By Terence G. Connor and Kevin J. White, Hunton & Williams LLP
(Terence G. Connor and Kevin J. White are partners in the firm of Hunton & Williams LLP. Mr. Connor works out of the firm’s Miami offices and Mr. Whiteworks out of its Washington, D.C., and Houston offices.)1
On April 25, the Equal Employment Opportunity Commission adopted its Enforcement Guidance: Consideration of Arrest - Conviction Records in Employment Decisions under Title VII of the Civil Rights Act of 1964(“2012 Guidance”), expanding on its 1987 and later policy statements to its field offices.
It adopted this new Guidance to instruct district offices in making “reasonable cause” determinations on claims of race or national origin discrimination stemming from the exclusion from employment because of prior crimes. Making these determinations is the EEOC’s first statutory obligation when investigating charges. 42 U.S.C. § 2000e-5(b).
Citing sparse judicial history on the subject, the 2012 Guidance addresses the use of conviction records as a screen for employment that may have an adverse impact by race or national origin. This Guidance introduces further complexity into employers’ risk-avoidance efforts and their obligation to avoid foreseeable harm to customers and other employees.
In this action the Commission has attempted to address the high rate of incarceration in America, particularly among African-Americans and Hispanics, see 2012 Guidance at nn. 65–74, a social problem worthy of significant attention.
Among the questions that will be litigated is whether the EEOC’s approach to the subject goes beyond its statutory mission to address discrimination and intrudes on matters reserved to Congress. Whether the Guidelines will receive judicial deference as an interpretation of the law the EEOC has been assigned to enforce remains to be seen.
Federal statutes require that the federal government, acting as an employer, does precisely the kinds of background checks that this Guidance addresses. Congress has concluded that certain government positions are sufficiently sensitive to require that applicants for those positions undergo a criminal background check aimed at evaluating whether an applicant’s appointment may present a risk of harm. See id. at nn. 130–352. Several state and local governments have adopted similar requirements. Id. at n.165.
Criminologists agree that at least two factors are highly predictive of future criminal misconduct that might give any employer pause:
• The arrest and convictions history of the applicant; and
• The age of the offender at the time of first conviction and at time of application for employment.
See Shawn D. Bushway, Paul Nieuwbeerta & Arjan Blokland, The Predictive Value of Criminal Background Checks: Do Age and Criminal History Affect Time to Redemption, 49 Criminology 27, 52 (2011).
Applying common experience has led juries to find a common law duty as well. An employer is negligent if it ignores available information about an applicant’s criminal history. See, e.g., Tallahassee Furniture Co. v. Harrison, 583 So. 2d 744 (Fla. 1991) (absence of background check supported $2.5 million jury verdict for customer brutally assaulted by employee with multiple unexamined convictions).
Of course, Congress and many state legislatures have also prohibited discrimination against employees and applicants based on their race or national origin, etc. 42 U.S.C. § 2000e-4. The EEOC seeks judicial deference for its interpretations of the statute and has extensive enforcement authority through investigation, conciliation of charges filed and, in appropriate cases, litigation. 42 U.S.C. § 2000e-5. This 2012 Guidance is part of its exercise of that enforcement authority. It will likely seek court deference to its Guidance in prosecuting cases under this theory.
Title VII was enacted in 1964 primarily to address persistent post-slavery discrimination against, and segregation of, African-Americans in the workplace. Its principal focus was to make it “an unlawful employment practice for an employer –
(1) to fail or refuse to hire or to discharge any individual, or otherwise discriminate against an individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s race, color, religion, sex, or national origin; or (2) to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s race, color, religion, sex, or national origin.
42 U.S.C. § 2000 e-2(a). Intentional discrimination against workers on these bases is a clear violation of law, and courts have broad equitable and legal remedies to address violations.
But in its unanimous decision in Griggs v. Duke Power Co., 401 U.S. 424 (1971), the Supreme Court went further, holding that practices that discriminate in their effects, regardless of intent, are also prohibited unless job related. A facially neutral requirement that any employee who wished to transfer into another department from a previously all-black unit must have achieved a high school diploma violated Title VII if that requirement had the effect of excluding significantly more African-American applicants for those positions, unless the employer could establish that the requirement was “related to job performance.” Griggs, 401 U.S. at 431.This was the Supreme Court’s first application of Title VII to neutral policies that adversely affected a protected group regardless of intent. Later cases in the Supreme Court arguably relaxed Griggs’ rule in different fact situations.
