The Economics of Discrimination
By Robert P. Murphy
“Remember they weren’t called Jim Crow ‘best practices,’ but were Jim Crow laws.”
Before exploring the economics of discrimination, we first need to distinguish the term from several related ones. For example, racism, bigotry, and prejudice refer to someone’s beliefs; they are mental phenomena. In contrast, discrimination refers to an action. The two often go hand-in-hand. For example, a bigoted employer might harbor prejudice against Muslims, and so he discriminates against them when hiring staff for his firm.
Most Americans think there is a prima facie case for outlawing some kinds of business discrimination, but do not want to criminalize racism or prejudice itself. Most people have no problem imposing legal penalties on outward actions that harm minorities, but, at the same time, would recognize the Orwellian nature of actually punishing people for forbidden thoughts.
What People Have in Mind When Opposing “Discrimination”
We can push the analysis further: Most people don’t actually oppose all forms of business discrimination.
A familiar example is the disproportionate number of black players in the NBA. Hardly anyone seriously believes that the owners of professional basketball teams are systematically biased against whites and Asians. The mere existence of a disparate outcome is not proof of the type of discrimination that most people oppose. So long as the “under-representation” of a particular group can be correlated with other factors—so that the person’s membership in the group is not the cause of the under-representation—the outcome does not qualify as the type of discrimination that most people want to criminalize.
In some cases, however, employers really do discriminate based on the outward traits defining membership in a group, and yet most people would shrug off even these examples as benign. For example, movie stars and pop musicians, especially females, tend to be thin and very good-looking. Just as white people and Asians tend to be under-represented in the NBA, we could say that the obese are under-represented in Hollywood. Yet, unlike the NBA outcome—where other factors are at work that are merely correlated with skin color—in the case of Hollywood we have direct causation: Casting directors would reject obese actors and actresses for many roles simply because of their obesity, even if they had strong acting skills.
To drive home the point, consider an even more fanciful example. The film The Hours required someone to play Virginia Woolf. Nicole Kidman landed the part. But suppose that Dustin Hoffman had tried reading for it. Even though he can play a woman—as he proved in Tootsie—the director obviously would have rejected Hoffman for the role because he was a man. In other words, Hoffman would have been denied a job opening because of his sex, a clear-cut case of discrimination.
For a different example, suppose that Will Smith had read for the part of the young Obi-Wan Kenobi in Star Wars, Episode I, a part that in fact went to Ewan McGregor. Will Smith probably would have been rejected from the job because he was black. Alec Guinness had already played the older Kenobi in the original trilogy, and the producers wouldn’t have wanted a needless distraction from the release of the eagerly anticipated movie by provoking a national discussion on race in film.
Although some purists might claim that even the above examples are unfair discrimination, most Americans would dismiss them as harmless. Further, most Americans don’t mind that the employees greeting customers at restaurants or staffing trendy clothing stores in the mall tend to be much better looking than the employees working in the kitchen at restaurants or cleaning the bathrooms in the mall. Again, this outcome isn’t a coincidence: the employees interacting with customers are selected, in part, because of their good looks.
Those who support laws against employer-based discrimination might scoff at the above examples as silly, but it’s important to think through why they don’t strike the average person as examples of social injustice. Once we put our finger on exactly what makes some types of discrimination objectionable—and ostensibly worthy of criminal punishment—we will see that the free market already has built-in penalties for it.
The primary reason these examples do not strike most people as illegitimate forms of discrimination is that the preference for a female actress (to play Virginia Woolf), or a white actor (to play a young Ben Kenobi), or an attractive female (to be the hostess at the front of the restaurant) were all related to job performance. In other words, the director of Star Wars didn’t “have something against blacks”; his decision to not cast Will Smith as Obi-Wan Kenobi “because he was black” would simply reflect the obvious fact that the audience would find it distracting. In the director’s view, casting Will Smith as Obi-Wan would have diminished the quality of the final product.
In the same way, the owner of a trendy clothing store in the mall will tend to hire attractive employees who dress well because they are likely to move more merchandise. Most people understand that this is a straightforward business decision and don’t hold it against the owner of the store.
Our analysis leads us to conclude that people who oppose “discrimination” really have in mind employers who treat some job applicants differently based on irrelevant characteristics. For example, if an accounting agency didn’t hire a CPA because he was black, then most people would say this definitely does qualify as “bad” discrimination that should be punished by law. The difference between this case and our fanciful example of Will Smith reading for the role of Obi-Wan is that there’s no good reason a black person can’t be an accountant. If an employer thinks so, then he is acting on a baseless prejudice or stereotype, and that’s just the type of practice that anti-discrimination laws are designed to eliminate.
The Free Market Precisely Punished “Bad” Discrimination
The previous section laid out exactly what (most) people have in mind when they say that the government should punish employment discrimination. Specifically, we saw that if an employer harbors negative views about a job applicant because of his or her membership in some group, and that membership has nothing intrinsically to do with job performance, then to deny the applicant a job constitutes “bad” discrimination.
Yet, in a free market, it is precisely this kind of discrimination that is swiftly and automatically penalized. If an employer discriminates against a job applicant on the basis of factors that are truly irrelevant to job performance, then the employer necessarily incurs a financial penalty. Even better, the penalty is directly proportional to how much the employer’s decision was based on prejudice, rather than on merit.
Consider a numerical example. Suppose that a restaurant owner needs a new busboy, a position that pays $8 per hour. One applicant is clearly qualified and has had previous experience. The owner reckons that by hiring the young man, his turnover of customers will be increased such that his revenues, net of other expenses, will rise by $10 per hour. In other words, hiring this stranger will add $2 per hour to the restaurant’s bottom line.
