Sam Barr’s most recent post makes the rather shocking claim that Rand Paul, the Republican nominee for the U.S. Senate seat in Kentucky being vacated by the retiring Jim Bunning, is a racist, or at least that he is not a non-racist. Sam deduces this from the fact that Mr. Paul is not a “consistent libertarian,” that he “picks and chooses” appropriate targets for government intervention and contends that eliminating racism in the workplace is an illegitimate function of government. Specifically, Rand Paul is pro-life and supports laws against abortion, but says he would have opposed the Civil Rights Act had he been in Congress in 1964.
In Sam’s estimation, Mr. Paul is not racist in the sense that he wears a white hood and burns crosses, but in the sense that, in the words of the illustrious Mr. Kanye West, he “does not care about black people.” Sam’s post rightly implies that the bar for calling someone the “R-word” should be relatively high, a standard many of his fellow partisans have often ignored: witness Sen. John Edwards’s absurd insinuation in the 2004 vice presidential debate that Dick Cheney was racist because he voted against the holiday for Martin Luther King, Jr., while in Congress, or Rep. Diane Watson, an African American congresswoman who condemned her Republican colleagues for the mere act of opposing “the first president who looks like” her, or the relentless attempts by the media to indict the Tea Partiers as a reincarnation of the KKK (we know this is not true, by the way, since if it were, the President pro tempore of the U.S. Senate would be scrambling to join its ranks). Sam’s accusation, unlike many leveled by trigger-happy race-baiting Democrats, is reasonable and deserves an answer. I should explain at the outset that I am not a libertarian per se, that I supported Trey Grayson, Rand Paul’s erstwhile opponent in the Republican primary, and that I have no intention of passing either positive or negative judgment on the Civil Rights Act. What I will argue here is that simultaneous opposition to anti-discrimination policies and support of anti-abortion laws does not a racist make. The reader will have to excuse the length of this post; such a serious charge requires a thorough response.
In Economics 1017, Professor Jeffrey A. Miron, Harvard’s foremost authority on libertarianism, provides an analysis of racial discrimination in the workplace from a libertarian perspective (and luckily for the HPR, I retained my lecture notes from the course). The economic model of discrimination, he explains, begins with the assumption that some people have a “taste” for discrimination, which in this case means people prefer hiring or buying from only persons of a certain race. As any graduate of Ec-10 knows, a free market will in theory drive racist employers out of business. Assume, for example, that some white employers do not like hiring blacks. This preference initially reduces the demand for black employees and reduces their wages, but this results in any employer with non-discriminatory preferences obtaining a cost advantage by hiring black employees. Since the non-discriminating firms have lower costs, they can set lower prices and take profits away from the discriminating firms. The discriminating firms exit the industry as they lose money, which then reduces the demand for white relative to black employees, and results in equal wages for blacks and whites in equilibrium. I would add to this that the economic disincentive to refrain from serving minority customers is even more obvious: business owners who choose not to serve an arbitrary segment of the population put themselves at a competitive disadvantage by depriving themselves of access to a broad swath of the market. In theory, therefore, competitive markets provide a potentially strong counterweight to employer discrimination.
An alternative assumption is that discriminatory preferences come from customers. For example, suppose restaurant patrons prefer to be served by white waiters, meaning they are willing to pay a higher price even if the quality of service is the same. In this case, Miron notes, a higher wage for white waiters can persist in equilibrium, but even here there are economic pressures that counteract the discriminatory preferences of customers. For one thing, restaurant owners face higher costs than they would if they could use both white and black wait staff, so they might still use both if customer discrimination is weak. And since some customers presumably do not care, the benefits of accommodating the customers with discriminatory preferences are potentially small. The same logic, incidentally, applies to situations in which some whites prefer not to be served in an establishment that accommodates blacks. A priori reasoning thus indicates that economic forces are likely to hinder discrimination in the workplace. As an example, Miron cites Levine, Levkov, and Rubinstein (2008), who determine that increased competition resulting from deregulation in the banking industry from the mid-1970s to the mid-1990s reduced both the racial wage gap and racial segregation in the workplace, particularly in states with a comparatively high degree of racial prejudice.