One example of the Supreme Court’s evolution in applying the “business necessity” defense is New York City Transit Authority v. Beazer, 440 U.S. 568 (1979). Significantly, the Beazer Court held that the New York City Transit Authority could exclude all current methadone users because that policy or practice had a “manifest relationship to the employment in question,” even if it adversely affected minorities. Beazer, 440 U.S. at 587 n.31 (1979).
The Court did not dwell on whether the absence of drug use was relevant to any specific element of the job(s) in question, but seems to have found the principle self evident that a transit authority should have drug free employees.
Congress codified and defined this theory in 1991. Where there is a racially adverse impact caused by selection standard, the employer must prove the standard to be “job related and consistent with business necessity.” 42 U.S.C. § 2000e-2(k). This language was adopted with a bipartisan congressional agreement that this terminology was to be applied in cases without references to any legislative history. Civil Rights Act of 1991, Pub. L. No. 102-166, § 105(b), 105 Stat. 1071, 1075. The courts are left to define the imprecise language “related” and “consistent” case by case with no help from legislative history.
Only two appellate courts have addressed the question of whether an employer’s consideration of an applicant’s criminal conviction history meets the business necessity requirements: Green v. Missouri Pacific Railroad Co., 523 F.2d 1290 (8th Cir. 1975), aff’d, 549 F.2d 1158 (8th Cir. 1977), and El v. SEPTA, 479 F.3d 272 (3d Cir. 2007). They did not advance the jurisprudence significantly.
In Green (a good example of the principle that bad facts yield bad law), the railroad used a no-felonies-ever-for-any-job standard to exclude a Vietnam era African-American applicant, who had been convicted of “dodging the draft,” from a position as a clerk. Addressing the “business necessity” defense, the 8th Circuit found that such an expansive exclusion could not be justified as job related. Green, 523 F.2d at 1298.
In El, the applicant had a 40-year-old conviction for murder and was rejected from a position driving a vulnerable population of disabled commuters because of that conviction. On a scant record, after presuming adverse impact on African-Americans, the 3rd Circuit affirmed summary judgment for SEPTA based on a finding that the policy excluding anyone with a felony conviction from that job was “consistent with business necessity,” at least on that court record. El, 479 F.3d at 235.
The court declined to follow the policy statement adopted by the EEOC in 1987 because of the absence of any indication that the Commission engaged in any thorough analysis when it simply adopted the Green standards to guide its agency decision makers in 1987. 479 F.3d at 244.
District courts have taken a deferential stance in a different direction and, in effect, deferred to employers’ risk assessments in their effort to balance non-discriminatory policies with other obligations they bear under the law. In the absence of any legislative history from the Civil Rights Act of 1991 on the business necessity defense, the district court in Rhode Island applied “consistent with business necessity” in a different setting and described it this way:
. . . the term “consistent with business necessity” requires something less than a showing that the challenged practice is essential to the conduct of the employer’s business but something more than a showing that it serves a legitimate business purpose. What it appears to require is proof that the challenged practice is reasonably necessary to achieve an important business objective.
Donnelly v. R.I. Bd. of Governors for Higher Educ., 929 F. Supp. 583, 594 (D. R.I. 1996) (emphasis added).
The cases have applied that more employer friendly standard generally to evaluations of “consistency with business necessity” when assessing risk from prior convictions. The Donnelly court seems to have captured the concept these courts apply. An employer’s evaluation of risk should be “reasonably necessary to achieve important business objectives” of the employer. The district courts have been nearly unanimous in upholding employer decisions to date. See, e.g.,Foxworth v. Pa. State Police, 228 F. App’x 151, 158 (3d Cir. 2007); Naugles v. Dollar Gen., Inc., No. 4:08CV01943, 2010 WL 1254645, at *5 (E.D. Mo. Mar. 24, 2010); Fletcher v. Berkowitz Oliver Williams Shaw & Eisenbrandt, LLP, 537 F. Supp.2d 1028, 1030 (W.D. Mo. 2008); Clinkscale v. City of Phila., No. Civ. A. 97–2165, 1998 WL 372138, at *2 (E.D. Pa. June 17, 1998); Williams v. Carson Pirie Scott, No. 92 C 5747, 1992 WL 229849, at *1 (N.D. Ill. Sept. 9, 1992). But cf. Field v. Orkin Exterminating Co., No. Civ. A. 00-5913, 2001 WL 34368768, at *1–3 (E.D. Pa. Oct. 30, 2001) (granting the plaintiff leave to amend her complaint to provide more evidence regarding whether the employer’s blanket policy against hiring persons with recent criminal records, but not deciding whether the plaintiff would succeed in her claim).
