Discrimination and Disparities

by Thomas Sowell

  According to the Chicago Tribune, the resistance of working-class and middle-class blacks “in some cases has been fierce.” Black homeowners have “protested, loudly” at public meetings that they “didn’t want ‘those people’ moving back into their rejuvenated neighborhood.” Often homeowners at public gatherings “would shout at officials that they’d worked hard to get where they were and that they didn’t want to live next door to people who would just tear up their homes. They called them ‘project people,’ ‘lowlifers’ and ‘freeloaders.’”70

  “Some blacks feel that ‘those people’ make it tough on those of us trying to make something of ourselves,” says Shirley Newsome, a homeowner in Kenwood-Oakland and a longtime voice of moderation. “That’s why white America doesn’t want me living next to them, because they look at me and figure I’m from a place like public housing.”71

  Like so many social patterns that are usually discussed in terms of race, this pattern of inserting underclass newcomers into neighborhoods where they are resented by the pre-existing residents also exists when the underclass newcomers are white, and are resented by white pre-existing residents. In the best-selling memoir Hillbilly Elegy, the author—a white man from a hillbilly background—reported that his grandmother saw the government’s placing underclass people in their midst “as a betrayal, ensuring that ‘bad’ people would move into the neighborhood,” even though they “looked a lot like us,” but they were the kind of hillbillies who “gave our people a bad name.”72

  Among other things, she resented “the drugs and the late-night fighting” among the new neighbors that the government had placed in their neighborhood, and said of the woman who lived next door: “She’s a lazy whore, but she wouldn’t be if she was forced to get a job.” More pointedly: “I can’t understand why people who’ve worked all their lives scrape by while these deadbeats buy liquor and cell phone coverage with our tax money.”73

  Advocates of unsorting neighborhoods, whether by race or by class, argue that living in a better neighborhood will produce benefits for both the adults and the children who are moved in, and benefits of diversity for society at large. But these expected benefits to the newcomers from housing projects and high-crime neighborhoods have repeatedly failed to show up in extensive empirical studies by a wide variety of researchers on the federal government’s “Moving to Opportunity” program.

  A study of that program published in the Journal of Human Resources concluded: “We did not find evidence of improvements in reading scores, math scores, behavior or school problems, or school engagement, overall or for any age group.”74 Another study of the same program published in the American Journal of Sociology concluded that “there is no evidence that extra time spent in low-poverty integrated neighborhoods improves economic outcomes.”75

  Yet another study of the “Moving to Opportunity” program, published in the economic journal Econometrica likewise concluded, “we found no significant evidence of treatment effects on earnings, welfare participation, or amount of government assistance after an average of 5 years since random assignment.”76 The American Economic Review, the official journal of the American Economic Association, reached similar conclusions about the same federal program—“no consistent detectable impacts on adult economic self-sufficiency or children’s educational achievement outcomes” from the movement of thousands of people into higher income neighborhoods than the ones they came from.77

  The Quarterly Journal of Economics, the oldest American journal in economics, likewise concluded that “the changes in neighborhoods induced by MTO [“Moving to Opportunity” program] have not affected the employment rates, earnings, or welfare usage by a statistically detectable amount for household heads.”78

  In addition to these scholarly journals, a study published by the U.S. Department of Housing and Urban Development (HUD) was based on research on that same program which “followed more than 4,600 very low-income families in five U.S. cities over a 10- to 15-year period to examine the short- and long-term effects of moving to low-poverty neighborhoods.” Its conclusion was: “No discernible benefit to economic self-sufficiency, employment outcomes, and risky and criminal behavior for adults and children was observed as a result of moving. Similarly, moving had few positive effects on educational achievement for youth.”79

  Nevertheless, Secretary of Housing and Urban Development Shaun Donovan in 2013 “vowed to help urban blacks relocate to suburban neighborhoods, where they can have access to ‘good schools, safe streets, jobs, grocery stores,’ among other things.” Secretary Donovan claimed that realtors and landlords still discriminate against blacks. “African-Americans,” Donovan said, “are being denied their freedom of choice.”80 According to Investor’s Business Daily:

  Earlier this year, HUD broadened the authority of two anti-discrimination laws—the Fair Housing Act and the Equal Credit Opportunity Act—making illegal any housing or credit policy that results in disproportionately fewer blacks or Latinos receiving housing or home loans than whites, even if those policies are race-neutral and evenly applied across all groups.81

  Here, yet again, we see the implicit assumption that there would be no disparate outcomes unless there were disparate treatment. Moreover, that assumption seems almost impervious to evidence.

  One major difference between people sorting or unsorting themselves, on the one hand, and government officials sorting or unsorting them, on the other hand, is that people who sort or unsort themselves receive both the benefits and the costs of doing so. But government officials receive neither the benefits nor the costs of unsorting other people—and so may persist in the process, in utter disregard of benefits or costs that fall on others. Indeed, the political costs of admitting to having inflicted socially counterproductive policies are a powerful incentive to keep on inflicting those policies and ignoring or denying their consequences.

