Model Of Statistical Discrimination Research Articles

Biological Gender Differences, Absenteeism, and the Earnings Gap

In most countries, women are absent from work more frequently than men. Using personnel data, we find that the absences of women below the age of 45 follow a 28-day cycle, while the absences of men and of women over the age of 45 do not. We interpret this as evidence that the menstrual cycle increases female absenteeism. To investigate the effect on women's earnings, we use a simple model of statistical discrimination. Consistent with the model, we find absenteeism has a more negative effect on men's earnings and this difference declines with seniority. The increased absenteeism induced by the 28-day cycle explains at least 14 percent of the earnings gender differential. (JEL J16, J22, J31)

Education and Labor Market Discrimination

Using a model of statistical discrimination and educational sorting, we explain why blacks get more education than whites of similar cognitive ability, and we explore how the Armed Forces Qualification Test (AFQT), wages, and education are related. The model suggests that one should control for both AFQT and education when comparing the earnings of blacks and whites, in which case a substantial black-white wage differential emerges. We reject the hypothesis that differences in school quality between blacks and whites explain the wage and education differentials. Our findings support the view that some of the black-white wage differential reflects the operation of the labor market. (JEL I21, J15, J24, J31, J71)

Distribution Variability, Within-Group Competition and Statistical Discrimination

This paper develops a statistical discrimination model concerning the second order moment. It is shown that with heterogeneous individuals within a group, this form of statistical discrimination has differential impact on individuals. Specifically a belief that one group has a smaller variability may be self-fulfilling and persist without policy intervention, so two groups are identical ex ante but different ex post. The history-dependent belief offers an explanation why there may be smaller variability in observed qualifications among female, even though its underlying distribution of innate ability may be the same as male.

Towards a unifying heat transfer correlation for the entire boiling curve

The mechanistic understanding of boiling processes is still inadequate. Major physical effects determining the heat transfer in high heat flux nucleate and transition boiling regions have not yet been captured adequately. Thus, existing design correlations are often valid only for one of the boiling regimes. In this paper, the wetting structure close to the boiling surface is identified using the experimental data from an optical probe, obtained during pool boiling of FC-72 on a horizontal surface, together with a mathematical model for the interfacial geometry based on two-phase flow averaging theory. In the same framework, a unifying correlation to describe the heat flux along the entire boiling curve is presented. The suggested correlation is based on the same physical quantities regardless of the boiling regime; it employs only a single fitting parameter in its most simple form. Alternative correlations are compared to the suggested correlation and their relative merit is assessed by statistical model discrimination techniques. The results suggest that transfer phenomena associated with the interfacial evolution, in particular the volumetric presence of interface close to the heater surface, together with the superheat, play an important role for the overall boiling heat transfer mechanism.

Information and Racial Exclusion

This paper presents several economic models that explore the relationships between imperfect information, racial income disparities, and segregation. The use of race as a signal arises here, as in models of statistical discrimination, from imperfect information about the return to transactions with particular agents. In a search framework, signaling supports not simply a discriminatory equilibrium, but a pattern of racially segregated transactions, which in turn perpetuates the informational asymmetries. Minority groups necessarily suffer disproportionately from segregation, since the degree to which transactions opportunities are curtailed depends upon group size, as well as the informational distance between racial groups. However, in some variants of the model, minority agents will self-segregate since they face an adverse selection of majority agents who are willing to trade with them. We also show that, if agents are able to learn from transactions, racial signaling can emerge with only minimal assumptions about the ex ante importance of race.

Uncertainty, Hiring, and Subsequent Performance: The NFL Draft

In this article, we analyze the impact of uncertainty on the hiring process. We show the connection between models of statistical discrimination where uncertainty can work against groups that have less reliable indicators of future productivity and models of option value where uncertainty about future productivity can be beneficial for these groups. These models generate hypotheses about the relationship between ex ante hiring patterns and ex post productivity. This is applied to the market for NFL football players. We provide various estimates of NFL success, which suggest that statistical discrimination and option value influence choices in this market.

A general equilibrium model of statistical discrimination

We study a general equilibrium model with endogenous human capital formation in which ex ante identical groups may be treated asymmetrically in equilibrium. The interaction between an informational externality and general equilibrium effects creates incentives for groups to specialize. Discrimination may arise even if the corresponding model with a single group has a unique equilibrium. The dominant group gains from discrimination, rationalizing why a majority may be reluctant to eliminate discrimination. The model is also consistent with “reverse discrimination” as a remedy against discrimination since it may be necessary to decrease the welfare of the dominant group to achieve parity.