As the 3rd Circuit observed in El v. SEPTA, in 1987 the EEOC simply adopted the Green rule and issued guidance urging that three factors might support a defense of job relatedness:
• The nature and gravity of the offense or conduct;
• The time that has passed since the offense, conduct and/or completion of sentence; and
• The nature of the job held or sought.
See EEOC, Policy Statement on the Issue of Conviction Records Under Title VII of the Civil Rights Act of 1964 (Feb. 4, 1987). Since then the incarceration rates have increased, especially among African-Americans and Hispanics, and the Commission felt compelled to act.
Twenty-five years after its adoption of the Green factors, following two meetings on the subject with what some will argue was limited input and flawed analysis of how these standards might be a surrogate for racial discrimination, the Commission issued its current Guidance on April 25. Its purpose was to “consolidate and update” its policy statements on the issue. 2012 Guidance, at p. 1.
The new Guidance starts with the assumption that national criminal incarceration statistics are the relevant measure for adverse impact on employment everywhere. Id. at nn. 11–14. Therefore, since convictions and incarcerations adversely impact African Americans and Hispanics nationally, the EEOC focus turns quickly to discussion of the employer’s burden to defend.
The Commission does acknowledge that there may be other ways to define the relevant labor market — a perhaps obvious observation since data for Omaha or Des Moines likely differ from New York City or Los Angeles on all relevant measures. Nevertheless, the focus of the Guidance is on nationwide statistics and on the statutory standard of “job related and consistent with business necessity.”
The Commission identified two ways an employer could defend rejecting a convicted offender:
(1) validation of the employer practice consistent with the 1978 Uniform Guidelines on Employee Selection Procedures, or
(2) development of a “targeted exclusion policy” and a program for “individualized assessment” of a candidate who presents with a conviction that might otherwise preclude employment.2
The Commission recognizes that validation under the Uniform Guidelines is impractical. It offers the understatement that “such studies are rare at the time of this drafting.” Id. at V.B.5. There are none. In fact, those Uniform Guidelines apply content, criterion, or construct validity standards to tests or devices that measure specific performance criteria. Employers considering an applicant’s suitability against a criminal record are considering general qualifications and risk of foreseeable harm.
It is not possible, for example, to validate exclusion of convicted burglars from sales jobs for security alarms. The absence of burglary convictions is not a necessary qualification for the performance of that job. It could be argued that burglars are better “qualified” by their experience for the specific features of the job.
But it is a “legitimate business objective” for an employer to seek at least two general characteristics in its employees that various other principles of law may require them to consider. When an employer reviews a candidate, it is at least legitimate to consider: (1) his honesty, or (2) any indications that she might pose a physical threat to others. It is here that the 2012 Guidance where the EEOC’s reading of criminology and the employer’s duty of care may collide. An employer’s reasonable perception that an applicant poses a threat to customers, employees or its business property advances its legitimate interests and seems to receive court deference at least in the absence of proof that it was applied differently among races or ethnic groups.
In the 2012 Guidance, the Commission “believes” that employers will “consistently meet the ‘job related and consistent with business necessity’ defense,” when they have a “targeted exclusion” policy with a “demonstrably tight nexus to the position in question,” and when they allow the applicant to provide additional information or argument that may justify an exception to their exclusion policy, i.e., the “individualized assessment.” Id at V.B.8 to V.B.9. The demonstrable tightness of nexus may be an area of disagreement. The Guidance says this individualized assessment should consider information that the applicant was not correctly identified in the criminal record or that that record is otherwise inaccurate and maybe at least:
• the facts and circumstances surrounding the offense;
• the number of offenses;
• age at the time of conviction or release from prison;
• evidence that the individual performed the same type of work post-conviction with the same or other employers without incident;
• the length and consistency of employment history before and after the offense;
• rehabilitation efforts;
• employment or character references; and
• whether the individual is bonded under federal, state or local programs.
