggests there are still clear gaps between traditional and emerging roles that are easily identified and recognized. Contrary to expectations, the research suggests there is no a clear model of emerging gender roles, and instead considerable diversity in the construction of individual aspirational models
Research limitations/implications
This is a first exploratory investigation limited to a reduced sample in the Spanish context. To confirm the findings, the research would benefit from a quantitative approach based on larger samples. Moreover, additional surveys in different cultural contexts will provide a broader understanding of the proposed research questions.
Practical implications
The gender gap framework can enable policy makers to correctly diagnose the barriers women face in their professional lives and to come up with efficient instruments to correct existing inequalities.
Originality/value
The principal contribution of this paper is to provide important insights into traditional and aspirational gender gaps that constitute one important internal barrier for women’s development.
Evidentiary rules for criminal trials disallow various forms of probative evidence. Conventional wisdom assumes these rules benefit all defendants, whether guilty or innocent, and thus reduce wrongful convictions at the price of more wrongful acquittals. We show the conventional view only holds under stylized conditions. We further identify properties of evidence generation mechanisms under which the conventional view is backward: an evidentiary rule will harm the innocent and protect the guilty. However, if adjudicators place too much weight on the evidence, its exclusion can reduce both wrongful convictions and wrongful acquittals.
Previous research has demonstrated that social stereotypes associated with women's gender can preclude them from leadership positions. It remains unclear whether these stereotypes affect how people perceive male and female leaders, however. To examine people's stereotypes, we extracted their mental representations of male and female leaders and typical men/women (referred to as nonleaders) using reverse correlation. We then asked perceivers to rate these prototypes’ apparent leadership ability and traits related to power and warmth across contexts that represented typically masculine, feminine, or neutral domains. Leaders in a feminine context appeared more leaderlike than nonleaders, but as equally leaderlike in neutral and masculine contexts. Moreover, female leader faces appeared more powerful than female nonleader faces but male leader and nonleader faces appeared equally powerful. Male leaders were perceived as warmer than male nonleaders, however, whereas female leaders and nonleaders were perceived as equally warm. Thus, people’s gender, social stereotypes, and the context in which leaders are judged influence how people conceive of male and female leaders, with counterstereotypical attributes distinguishing leaders within their gender.
Among the countries of the Organization for Economic Co-operation and Development, South Korea shows the worst female earnings disadvantage. Women's career disruption associated with marriage and childbearing is said to be the primary factor behind the huge female disadvantage in Korea. Recent studies, however, demonstrated that substantial female disadvantage appears prior to women's career disruption, even net of human capital covariates. In this study, we examine whether taste-based gender discrimination is a source of female earnings disadvantage. We use sex ratios of births across regions during the 1990s as a proxy of prejudice against women in current labor markets. Our empirical results show that female earnings disadvantages among 2009–2017 college graduates are larger in the regions where sex ratios of 1990–1999 newborns were higher. Our results are robust to the control of an extensive set of human capital variables, including concrete college names, detailed fields of study, high school types, and more. Depending on models, one-fourth to one-third of female earnings disadvantage is attributable to prejudice against women. Implications of these findings are discussed.
For the first time, the Max Planck Society is publishing an anthology on gender research, "Fundamental Questions". Thanks to the broad spectrum of disciplines and cultures represented in the Research Association, the authors, who come from various institutes, present their findings in numerous fields of research: law, art history, history of science, neuroscience and computer science. The approaches, topics, issues and methodology of the collected contributions are equally diverse. This diversity shows in the best possible way that the integration of the gender perspective is beneficial not only for applied science and development, but also for basic research. With contributions by Dr. Laura A. Bechthold, Elifcan Celebi, Dr. Marina Chugunova, Dr. Luisa Stella de Oliveira Coutinho Silva, Svenja Friess, Ph.D. Giorgia Gastaldon, Dr. Lisa Hanstein, Dr. Philine Helas, Prof. Karin Hoisl, Ph.D. Michael E. Rose, Esra Sarioglu, Isabel Valera and Dr. Ulla Weber.
This chapter considers the theoretical developments and the empirical evidence around ageism, age discrimination, and age-related stereotypes in the workplace. It covers the economics of discrimination in hiring and against older workers in micro and meso contexts. Finally, it describes the work-related, physical, and cognitive consequences of age discrimination.
This paper analyses the ways in which ethnic identity and labour institutions shape favouritism and discrimination among workers. We conduct a lab experiment in the field with South African coal miners from various ethnic groups and with different trade union membership status. Our analysis suggests that union identity and ethnic identity are two social constructs that operate in a distinct and opposite fashion. Unionization acts as a factor of workers solidarity beyond the confine of union membership. Conversely, ethnicity operates as the linchpin through which discrimination among workers is infused not only between ethnic majority and minorities, but also within the majority group itself. We find that the widespread practice of subcontracting in the mining sector exacerbates ethnic discrimination among workers both between and within ethnic groupings.