  It would be wrong to say that there have been literally no benefits at all to anyone from government-subsidized or government-enforced unsorting of people. While some studies have found some benefits to some segments of the low-income groups placed into middle-class neighborhoods by the government,82 these have seldom, if ever, been of the scope or magnitude envisioned when these programs were instituted.

  More fundamentally, negative consequences to the pre-existing residents of the communities into which they have been placed are seldom, if ever, mentioned—much less measured—in these studies. It is as if any benefit, however small, to the new residents automatically outweighs any costs, however large, to the pre-existing residents.

  “Disparate Impact” in Employment

  Although disparate treatment of individuals, because of the group to which they belong, is what is meant by Discrimination II, this is not always easy to prove in a court of law. Nor does anti-discrimination law, as applied in American courts, require such proof. If a given prerequisite for employment or promotion—a high school diploma, for example—has a “disparate impact” on some group, such as ethnic minorities, then the burden of proof falls on the accused employer to provide a justification of the requirement or else be judged guilty of discrimination.

  This process represents a major departure from American legal principles in both criminal and civil cases, where the burden of proof is usually on those making an accusation, rather than expecting the accused to prove their innocence. There are serious practical consequences of this very different legal standard in civil rights cases. There are costs to both employers and workers seeking employment, when the assortment and proportions of employees differ from the assortment and proportions of groups in the surrounding area.

  For the employer, the fact that a charge of discrimination can be made solely on the basis of statistics about his employees, without even a single flesh-and-blood human being actually claiming to be discriminated against, means that employers can be put through a costly and time-consuming legal process that can drag on for years, consuming millions of dollars in legal costs alone,
quite aside from costs imposed if this uncertain process leads to an unfavorable verdict.

  For example, a case charging the Sears department store chain with sex discrimination cost the company $20 million in legal fees83 and took 15 years to resolve through the federal courts—without the government having to produce even one woman, from any of Sears’ hundreds of department stores around the country, claiming to have been discriminated against. Statistical disparities alone were sufficient to keep this costly process going for more than a decade.

  In the end, Sears prevailed in the appellate courts. But few employers are in any position to sustain such financial costs for so many years, all the while operating under the public stigma of discrimination accusations that can affect public opinion and the sale of the company’s products.

  Most employers, including large corporations, find it expedient to settle such cases out of court, even when they have not violated anti-discrimination laws—and the number of such settlements is then used by critics to claim that employment discrimination is widespread. In 2012, for example, PepsiCo paid more than $3 million to settle a charge by the Equal Employment Opportunity Commission that the big soda and snacks company’s use of criminal background checks was discrimination against blacks.84

  This was a bargain compared to the cost to Sears of fighting a charge of discrimination against women, even though Sears eventually won the case. Moreover, having a charge of racial discrimination hanging over PepsiCo for years, while the case dragged on through the federal courts, could have cost more millions, as individuals and institutions decided to buy their sodas and snacks from some other company.

  In short, the outcome of “disparate impact” cases does not necessarily depend on either the quantity or the quality of the evidence. By the time of the PepsiCo settlement, an empirical study had already shown that companies using criminal background checks tended to hire more blacks than companies which did not use such checks.85 The crucial factor in such cases is not the trial, but the costliness of going to trial, both in legal fees and in the loss of business due to bad publicity. The only way for the accused to win, in any economically meaningful sense, is for the case to be thrown out of court instead of going to trial.

  Rarely does a judge refuse even to let a case go to trial, though that did happen in 2013 when the evidence presented by the Equal Employment Opportunity Commission was called by District Court judge Roger Titus “laughable” because of its “mind-boggling-number of errors” and because of the inconsistency of EEOC’s lawsuit against a company for using criminal background checks on job applicants, when the EEOC itself uses such checks.86

  The implications of the use of a “disparate impact” basis for costly lawsuits in civil rights cases does not end with employers. Workers can also be adversely affected, and not just with reduced employment opportunities for black workers who have no criminal record.

  When a federal agency can so easily make charges of discrimination on behalf of workers from racial or ethnic minorities—charges that can be costly and time-consuming to defend against in the courts, or charges that can force costly settlements out of court—that reduces the value of hiring black or other minority workers, even when their job qualifications are equal to the job qualifications of other workers who present no such legal risk.

  Employers therefore have incentives to locate their businesses away from concentrations of minority populations, so that they will not be as legally vulnerable to costly charges of discrimination if their work force does not end up with the same demographic makeup as that of the surrounding population.

  Japanese firms seeking to find locations for their first businesses in the United States have specified that they do not want to locate near concentrations of blacks in the local population.87 American firms that do the same thing, being more familiar with both the legal and the social atmosphere in the United States, may be less likely to leave a paper trail. Nevertheless, this raises the question whether anti-discrimination laws, as applied in the courts, provide incentives to discriminate against racial minorities as well as incentives not to discriminate, with their net effect being uncertain.