The Effect Of Statistical Discrimination On Black–White Wage Inequality: Estimating A Model With Multiple Equilibria*

This article presents the structural estimation of the parameters of a statistical discrimination model. Although the model is capable of displaying multiple equilibria, an estimation strategy that identifies both the model parameters and the equilibrium selected by the economic agents is developed and empirically implemented. A comparison between the selected equilibria and the other potential equilibria reveals that the decline in wage inequality experienced in the U.S. economy cannot be attributed to changes in the equilibrium selection. Nonetheless, a counterfactual experiment shows that in a color‐blind society blacks' wage would have been on average more than 20% higher.

Uma resenha teórica e empírica sobre economia da discriminação

O artigo faz uma resenha dos principais modelos de discriminação e examina as possíveis fontes de discriminação racial por sexo no mercado de trabalho. Do ponto de vista teórico, o modelo de discriminação por preferência mostra que o comportamento discriminatório do empregador está inversamente relacionado aos lucros das firmas. Os modelos de discriminação estatística partem do pressuposto de que o credor ou empregador não tem informação completa sobre os indivíduos. Estes modelos utilizam as características dos grupos que sofrem discriminação, como raça ou seu sexo, para reduzir o valor do crédito ou do salário.

Uncertainty, Hiring and Subsequent Performance: The NFL Draft

In this paper we analyze the impact of uncertainty on the hiring process. We show the connection between models of statistical discrimination in which uncertainty can work against groups who have less reliable indicators of future productivity and models of option value where uncertainty about future productivity can be beneficial for these groups. We show that these models generate testable hypotheses about the relationship between ex ante hiring patterns and ex post productivity. Testing these hypotheses requires a special data set, one that permits accurate indicators of preemployment valuations as well as post-employment productivity. The data must also not be confounded by a matching problem where the highest valued candidates are matched with jobs where they have the highest probability of succeeding. We chose to use data from the market for professional football players. This market is special because abundant data exist on the preemployment evaluation through the observed draft choice ordering. There are good data on ex post realized productivity as the performance of each player is well documented. In addition, matching occurs through a preset process that does not allow firms with the most resources to gather the highest ability applicants. We provide various estimates of an array of different measures of NFL success, which all suggest that both statistical discrimination and option value influence choices in this market.

Networks, culture, transaction costs and discrimination

Starts out with a survey of various formal theories that have focused on discrimination in the labor market. Argues that Becker’s traditional taste for discrimination model, the various statistical discrimination models and the new cultural communication cost models ultimately yield analytically and observationally equivalent predictions. In particular, these models all imply that we may find occupational segregation across firms. This, in turn, suggests that it is not easy to identify the true causes of discriminatory wage differentials in the labor market and thus that we may have a very hard time sorting out which of these models applies best. Finally, speculates, in the context of Kremer’s model of economic growth, about how changing technologies and structure of production could possibly exacerbate the inequalities predicted by these models of discrimination in the labor market.

On the Economics of Discrimination in Credit Markets

This paper develops a general equilibrium model of both taste-based and statistical discrimination in credit markets. We find that both types of discrimination have similar predictions for intergroup differences in loan terms. The commonly held view has been that if there exists taste-based discrimination, loans approved to minority borrowers would have higher expected profitability than to majorities with comparable credit background. We show that the validity of this profitability view depends crucially on how expected loan profitability is measured. We also show that there must exist taste-based discrimination if loans to minority borrowers have higher expected rate of return or lower expected rate of default loss than to majorities with the same exogenous characteristics at the time of loan origination. Empirical evidence on expected rate of default loss cannot reject the null hypothesis of non-existence of taste-based discrimination.

On the Economics of Discrimination in Credit Markets

This paper develops a general equilibrium model of both taste-based and statistical discrimination in credit markets. We find that both types of discrimination have similar predictions for intergroup differences in loan terms. The commonly held view has been that if there exists taste-based discrimination, loans approved to minority borrowers would have higher expected profitability than to majorities with comparable credit background. We show that the validity of this profitability view depends crucially on how expected loan profitability is measured. We also show that there must exist taste-based discrimination if loans to minority borrowers have higher expected rate of return or lower expected rate of default loss than to majorities with the same exogenous characteristics at the time of loan origination. Empirical evidence on expected rate of default loss cannot reject the null hypothesis of non-existence of taste-based discrimination.

A General Equilibrium Model of Statistical Discrimination

We consider a general equilibrium model with endogenous human capital formation in which ex ante identical groups may be treated differently in equilibrium due to informational externalities. Unlike earlier models of statistical discrimination, group inequalities may arise even if the corresponding model with a single group has a unique equilibrium. The dominant group gains from discrimination, rationalizing why a majority may be reluctant to eliminate discrimination. The model is also consistent with reverse as a remedy against discrimination since it may require to decrease the welfare of the dominant group to achieve parity.