Although numerous studies have focused on ethnic discrimination in the labour and housing market, carpooling has barely been covered. Yet carpooling is indispensable for many minorities because they face severe economic inequalities. In order to test for ethnic discrimination in the carpooling market, we conducted a correspondence test. We sent out requests from four fictitious user profiles to drivers listed on one of Germany's carpooling websites. All users were aged mid‐20s: two males and two females. One man and one woman had German names, the second set had Turkish names. In total, we collected 952 observations. Our field experiment reveals substantial discrimination by ethnicity and gender. Whereas women with German names are least likely to experience discrimination, men with Turkish names are the most likely to be discriminated against. At first sight, counterintuitively but in line with ethnic competition theory, users with Turkish names are not significantly more discriminated against by drivers on routes in East Germany where xenophobia is more prevalent compared with West Germany.
We study a model where an employer, trying to fill a vacancy, engages in optimal sequential search by drawing from two subpopulations of candidates who differ in their "discourse systems": during an interview, a minority candidate with a discourse system not shared with the employer conveys a noisier unbiased signal of ability than does a majority candidate. We show that, when the employer is "selective," minority candidates are underrepresented in the permanent workforce, fired at greater rates, and underrepresented among initial hires, even though the employer has no taste for discrimination and the populations are alike in their average ability. Furthermore, workplace diversity is increased if: (1) the cost of firing is reduced, (2) the cost of interviewing is increased, (3) the opportunity cost of leaving the position unfilled is increased, or (4) the prior probability that a candidate can perform the job is increased. Indeed, if the prior probability is sufficiently high, or the cost of firing sufficiently low, then minority candidates may be overrepresented in the permanent workforce.
Adults evaluate others based on their speech, yet little is known of the developmental trajectory by which accent attitudes are acquired. Here we investigate the development of American children's attitudes about Northern- and Southern-accented American English. Children in Illinois (the "North") and Tennessee (the "South") evaluated the social desirability, personality characteristics, and geographic origins of Northern- and Southern-accented individuals. Five- to 6-year-old children in Illinois preferred the Northern-accented speakers as potential friends, yet did not demonstrate knowledge of any stereotypes about the different groups; 5-6-year-old children in Tennessee did not show a preference towards either type of speaker. Nine- to 10-year-old children in both Illinois and Tennessee evaluated the Northern-accented individuals as sounding "smarter" and "in charge", and the Southern-accented individuals as sounding "nicer." Thus, older children endorse similar stereotypes to those observed in adulthood. These accent attitudes develop in parallel across children in different regions and reflect both positive and negative assessments of a child's own group.
We review theories of race discrimination in the labor market. Taste-based models can generate wage and unemployment duration differentials when combined with either random or directed search even when strong prejudice is not widespread, but no existing model explains the unemployment rate differential. Models of statistical discrimination based on differential observability of productivity across races can explain the pattern and magnitudes of wage differentials but do not address employment and unemployment. At their current state of development, models of statistical discrimination based on rational stereotypes have little empirical content. It is plausible that models combining elements of the search models with statistical discrimination could fit the data. We suggest possible avenues to be pursued and comment briefly on the implication of existing theory for public policy.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.
We briefly review the evolution of empirical work on discrimination. We discuss why traditional regression‐based approaches neither convincingly measure market discrimination nor disentangle the relative importance of animus versus statistical discrimination in explaining such discrimination as exists. We describe the development of modern correspondence studies. We argue that these studies have the promise to credibly identify the presence of discrimination if not its magnitude, can inform us about the underlying mechanism generating discrimination and can also point to avenues for new theoretical and empirical work on discrimination. We discuss two articles with exemplary applications of these new methods.
This study contributes to the economics literature that links physical characteristics to labour market outcomes, by investigating the influence of hair colour on women's own wages and also their spouse's wages. Using U.S. panel data, we find that blonde women receive large wage premiums.
In this paper we explore whether recruiters prefer applicants who are relatively strong in the skills in which the recruiters themselves excel. We analyze evidence from all entry exams to the Spanish Judiciary held between 2003 and 2007, where applicants are randomly assigned across evaluation committees. We find that applicants who excel in the same dimensions as recruiters are significantly more likely to be hired. Our findings have important strategic implications for both public and private sector recruitment practices.
In many dimensions the ability to assess knowledge depends critically on the observer's own knowledge of that dimension. Building on this feature, this paper offers both theoretical and empirical evidence showing that, in those tasks where multidisciplinary knowledge is required, evaluations exhibit a similar-to-me effect: candidates who excel in the same dimensions as the evaluator tend to be ranked relatively higher. It is also shown that, if races or genders differ in their distribution of ability, group discrimination will arise unless evaluators (i) are well informed about the extent of intergroup differences and (ii) they may condition their assessments on candidates' group belonging.