However, the restaurant owner’s wife complains that her shiftless nephew is once again unemployed. She urges her husband to hire him as a busboy. The owner reckons that the nephew will not clean tables as quickly as the other applicant and is more likely to drop dishes. All in all, hiring the nephew at the same pay will add only 50 cents per hour to the restaurant’s bottom line.
In a free market, the owner has the legal right to hire whomever he wants. After all, it’s his restaurant, and the wages he pays are (initially) his property. But if he allows nepotism to influence his hiring decision, he suffers an implicit fine of $1.50 per hour. He makes less money if he makes employment decisions based on factors, such as family relationships, that have nothing to do with the job.
Let’s change the example. Suppose that instead of looking for a new busboy, the owner has to find a new chef, a position that pays $80,000 per year. He interviews one qualified applicant with years of experience and figures the man would add $85,000 in revenue over the course of a year. In contrast, if the owner hired his wife’s nephew as a chef, his revenue would increase by only $10,000 per year. In this case, a nepotistic decision would cost the owner $75,000 per year in foregone revenues.
We see that in a free market, employers suffer automatic financial penalties when they engage in discriminatory practices that qualify as “unfair” in the popular mind. It’s true that employers are still free to discriminate, even in the pejorative sense, but they are “fined” in direct proportion to the severity of their arbitrariness. As David R. Henderson points out, critics of the free market can’t have it both ways: Are we to condemn businesspeople as ruthless profiteers, who sacrifice all else to profit? Or are we to condemn them as bigots who won’t hire the best applicant for a job opening? The critics must decide because these complaints are mutually exclusive.3 Notice that the penalty from bad discrimination does not prevent that discrimination: it just discourages it. Employers who are willing to pay the penalty can still discriminate. However, they will find themselves losing market share to employers who don’t discriminate.
What If the Customers Are Bigots?
A critic might retort that in a sufficiently bigoted society, business owners wouldn’t be penalized for discriminatory practices. For example, suppose that a large number of white consumers in a certain town prefer to eat at restaurants that hire only white staff and serve only white customers. Imagine that their preference is so strong that these racist consumers are willing to pay higher prices for “the same” food, just so long as no black people are in the restaurant with them.
In this case, it’s true that the owner of at least one restaurant in our hypothetical town would not suffer financial penalties for instituting segregation on his private property. The principle would be analogous to today’s upscale restaurants that require all men in the building—including customers—to wear a coat and tie. That is “discrimination” too, and, for it to work, enough customers must be willing to pay for the policy of excluding potential customers who don’t dress formally for dinner. I am not claiming that the customers’ underlying preferences—for a whites-only environment in the one case, versus a coat-and-tie-only environment in the other—are morally equivalent. Instead, I am dispassionately analyzing a “worst-case scenario” to show how market forces work.
In the case of a segregated restaurant in the free market, the penalty now is not on the employer but on the discriminating customers. By hypothesis, the only way the owner evades financial loss from discriminatory practices is if his customers, in turn, are willing to pay higher prices. This means that the bigoted whites in our hypothetical community are paying more for eating out (in whites-only restaurants) than their colorblind neighbors, who are happy to patronize restaurants with black employees and customers. Again, the free market doesn’t prohibit people—whether as employers or consumers—from acting on their prejudices, but it does make them pay for it.
Is Government the Answer?
Those who support laws against discrimination might reject the above arguments as hypothetical and risky. They might think that there are so many bigots in society that the financial penalties of a free market would still allow systematic and unfair discrimination. Rather than trusting the profit motive to stamp out the practice, therefore, they want government to ensure neutrality in business decisions.
For more on how the market resists discrimination, see Discrimination, by Linda Gorman in the Concise Encyclopedia of Economics. The boxed material discusses how streetcar companies resisted discrimination.
This world view rests on a basic contradiction: If society is filled with so many racists and sexists that minorities and women don’t stand a chance in the free market, then why in the world would they want to empower a government elected by majority rule? In the market, at least, the bigoted employer faces immediate and personalized penalties when he acts on his views. But when he votes for politicians who institute discriminatory policies, the harms are passed on to millions of others.
This is not a mere hypothetical. The most shocking abuses of minorities and women in history occurred under the regimes of tyrannical governments. No matter how anti-Semitic the business leaders in a community might be, they would never have instituted the racial policies of the Nazis because those policies would have been too unprofitable.
It’s also important to mention that in the case of segregated lunch counters—the issue at the heart of the Rand Paul controversy—this was not a purely free-market outcome. Remember they weren’t called Jim Crow “best practices,” but were Jim Crow laws. We can never know for sure what the racial policies of Southern businesses would have been in the 1950s in a genuine laissez-faire environment. What we do know is that discriminatory business owners would have more acutely felt the costs of their policies without government mandates enforcing them on the whole industry.
The typical discussion of discrimination takes it for granted that the practice is odious and should be penalized. Yet, closer inspection reveals that most people have no problem with discrimination that caters to their preferences. When we specify precisely what we mean by “bad” discrimination, we find that the free market offers automatic and swift financial penalties in proportion to the severity of the practice.
For example, see Krissah Thompson and Dan Balz, “Rand Paul comments about civil rights stir controversy”. The Washington Post, May 21, 2010.
Gary Becker, The Economics of Discrimination. (Chicago: University of Chicago Press, 1957).