That said, Miron explains that what is known as “statistical discrimination” in employment may be rational as a result of the correlation of unobservable factors, such as educational achievement and general competence in the workplace, with observable factors like race. If African Americans are disproportionately likely to be poor workers because they receive disproportionately poor educations, in other words, then it can be rational for employers to use race as a proxy for the less observable characteristics of intelligence and competence. The fact that statistical discrimination might be rational, as Miron points out, does not mean it is acceptable. But if statistical discrimination is the underlying cause of workplace discrimination, improving the quality of education offered to minorities is likely to be more effective than direct anti-discrimination policies à la the Civil Rights Act. Incidentally, Republicans have long advocated school choice through government vouchers to improve education within minority communities, an effort that has been blocked by teachers’ unions and their allies among congressional Democrats.
Assuming workplace discrimination is based on employer or consumer preferences rather than statistical correlation with unobservable traits, policy may undertake to counteract discrimination either by prohibiting it in hiring, promotion, firing, establishing wages, selecting customers to serve, and so forth, or, in the job market, by “affirmatively” promoting the hiring of targeted groups through quotas. The Civil Rights Act of 1964, among other things, ended racial discrimination in all federal government agencies and organizations receiving federal support, and prohibited discrimination in the private sector to the extent permitted under the Constitution. While the private sector provisions probably had some impact, Sam’s opinion that “Paul gets the Civil Rights Act completely wrong” because the “ban on private discrimination was absolutely central to its achievement” is hardly a matter of scholarly consensus. I couldn’t agree more, by the way, with Max Novendstern’s comment in his response to Sam’s post that the Civil Rights Act should be judged based on its “material consequences, not just (and not primarily) the soundness of its ethical claims.” Although black-white wage differentials have declined substantially over the past fifty years, Miron points out that the gap began declining before the federal government adopted anti-discrimination policies, and that there is little dispute that “forces other than anti-discrimination policy played a significant role in reducing race … wage differentials.” One plausible candidate that he mentions is increasing educational attainment by African Americans. In “Catching Up: Wages of Black Men” in The American Economic Review, Finis Welch notes that he and James P. Smith observed that although the relative wages of blacks increased in the decades after 1960, “there was little evidence of improvement within cohort”; the narrowing wage gap was in other words a result of younger African Americans receiving better wages rather than increased wages for blacks already in the workforce (Smith and Welch, 1977, 1984, 1989). Although there were clear employment shifts toward industries with concentrations of firms presumed to be more sensitive to affirmative-action pressures, the wage gains were “pervasive and not restricted to these industries.” For these reasons, Smith and Welch conclude that improvements in the quantity and quality of schooling were more important in decreasing workplace discrimination than federal legislation. Miron further points out that the Civil Rights Act was accompanied by Justice Department suits against Jim Crow laws (which, let’s not forget, were racist government interventions frequently opposed by profit-seeking private firms) as well as private actions including boycotts and protests in the South. So while some academics and policymakers contend that the Civil Rights Act was crucial to eliminating racism, others have argued that it was unnecessary, and, as Miron points out, “reasonable people can disagree.”