Id. at V.B.9. These are legitimate suggestions, some of which are consistent with current criminological research.
A question will be whether the EEOC has stepped over the traces and attempted to legislate new protections where its statutory duty is to address prohibited discrimination. An employer’s judgment on who presents risk, so long as the judgment is applied to everyone, has so far been receiving deference from the courts as described above.
In weighing their various legal and moral obligations, employers operate between the “rock” of putting their business or customers at foreseeable risk by not considering indications of violence or dishonesty, and the “hard place” of these standards. The prospect of a candidate’s recidivism to crime is a legitimate business concern for employers.
The data that drives criminology to its conclusions of “redemption” or desistence from crime are incomplete, and the “statistic” that ultimately matters when an employee appears with a gun is whether the employer made a diligent use of available information to assess the risk that the employee foreseeably presented before that moment arrived.
So questions will continue in this area. They include the facts that many criminologists do not believe that “crime type” is particularly predictive of recidivism, that criminologists may not have sufficient accurate data to make their assessments reliably enough to override on-scene observations and assessments, and that their measure of recidivism in the United States looks at later arrests (not convictions) as evidence of a failure to desist from criminal activity or to achieve redemption. See Alfred Blumstein & Kiminori Nakamura, “Redemption” in an Era of Widespread Criminal Background Checks, 263 Nat’l Inst. of Just. J. 10, 14 (2009).
Moreover, to add to the mischief, the Bureau of Justice Statistics’ data on recidivism reveals the uncomfortable fact that race itself actually is a factor of some moment in the prediction of recidivism. Bureau of Justice Statistics, Prisoner Recidivism, available at http://bjs.ojp.usdoj.gov/index.cfm?ty=datool&surl=/recidivism/index.cfm.
The bottom line is that employers navigate in uncertain waters when trying to assess any risk of harm to themselves or their customers, employees, and visitors, and they assess those risks at some significant peril that they will unintentionally affect more people of one race or gender than another.
While this Guidance offers some potentially helpful factors for an employer to consider, the Commission appears not to have considered other facts that lead in a different direction. For example, studies find that doing background checks actually enhances minority employment prospects,3 and employment does not seem to bear on recidivism rates.
Ultimately the correction of the national over-incarceration rate is a matter that should be addressed by legislative process and the criminal justice system. It will be for the courts to determine whether that correction belongs at the door of the private employer. Indications of past voluntary criminal conduct cannot safely be ignored and may not be found to be comparable to imposing a new requirement for placement in a previously segregated workforce as in Griggs.
Given the direction the Commission has taken, it appears that there will be protracted litigation before judges who make the same analysis as employers every day in deciding sentencing issues. Otherwise, correction through legislative action may resolve the issues.
A simple amendment to Section 703(h) of Title VII could offer clarity by providing that “. . . it shall not be an unlawful employment practice for an employer . . . to take adverse action against employees or applicants because of their having been convicted of a crime, provided that such actions are not the result of an intention to discriminate . . .” See 42 U.S.C. §2000e-2(h). This straightforward principle would accommodate both the employer’s freedom to act to avoid legitimate risks and its duty to apply its standards equally among the races.
But even if Congress does not act, the district courts have already shown a proclivity for giving deference to employers on this issue, and ultimately it may be settled in the Supreme Court, where at least one Justice questions whether the disparate impact theory itself should survive a review for constitutional muster. See Ricci v.DeStefano, et al., 557 U.S. 557, 594(2009) Scalia, J. concurring.
1 The authors acknowledge the assistance of Douglas C. Dreier, a student at the Duke University School of Law and a 2012 summer associate at Hunton & Williams for his contribution to this article.
2 It is not discussed anywhere in the Guidance that criminal conviction standards will likely affect men significantly more adversely than women, or that research indicates that the use of background checks actually correlates with a higher level of hiring African-Americans. See Holzer, Raphael & Stoll, Perceived Criminality, Criminal Background Checks and the Racist Hiring Practices of Employers, 49 J.L. & Econ. 451 (2006).
3 See supra, n.2, Holzer et al.