  Many observers who see racism as both widespread and widely effective in the job market fail to account for the fact that employers in competitive markets have actively sought out black workers, even in places and times where racism was rampant and undisguised, such as in South Africa during the era of apartheid, under a white minority government openly proclaiming white supremacy.

  Similarly, black American workers were sufficiently in demand more than a century ago, in the Jim Crow South, that the organized attempts of white employers and landowners to suppress black earnings often collapsed under the pressure of that demand for black workers and sharecroppers.

  Northern white employers sent recruiters into the South during the Jim Crow era to recruit black workers, on such a scale as to cause many laws to be passed in the South, restricting the activities of these recruiters by charging them licensing fees and imposing other restrictions, and with serious penalties for violating those restrictions.88 This clearly indicated a strong demand for black workers in both regions of the country.

  Within Northern communities, the demand for black workers was sufficient in the 1920s to cause Henry Ford and his executives to establish connections with clergy in Detroit’s black community, in order to get their help in sorting black job applicants. Similar arrangements existed in Chicago and Pittsburgh.89 The Ford Motor Company was, in effect, seeking low-cost access to knowledge of individual qualities, in order to judge each individual individually, instead of having to rely on information about group characteristics.

  In short, racism has not been sufficient to prevent a demand for black workers in a competitive market. It would be painfully ironic if anti-discrimination laws have been among the factors which reduced that demand in later times. Intentions, whether good or bad, do not predestine outcomes.

  Chapter 4

  THE WORLD OF NUMBERS

  When trying to understand economic and social disparities, statistics are often used, both to convey the magnitude of those disparities and to try to establish their causes. To some, numbers may convey a sense of objective, hard facts. But, even when the numbers are correct, the words that describe what the numbers are measuring may be incorrect or misleading. These include such basic numbers as income, unemployment rates and rates of arrest for violations of laws.

  Numbers may also be misleading, not because of any intrinsic defects in either the numbers themselves or in the words describing them, but because of implicit assumptions about the norms to which those numbers are being compared. Here the seemingly invincible fallacy of assuming an even or random distribution of outcomes as something to expect, in the absence of such complicating causes as genes or discrimination, can make many statistics that show very disparate outcomes be seen as indicating something fundamentally wrong in the real world, rather than something fundamentally wrong with the assumptions behind the norms to which those outcomes are being compared.

  Neither logic nor empirical evidence provides a compelling reason for expecting either equal or random outcomes among individuals, groups, institutions or nations.

  When used with an awareness of their pitfalls, statistics can be enormously valuable in testing competing hypotheses about disparate outcomes. But statistics may nevertheless be grossly misleading when they are distorted by errors of omission or errors of commission.

  ERRORS OF OMISSION

  The mere omission of one crucial fact can turn accurate statistics into traps that lead to conclusions that would be demonstrably false if the full facts were known. This often happens in comparisons of different ethnic groups and different income classes, among other comparisons.

  Group Disparities

  During a long and heated campaign in politics and in the media during the early twenty-first century, claiming that there was rampant discrimination against black home mortgage loan applicants
, data from various sources were cited repeatedly, showing that black applicants for the most desirable kinds of mortgages were turned down substantially more often than white applicants for those same mortgages.

  In the year 2000, for example, data from the U.S. Commission on Civil Rights showed that 44.6 percent of black applicants were turned down for those mortgages, while only 22.3 percent of white applicants were turned down.1 These and similar statistics from other sources set off widespread denunciations of mortgage lenders, and demands that the government “do something” to stop rampant racial discrimination in mortgage lending institutions.

  The very same report by the U.S. Commission on Civil Rights, which showed that blacks were turned down for conventional mortgages at twice the rate for whites, contained other statistics showing that whites were turned down for those same mortgages at a rate nearly twice that for “Asian Americans and Native Hawaiians.”

  While the rejection rate for white applicants was 22.3 percent, the rejection rate for Asian Americans and Native Hawaiians was 12.4 percent.2 But such data seldom, if ever, saw the light of day in most newspapers or on most television news programs, for which the black-white difference was enough to convince those journalists that racial discrimination was the reason.

  This conclusion fit existing preconceptions, apparently eliminating a need to check whether it also fit the facts. This one crucial omission enabled the prevailing preconception to dominate discussions in politics, in the media and in much of academia.

  One of the very few media outlets to even consider alternative explanations for the statistical differences was the Atlanta Journal-Constitution, which showed that 52 percent of blacks had credit scores so low that they would qualify only for the less desirable subprime mortgages, as did 16 percent of whites. Accordingly, 49 percent of blacks in the data cited by the Atlanta Journal-Constitution ended up with subprime mortgages, as did 13 percent of whites and 10 percent of Asians.3