One might assume that the Civil Rights Act was justified if it succeeded even to a very minor degree in eradicating racism in the private sector. This is a legitimate position, to be sure, but the fundamental libertarian philosophy as articulated by Professor Miron, which Sam appears to overlook, is that while the free market often delivers imperfect outcomes, government intervention generally does more harm than good. It is therefore necessary to examine the potential costs of anti-discrimination policy, including those that may accrue to the very minority communities they are intended to help, before arriving at a final evaluation. For one thing, libertarians often contend that the distinction between merely banning employment discrimination on the one hand, and implementing racial quotas on the other, is not meaningful in practice, since without affirmative action there is no way to enforce fair hiring practices (employers, in other words, can always claim that whites are simply more qualified). And Miron explains that affirmative action entails potentially draconian costs, including perpetuated negative stereotypes of minority communities (i.e., the perception that minorities are unable to find employment without the government’s help), resentment among non-minorities (i.e., whites who feel cheated out of positions for which they are more qualified), reduced educational attainment and effort within minority communities (i.e., reduced incentive for minority communities to improve themselves if the government guarantees them a certain number of jobs), and reduced efficiency (i.e., firms cannot hire the most qualified employees). But Miron explains that even anti-discrimination measures by themselves might do more harm than good. If an employer knows he might get penalized for firing, or not promoting, or not giving a raise to a minority employee, it might make sense to avoid hiring members of the protected group in the first place. One prominent example of this is the Americans with Disabilities Act (ADA) of 1990, which requires employers to accommodate disabled workers and outlaws discrimination against the disabled in hiring, firing, and pay. Acemoglu and Angrist (2001) observe a sharp drop in the employment of disabled workers after the ADA went into effect, and ironically isolate the ADA itself as the likely cause. Sam’s libertarian straw man advocates the right to do “whatever you want with what’s yours,” even if it means “perpetuating a system of race-based subordination.” While it would be difficult to oppose a government effort to eradicate discrimination that reliably produced results in excess of its costs, Miron points out that policy cannot ban discrimination without endorsing the view that firms are partially “public” and can be told to operate in “socially” approved ways. Even if this does more good than harm in the context of discrimination, he says, blurring the private/public distinction might legitimize ill-advised government intervention in other areas. All this is to say nothing of the deadweight loss from taxation needed to fund the personnel who enforce the laws. So the notion that anti-discrimination legislation in general, and the Civil Rights Act in particular, is an unequivocal good is far from accurate.
For pro-life libertarians like Rand Paul, the cost-benefit equation with respect to abortion is substantially different. Abortion restrictions, like anti-discrimination laws, undoubtedly entail costs. The difference is that pro-lifers equate the termination of unwanted pregnancies (at least those that do not result from rape or incest and do not threaten the mother’s life) with murder. Prohibiting abortion is therefore an attempt to prevent the needless slaughter of innocent human life, which, unlike ensuring equal employment opportunities for every citizen, is justified at virtually any cost. In the most recent issue of the Harvard Salient, Roger G. Waite notes that the United States has the highest abortion rate in the developed world, as well as a legal system extremely permissive of abortion. This does not imply causation, but it’s difficult to imagine that laws against abortion would increase the number of abortions, or that our high abortion rate results from Americans’ moral depravity (or, as Mr. Waite posits, the structure of our health care system). A true libertarian who opposes the Civil Rights Act and supports anti-abortion laws, therefore, is not being inconsistent, but making a rational cost-benefit analysis of government intervention in two distinct cases.
Sam ends his post by invoking the questions Ezra Klein poses to Mr. Paul as to whether the federal government can set the private sector’s minimum wage, tell private businesses not to hire illegal immigrants, tell oil companies what safety systems to build into an offshore drilling platform, tell toy companies to test for lead, or tell liquor stores not to sell to minors. I’ll spare you the explanations, but I can assure you that what Sam calls “consistent libertarians” can indeed oppose each of these forms of intervention, or oppose some and not others, according to a rational cost-benefit analysis. Sam and Mr. Klein might reach different conclusions, but this does not imply that libertarians or Mr. Paul are “willfully blind and insensitive to racism.”
None of this is to say that I endorse or condemn either anti-discrimination policies or the libertarian response thereto. The point is that it is entirely possible for Rand Paul to be a consistent libertarian, and not to be a racist, while both opposing the Civil Rights Act and supporting legal restrictions on abortion. Perhaps Mr. Paul is racist, but not by virtue of anything he has said about the Civil Rights Act. It is important to hold politicians accountable on an issue as fundamental as race, and I do not fault Sam for raising this accusation given Mr. Paul’s opposition to landmark civil rights legislation. I would advise Sam, however, that a valid charge of racism must withstand the strictest of scrutiny. This one does not.
Photo Credit: Wikipedia