March 5, 2013

Now Hiring Ex-Cons

Can the government force firms to employ convicted criminals just because they are minorities?

My column for Defining Ideas last week stressed that antidiscrimination laws can wreak havoc on job creation. The Equal Opportunity Employment Commission, a federal agency tasked with enforcing antidiscrimination laws, has demonstrated just how destructive such laws can be. One instance of its folly is its “Enforcement Guidance” of April 2012, which has come to prominence after recent public hearings before the U.S. Commission on Civil Rights.

Epstein
Illustration by Barbara Kelley

With the Enforcement Guidance, all private employers and all state employers must use detailed and particularized inquiries before turning down a minority applicant who has a criminal arrest or conviction on his record, even though employers can turn down a white applicant with the same past record without going through such hoops.

An Upside Down Civil Rights Case

To the unpracticed eye, the EEOC ruling looks genuinely perverse. The law that was intended to end discrimination by private parties now institutionalizes it by government. Title VII of the 1964 Civil Rights Act has, as its purpose, to make it “an unlawful employment practice for an employer to fail or refuse to hire or to discharge any individual…because of such individual’s race, color, religion, sex, or national origin.”

Thus, the newest confection out of the EEOC orders most employers to do exactly what the law forbids. It introduces an explicit classification into the hiring equation by imposing a higher standard for refusing to hire minority workers than for others. The Enforcement Guidance also applies even when it is clear that the employer’s refusal to hire certain workers is not because of race but because of the evident risk that a criminal record could present to the employer, its other employees, and its customers.

The EEOC introduces what is termed “disparate treatment” by race in its supposed effort to prevent discrimination. The results are perverse at best. To take just one example, James Bovard, writing in the Wall Street Journal, reports that in 2010, the EEOC initiated litigation against G4S Secure Solutions “after the company refused to hire a twice-convicted Pennsylvania thief as a security guard.” Needless to say, the EEOC did not offer to indemnify G4S should they be held liable for any torts of their employee while on the job.

Nothing is more dangerous in public or private affairs than power without responsibility. It is therefore a fair question to ask how the EEOC guidance took its final shape. The story has both a procedural and a substantive dimension.

Administrative Legerdemain

On the procedural issue, it is important to understand that the innocent word “guidance” carries with it explosive administrative law implications. One common feature of modern regulatory statutes is that they delegate broad powers to agencies to make regulations. The Administrative Procedure Act of 1946, passed after the Republicans retook control of the Congress at the end of World War II, sought to impose orderly limits on the authority Congress delegated to administrative agencies.

At a minimum, regulations were supposed to go through a “notice and comment” procedure. Proposed regulations are published in the Federal Register so that individuals and groups can comment on their merits. But the EEOC’s Enforcement Guidance blocks that form of public disclosure. In fact, agencies can now formulate their positions in the dark of night without any public input at all.

Though the Enforcement Guidance is said not to have any statutory force to it, employers ignore it at their peril. Doing so, after all, could ignite a costly EEOC investigation. The government thus gets full in terrorem effect while bypassing all procedural safeguards. For years, the courts have tolerated this device—the guidance—even though they should have given it no legitimacy.

The Substantive Transformation

The EEOC’s broad reading of the Civil Rights Act would not be possible without the historical willingness of the Supreme Court to adopt an expansive civil rights theory that reads the “because of” out of Title VII. (As a reminder, the relevant part of the law makes it “an unlawful employment practice for an employer to fail or refuse to hire or to discharge any individual…because of such individual’s race, color, religion, sex, or national origin.”)

The legislative history makes it painfully clear that in order to overcome political opposition, the Act did nothing to prevent an employer from using whatever tests it liked to select those employees whom it thought best for the job. Indeed the law contained an explicit protection for any employer “to give and to act upon the results of any professionally developed ability test provided that such test, its administration or action upon the results is not designed, intended or used to discriminate because of race…”

Read in sequence, the term “used” works in parallel with “designed” or “intended,” and thus requires a bad state of mind. This statutory language seems to block the selection of a test that has an eye toward getting the preferred racial result. But in the fabled 1971 case of Griggs v. Duke Power Co., a unanimous Supreme Court, speaking through Chief Justice Warren Burger, upset that carefully drawn legislative compromise. It tortured the language of the statute against the entire weight of the legislative history. The vital intent requirement was excised, so that a test was used improperly if it had a disparate impact by race—that is, if the test had a heavier negative effect on one group than the other, even when its use was only intended to sort out the best workers on the job.

Owing to the major differences in education and job history, most tests do have disparate impact by race. Under the Griggs rule, the employer could use that test only if some “business necessity” required its use, and then only if it had a near-perfect predictive power, which virtually no test ever has.

At that point, it was off to the races. We know that tests are valuable because they are used as diagnostic tools. But the protracted effort to get tests that are both predictive and valid has led to inordinate and wasteful expense. What makes the situation inexcusable is that by the early 1980s, no one could credibly claim that the hard look on tests was intended to ferret out conscious discrimination that went below the radar.

