Discrimination

In part, the confusion about the effectiveness of market penalties for discrimination results from confusion about the meaning of the term discrimination. Although most people abhor discrimination on the basis of characteristics such as race and sex, they generally applaud those who discriminate against the lazy, the dishonest, and the unproductive by paying them lower wages, firing them, or refusing to hire them in the first place. The problem thus becomes one of distinguishing between unfair discrimination based on simple prejudice, which many people wish to prohibit, and productive discrimination based on merit, which most people wish to encourage. To understand how the market penalizes unfair discrimination while rewarding the productive kind, and why government often indiscriminately punishes both, one must first understand how markets constrain people's behavior.

To begin with, a business can afford to hire an additional employee only if the additional output made possible by hiring him sells for a price that equals or exceeds his wages. The employee's addition to output depends on the tools that he has to work with and on his ability. The price that the additional output sells for depends upon how much consumers are willing to pay. Suppose, for example, that a concessionaire at a baseball stadium decides to hire another hot-dog seller at a wage of $5.00 per hour. If hot dogs sell for $1.00 and cost $.90 to produce (excluding the seller's wage), the new salesman must sell at least fifty hot dogs an hour to cover the cost of paying him. Because the concessionaire keeps any revenues from hot-dog sales in excess of his costs, he naturally prefers employees who can sell seventy hot dogs per hour to those who sell only the minimum fifty.

Employers wishing to maximize their profits attempt to hire people with the highest possible expected productivity. Suppose that the concessionaire notices that his highest producers have little body fat and wear road race T-shirts advertising marathons and ten-kilometer runs. Suppose also that he has noticed that the majority of those who have failed to meet the standard of fifty hot dogs per hour were overweight smokers. Faced with a choice between a smoker and a marathon runner, a rational employer would hire the runner.

Of course, an exceptionally motivated smoker might outperform the average runner. Unfortunately, without requiring extensive physical examinations (the cost of which could wipe out all the profits from hot-dog sales and make any hiring moot), the manager cannot separate exceptional smokers from ordinary ones. So he must base his decision on his experience with smokers in general. As a result most of his employees will be lean and fit. An observer ignorant of the correlation between physical conditioning and productivity would condemn him for unfairly discriminating against overweight tobacco lovers.

In this example the correlation between personal habits and productivity makes it impossible for an outside observer to judge whether the employer discriminated fairly (for a bona fide business reason) or simply has a personal bias against smokers. In either case, because physical appearance and personal habits are correlated with productivity, overweight smokers will be underrepresented in the ranks of hot-dog salesmen. The employer's work force will look the same whether the manager discriminated fairly on the basis of real differences in productivity (no smoker will ever cover as much area as a marathon runner), fairly on the basis of incomplete information (the smoker was exceptionally fit but the manager did not know it), or unfairly on the basis of managerial taste (lean men are better looking than fat ones).

Correlations between personal characteristics and productivity abound in the real world, and their abundance makes it virtually impossible for outside observers to separate productive discrimination from unfair discrimination. On average, blacks with high school diplomas score much lower on achievement tests than whites or Hispanics. On average, recent immigrants have a worse command of English than natives. On average, women take more respites from continuous employment than men. Does the "underrepresentation" of blacks on college campuses stem from unfair discrimination or from their lower average academic achievement? Does the "underrepresentation" of recent immigrants in higher-paying jobs stem from bigotry or from the fact that in the United States fluent English substantially reduces the cost of communicating? Are women "overrepresented" in particular occupations because they are discriminated against or because women choose occupations that give them more flexibility to raise a family? An outsider cannot answer these questions.

The statistical measures of discrimination used by government agencies like the Equal Employment Opportunity Commission depend upon easily measurable individual characteristics, such as race, sex, and age. Although characteristics such as commitment, cooperativeness, motivation, and trustworthiness make large contributions to individual productivity, these same characteristics defy accurate quantitative measurement. For this reason they are often left out of the equation altogether. Evidence on the distortion that this causes suggests that the resulting estimates of discrimination are too unreliable for policy purposes. One study, for example, found that wages for women were either 61 percent lower or 19 percent higher than those of comparable men, depending on how one controlled for unobservable characteristics.

Leaving out the unobservables and basing government-imposed hiring guidelines on easily measured characteristics has the effect of basing hiring and promotion criteria on ethnic background, religion, and sex instead of merit. This punishes those innocent of discrimination along with the guilty and exacerbates social friction between favored and unfavored groups. To the extent that it substitutes less competent people for more competent ones and encourages defensive hiring, it also wastes resources and lowers the average standard of living. In extreme cases, hiring quotas based on caste membership in India and on Sinhalese extraction in Sri Lanka have provoked civil wars.

In contrast, the market mechanism penalizes only those who discriminate unfairly. Intolerant employers find their profits reduced, and bigoted customers must pay more for their shabby tastes. Suppose that black and white hot-dog salesmen are equally productive, but that the concessionaires at all of the stadiums want to hire whites rather than blacks. The stiffer competition for whites will force employers to pay more for a white worker than for an equally productive black one. In effect, employers insisting on white workers make themselves higher-cost producers. Unless customers are willing to pay more for a hot dog delivered by a white than a black, higher costs mean smaller profits. Concessionaires interested in maximum profit will hire blacks, make more money, and be able to underprice the bigots. Even if the white concessionaires collude in refusing to hire blacks, they could still be undercut by new black firms exploiting their lower operating costs. Competition will ultimately force a firm to hire people of either color unless the owner accepts a cut in his profits. As the more profitable equal-opportunity employers expand, the demand for black workers will increase and black wages will rise. Since a few owners may be willing and able to pay for their desire to discriminate against blacks, competition does not necessarily bankrupt all firms practicing unfair discrimination. But competition does make unfair discrimination expensive, thus ensuring that less will occur.

In South Africa in the early 1900s, for example, mine owners seeking profits laid off higher-priced white workers and hired lower-priced black workers, even in the face of penalties threatened by government and violence threatened by white workers. Only by lobbying for, and getting, extreme restriction on blacks' ability to work were the whites able to reserve higher-paying jobs for themselves (see Apartheid). The profit motive also ameliorated the discrimination of the McCarthy era: profit-maximizing producers, in defiance of the Motion Picture Academy's blacklist, secretly hired blacklisted screenwriters.

Government, on the other hand, remains unconstrained by any considerations of profit and incurs no costs for discriminating on the basis of race or other factors so long as the discrimination is politically acceptable, which it often has been. At the turn of the century, blacks, who had been making progress since the Civil War, began to compete for previously all-white jobs. Racial animosity increased. Voting power was in the hands of whites and, as economist Thomas Sowell points out, civil service hiring rules were amended to require a photograph of the applicant and to allow the hiring official to choose between the three top performers on civil service tests. The number of blacks in federal employment plummeted. It remained low until the political repercussions of the civil rights movement resulted in affirmative action sixty years later. Although affirmative action changed the group discriminated against, the government continues to discriminate, now against white men. In contrast, private businesses in the South were less eager to discriminate against blacks even at the height of segregation (see sidebar).

Although the market makes people pay for unproductive discrimination, many of the restrictions that government imposes on markets blunt the market mechanisms that make discrimination expensive. Barriers to hiring and firing make employers less likely to try out types of people with whom they have little experience. Minimum wage laws and union wage scales, by keeping wages higher than market wages, reduce the number of people that employers wish to hire while simultaneously attracting more applicants. With so many people to choose from, the cost of turning away applicants who meet the employer's productive requirements, but not his tastes, drops considerably.