The Effect of Statistical Discrimination on Black-White Wage Inequality: Estimating a Model with Multiple Equilibria

This paper presents the structural estimation of a statistical discrimination model where wage differentials can be generated by self-fulfilling expectations about group productivities. Although the model is capable of displaying multiple equilibria, an estimation strategy that identifies both the parameters of the model and the equilibrium chosen by the economic agents is developed and empirically implemented. The model is estimated three times using U.S. black and white male wage data from 1964 to 1996. In addition to recovering the selected equilibrium, all of the equilibria that the model could have displayed under the estimated set of fundamentals are computed. A comparison between the equilibria that were selected in the economy over time and the other potential equilibria reveals that the outcome with essentially the lowest wage inequality have been selected whenever the estimated parameters are consistent with multiple equilibria. This implies that self-fulfilling expectations did not exacerbate wage differentials in the U.S. and that the decline in wage inequality experienced in the U.S. economy in the last thirty years cannot be attributed to changes in the equilibrium selection. A simulation is also performed to compute equilibrium wages in a color blind society where employers cannot distinguish their employees for their ethnicity. Results show that black males' average wage would be 24.5% higher and whites' wage would be 2.2% lower.

Race and Ethnicity Assessment in Baseball Card Markets

There is a growing literature investigating fan discrimination revealed in markets for sports memorabilia. Such estimates miss the mark for two reasons. First, simply measuring race as a binary variable may be less insightful than an alternative measure. Second, although it is the race perceptions of market participants that matter, past studies rely exclusively on researcher race/ethnicity assessment. The authors provide an estimation framework using continuous, market participant reports of race/ethnicity. The results show that past race/ethnicity results in card markets suffer from errors in variables and that continuous, market participant reports of race/ethnicity provide (carefully defined) "better" estimation results. The authors find that race/ethnicity matters and in a way is consistent with a model of statistical discrimination by those baseball fans who trade in the card market. The discrimination is against Black and Hispanic hitters and Black pitchers but not against Hispanic pitchers.

Wage Differentials by Race and Sex: The Roles of Taste Discrimination and Labor Market Information

Using a unique dataset, this article first documents that gaps in starting wages by race and sex persist after accounting for performance on the job. Evidence suggests that simple statistical discrimination, and not just taste discrimination, is partly responsible for race differences in starting wages. But because women's average performance in the sample is higher than men's, simple statistical discrimination cannot explain the sex gap. In more complex models of statistical discrimination, worse information about a group can lower its average wage. Estimates of the quality of labor market information indicate that this may explain women's lower starting wages.

Race, Information, and Segregation

This paper presents several related economic models that explore the relationships between imperfect information, racial income disparities, and segregation. The use of race as a signal arises here, as in models of statistical discrimination, from imperfect information about the return to transactions with particular agents. If agents are able to learn from transactions, racial signaling can emerge with only minimal assumptions about the ex ante importance of race. In a search framework, this signaling supports not simply a discriminatory equilibrium, but a pattern of racially segregated transactions, which in turn perpetuates the informational asymmetries. Minority groups necessarily suffer disproportionately from segregation, since the degree to which transactions opportunities are curtailed depends upon group size. However, in some variants of the model, endogenous segregation will be mutual, since minority agents face an adverse selection of majority agents who are willing to trade with them.

Self-fulfilling beliefs about social status

Human beings care a lot about being viewed as `smart'. This paper defines social status as the public beliefs about one's `smartness' inferred from one's publicly observed social mobility experience. This status motive can easily generate multiple equilibria based upon self-fulfilling beliefs. It also tends to amplify the inequality of economic success between agents with different social origins. This theory of persistent inequality is compared with other theories of inequality based upon self-fulfilling beliefs, such as the model of statistical discrimination and the model of learning about one's chances of mobility in the absence of a status motive, as well the sociological theories of Merton, Boudon and Bourdieu.

Statistical discrimination in the labour market: an efficient vice?

On the presumption that more information improves economic decision-making, it is often claimed that statistical discrimination reduces inefficiencies. Outlawing discrimination then means foregone efficiency gains. This paper examines whether there is indeed an uncomfortable trade-off between equality and efficiency, by reconsidering some of the most significant statistical-discrimination models. The conclusion is that there is no firm basis for supposing that statistical discrimination is efficient. As well as disputing the efficiency claim, the paper casts further doubts on the common interpretation of the decomposition of empirical wage differentials.