Discrimination, Statistical

BIBLIOGRAPHY

Received economic theory of the firm admits only one motive: profit maximization. The received theory of the market is that it is most efficient when operating unconstrained. If this is so, hiring policies should seek to assemble the most efficient workforce, disregarding irrelevant factors, which cause loss due to the failure to employ the most competent available workers. (This assumes choice between candidates; namely, some unemployment. But the minimal level, often called friction unemployment, should suffice.) Despite the received theory, discrimination is rampant in recruitment against women and members of minority groups (Bertrand and Mullainathan 2004). By the theory of the firm, entrepreneurs should prefer the more productive workers over the better qualified ones, but they do not. (Even ignorance suffices to insure this; see Altonji and Plerret 1997.) This is amenable to statistical tests that may be viewed as empirical tests of the received theory. Also, it signifies discrimination being a burden on the economy. It looks as if data easily refute the received theory. It is possible to save it from refutation by two different claims. First, the perpetrator of discrimination is not the employers but, say, the employee organizations (the incumbent hypothesis). If so, then the theory is not refuted because it is inapplicable, as the (employment) market is not quite free. Advocates of received theory will not admit this defense; they will not suggest that the theory is inapplicable, at least not in the West. If the theory is applicable and is correct, then discrimination is profitable after all, even if it is disguised (Sattinger 1998). If one does not take the theory on faith, it is necessary to learn how discrimination can be profitable.

This problem concerns both the theory of the firm and the theory of the market. First, how does employment discrimination raise profits? Second, does the market mechanism operate optimally without the aid of antidiscrimination legislation? Since, admittedly, employment discrimination is clearly detrimental to the economy, the question is: Will it disappear faster with or without the aid of legislation? By the received market-mechanism theory, legislation against employment discrimination is harmful. Evidence suggests that at times this is so and at times not (Loury 1981; Neumark and Stock 1999). Hence the received theory of the market needs adjustment to allow for some government intervention. (This is scarcely news, since even the most ardent defenders of received theory do not oppose systematic monetary interventions, not to mention fiscal ones.) Does the same hold for the received theory of the firm?

Possibly, discrimination has a cost, and it should be considered a part of entrepreneurs’ preference. Their preferences thus become a part of the theory of consumers’ preference (Swinton 1977). This is a serious deviation from received theory that takes the theory of consumers’ preferences as no more than a theoretical nicety, since the market tends to aggregate demand only (Agassi 1992). The option left for the effort to rescue the received theory of the firm, then, is to explain how discrimination incurs no loss.

One explanation holds that discrimination is economically advantageous because it pleases community leaders, especially in underdeveloped countries where it is socially obligatory. This explanation interests students of business public relations and voluntary organizations such as the United Nations Volunteers and others who attempt to improve matters. Economists tend to ignore this view, since obviously not all countries maintain free markets. This is regrettable, since market freedom is a matter of degree. Economists tend to declare deviations of the market from received economic theory insignificant, praise the market for the advantages of modern liberal society, and blame deviations from it for the disadvantages. According to a second explanation, discrimination is due to employers who cling to refuted views as prejudices. This explanation does not help received economic theory that deems the market mechanism the best means for eliminating inefficient entrepreneurs through competition. A third explanation holds that the advantage of discrimination is in its reduction of the cost of hiring, since members of the preferred groups keep their jobs for longer periods. Such advantages are called statistical discrimination. Not surprisingly, this view of discrimination is pivotal (Collinson et al. 1990), and so it deserves attention. Is the hypothesis true that discrimination is advantageous?

Before examining the hypothesis we should ask the following questions: (1) Does it rescue the theory? (2) Does it suggest that the market mechanism still is the best way to overcome discrimination through recruitment? (3) Does it suggest leaving matters to the market mechanism anyway?

The theory allows for statistical discrimination only if it is cost effective. If so, then reducing the influence of discrimination on workers’ turnover should reduce it. Even then, allowing for statistical discrimination cannot be the best strategy, because reducing it by legislation is an improvement, even from the strictly economic viewpoint. Defenders of the received theory are familiar with such criticism, and do not deny it: certain legislations may outdo the market mechanism, but these are too costly and their advantages are temporary at best. In the long run, the market mechanism is the most efficient means for tackling social evils. This response is a reading of received theory as promising social benefits only in the long run.