Indeed, in the misguided 1982 decision of Connecticut v. Teal, the late Justice William Brennan, speaking for a five-member majority, missed an opportunity to stop this abuse when he insisted that the disparate impact analysis applied to state agencies that had adopted bona fide affirmative action programs, which embodied a form of reverse discrimination. The correct approach in that case would be to let the tests go so as to get the strongest candidates from both groups. Adopt that approach and the scope of employment discrimination law shrinks by ninety percent—the better for everyone.

The EEOC does not see matters that way, so it defends its Enforcement Guidance by explicit reliance on both of these Supreme Court cases. But even if we assume that the Griggs-Teal framework makes sense, the EEOC Enforcement Guidance still comes up short. The factual predicate for the decision rests on the differential conviction rates for African-American and Latino males. More specifically, as of 2010, the EEOC reports, “Black men had an imprisonment rate that was nearly 7 times higher than White men and almost 3 times higher than Hispanic men.”

Note, however, that the Enforcement Guidance does not make any further claim that the differential rates of incarceration were due to any error in the operation of the criminal justice system. At this point, the question should become this: why should persons who received the same treatment inside the criminal justice system now receive different treatment under the antidiscrimination laws?

The EEOC never addresses that question, but instead goes into painful detail as to the extensive requirements that are said to influence the decision. In so doing, it refers to three factors derived from a 1977 decision of the Eighth Circuit Court of Appeals, Green v. Missouri Pacific Railroadwhich held that in dealing with this issue, the court should consider the nature of the crime, the time elapsed since the crime, and the nature of the job.

Of course, it is one thing to identify the relevant facts, and another to apply a policy. In Green, all we know was that the plaintiff had been convicted of refusing military induction, for which he served twenty-one months in prison before he was paroled for about two years, after which he applied for an unspecified position.

That case is a clear outlier. Unfortunately, this pattern of facts gives no indication of how to treat potential security guards who have committed violent crimes. The EEOC’s useless Enforcement Guidance only offers far-fetched examples at best. Additionally, the guidance neglects to tell an employer, who may receive hundreds of applications for a single position, how to make a detailed “individuated assessment” of each applicant and still remain economically viable. Nor does it say whether an employer remains in violation of Title VII by turning down ten minority applicants with criminal records after hiring the eleventh.

What makes this entire approach even more bizarre is that many agencies of the federal government make the same use of criminal conviction records that the EEOC demands that all states and private employers reject. At no point, however, does the EEOC claim that these federal agencies discriminate in any way. Nor does the EEOC address why it thinks these federal agencies are misguided in making their own considered judgments regarding the hiring of employees.

The EEOC puts state and private employers in the crossfire of regulations while announcing blandly that it will “coordinate” with government agencies on “best practices,” i.e., the destructive rules it wants to impose on everyone else.

Who Benefits?

At the end of the day, what good does the EEOC hope to gain from this massive undertaking?

One of the great benefits of a competitive labor market is its self-corrective nature. The correct social question therefore is not whether this or that firm decides to hire a worthy applicant with a criminal record. It is whether any firm makes a positive hiring decision for a worthy candidate; if not, in competitive labor markets, any errors made by one potential employer can be corrected by favorable decisions by another.

Ironically, however, that redundancy is undercut by the EEOC’s uniform Enforcement Guidance. Some studies already suggest that firms are “much less likely to hire minority applicants when background checks are banned.” That result should not come as any surprise. The white male workers who are not protected by Title VII can offer employees this precious guarantee: the ability to hire and fire at will. Minority workers cannot waive their ill-conceived protections under Title VII, and thus are prevented from competing along this critical dimension. The EEOC Guidance may help some minority workers in a few cases, but it will hurt even more.

And by raising transaction costs, the EEOC will continue on its mindless job-killing path. Once again, the EEOC seems utterly oblivious to the harm that it causes to the groups that it most wants to help—and indeed to everyone else. 


Richard A. Epstein, Peter and Kirsten Bedford Senior Fellow at the Hoover Institution, Laurence A. Tisch Professor of Law at New York University, and senior lecturer at the University of Chicago, researches and writes on a broad range of constitutional, economic, historical, and philosophical subjects. He has taught administrative law, antitrust law, communications law, constitutional law, corporate law, criminal law, employment discrimination law, environmental law, food and drug law, health law, labor law, Roman law, real estate development and finance, and individual and corporate taxation. His publications cover an equally broad range of topics. His most recent book, published in 2013, is The Classical Liberal Constitution: The Uncertain Quest for Limited Government (2013). He is a past editor of the Journal of Legal Studies (1981–91) and the Journal of Law and Economics (1991–2001).


Letters to the editor may be sent to definingideas@stanford.edu. Editors reserve the right to reject or publish (and edit) letters.