To this view, John Maynard Keynes (1883–1946) responded with his memorable adage: in the long run we are all dead. This means that we cannot test this version in a lifetime. Moreover, the evils we are facing are pressing, and the promises of received theory are postponed and doubtful. This is still under debate, however, especially since statistical discrimination is less pressing than the massive unemployment that troubled Keynes. The central question then is: Is the statistical discrimination hypothesis true? We do not know. Is it at least testable? If not, should we give the market mechanism the benefit of the doubt or should we fight discrimination or wait for the market to do it? Waiting is questionable, since pockets of poverty and related ills are stagnant, and, notoriously, many social factors hardly benefit from the economy, which adds to the stagnation. Instances for this are understandable but intolerably self-reinforcing: the disposition of the police to seek criminals within the groups in which crime prevails, persistent poor education, underage pregnancy, and the absence of intergenerational mobility where it is needed most. The self-reinforcing of socioeconomic patterns ensures social stability at the cost of retaining various social evils, employment discrimination included.

The opinion that the market mechanism is the best means for the eradication of social ills is questionable. Opposing laws against discrimination requires particularly strong arguments. Arguing that such legislation is ineffective will not do: many laws are ineffective yet we still want them on the books. The paradigm case is the proscription of statutory rape by minors. The argument against the proscription of discrimination is therefore understandably philosophical. In principle, advocates of the market mechanism theory approve only of laws that are essential for the maintenance of the market. Thus, they approve of imposing the honoring of contracts, but not truth in advertising: the market can live with false advertisements and the market mechanism will eliminate them, they say. Even if we accept this kind of argument, advocates of the market mechanism theory cannot show that the market tolerates no social ills that are too objectionable. As discrimination is both objectionable and a burden on the market, and since laws against it are no burden on any social group, the theory of statistical discrimination is a priori too feeble—apart from its dependence on many questionable assumptions. Which is not to say that we need not test these assumptions: they may be interesting for other ends, especially as means of insight into the costs—economic and other—of social stability. As Karl Popper (1902–1994) suggested more than half a century ago, however advisable it is to rely on the market mechanism when it works, the view that it always works well is a metaphysical dogma. This squares with his proposal to replace the demand for social stability with the demand for democratic controls.

SEE ALSO Discrimination; Discrimination, Racial; Discrimination, Taste for; Discrimination, Wage; Discrimination, Wage, by Age; Discrimination, Wage, by Gender; Discrimination, Wage, by Occupation; Discrimination, Wage, by Race; Distribution, Normal; Expectations; Information, Asymmetric; Probability Distributions; Risk; Uncertainty

BIBLIOGRAPHY

Agassi, Joseph. 1992. Beyond the Static Theory of Tastes as Exogenous. Methodology and Science 25: 99–118.

Agassi, Judith Buber. 1992. Review of David L. Collinson, David Knights, and Margaret Collinson: Managing to Discriminate. Organization Studies 13 (3): 472–475.

Altonji, Joseph G., and Charles R. Plerret. 1997. Employer Learning and Statistical Discrimination. U.S. Department of Labor. http://www.bls.gov/ore/pdf/nl970020.pdf.

Arrow, Kenneth. 1973. The Theory of Discrimination. In Discrimination in Labor Markets, eds. Orley Ashenfelter and Albert Rhees, 3–33. Princeton, NJ: Princeton University Press.

Bertrand, Marianne, and Sendhil Mullainathan. 2004. Are Emily and Greg More Employable than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination. American Economic Review 94: 991–1013

Collinson, David L., David Knights, and Margaret Collinson. 1990. Managing to Discriminate. London and New York: Routledge.

Darity, William A., and Patrick L. Mason, 1998. Evidence on Discrimination in Employment: Codes of Color, Codes of Gender. The Journal of Economic Perspectives 12: 63–90.

Loury, Glenn C. 1981. Is Equal Opportunity Enough? American Economic Review 71: 122–126.

Neumark, David, and Wendy A. Stock. 1999. Age Discrimination Laws and Labor Market Efficiency. Journal of Political Economy 107: 1081–1125.

Popper, Karl R. 1945. The Open Society and Its Enemies. London: Routledge.

Sattinger, Michael. 1998. Statistical Discrimination with Employment Criteria. International Economic Review 39: 205–237.

Swinton, David H. 1977. Racial Discrimination: A Labor Force Competition Theory of Discrimination in the Labor Market. American Economic Review 67: 400–404.

Joseph Agassi