Last year, we wrote about the EEOC’s then-new guidance on the use of criminal-background checks in hiring decisions. [“Using Conviction Records As A Screening Tool,” Retail Industry Update, June 2012]. In December 2012, the Commission issued a strategic enforcement plan that included targeting background checks as a barrier to employment of minorities. In June of this year, the Commission trumpeted the filing of lawsuits against Dollar General and BMW North America claiming their use of criminal convictions in hiring violates Title VII.
But these latest lawsuits were not the EEOC’s first attempt to challenge an employer’s alleged blanket use of criminal-background checks in hiring. In 2009, prior to the publication of the latest guidance, it sued Freeman Companies in federal court in South Carolina alleging that the manner in which Freeman used background checks had a disparate impact on minorities. Recently, the district court sent the EEOC packing with its tail between its legs.
Title VII prohibits both intentional discrimination and disparate-impact discrimination. The EEOC’s criminal-background check guidance and lawsuits over the use of them all fall under the disparate-impact theory. In this model, an employer does not have to intentionally discriminate to have liability for violating Title VII. Rather, the employer must only use a neutral policy or practice that screens out a disproportionate number of a particular protected class. And even if a neutral practice does have such a disparate impact, a business will not violate Title VII if it can prove that the neutral practice is job related and consistent with business necessity.
The EEOC holds out that the use of criminal-background checks in the hiring process can have a disparate impact on African Americans. The basis for this conclusion appears to be extrapolated from generic figures related to the conviction rates of African Americans and other minorities versus whites. Additionally, the EEOC has for years contended that blanket exclusions based on criminal convictions are not a business necessity relying on the 1975 decision in Green v. MoPac RR. In that case the U.S. Court of Appeals for the 8th Circuit noted, “[w]e cannot conceive of any business necessity that would automatically place every individual convicted of any offense, except a minor traffic offense, in the permanent ranks of the unemployed.”
Disparate-treatment cases rely on statistical evidence. In Green, the proof of the disparate impact came from a very rudimentary statistical analysis of the difference in rejection rates of white applicants and black applicants. Since using the rule excluded black applicants two and a half times for every one time a white applicant was rejected, the court concluded disparate impact had been proven.
The Failure Of Proof
News stories continue to abound about the significantly higher percentages of African Americans, particularly African-American males, convicted in the U.S. criminal-justice system. Because of this, for years, the EEOC’s position that the blanket use of conviction records to screen applicants will result in a disparate impact against African Americans has been unchallenged.
If simple numbers and percentages were used to prove disparate treatment, as they were in 1975, this theory might be provable. But statistical analyses have come a long way in the last 30 years. Today most office employees have more computing power on their desks than an entire business might have had in 1975. Courts require expert witnesses to take into account multiple variables in their analyses. Problems that appear on the surface, disappear as data is chopped and sliced in different ways.
The EEOC litigated and conducted discovery against Freeman for three years. At the end of that period, Freeman filed a motion for summary judgment on the most common grounds, that the EEOC could not prove its case.
The direction of the district court’s decision was prefaced in the first sentence when it noted, “For many employers, conducting a criminal history or credit record background check on a potential employee is a rational and legitimate component of a reasonable hiring process.” The court went on to explain, as any retailer knows, that there are numerous legitimate reasons to employ criminal-background checks in the hiring process and noted that the EEOC itself performs criminal-background checks on applicants for every one of its positions. In light of these facts, the court concluded that “a disparate impact case must be carefully focused on a specific practice with an evidentiary foundation showing that it has a disparate impact because of a prohibited factor.”
Under this standard, the court found the EEOC’s proof wanting in virtually every respect. The EEOC presented testimony of an expert witness, whose statistical analysis of data obtained from Freeman showed that Freeman’s use of criminal-background checks had a disparate impact on African-American males. The district court concluded that the database upon which the expert had based his report contained so many fallacies and errors that it rendered any conclusion based on that database unreliable.
Some of the problems stemmed from the expert’s failure to use a random sample of the data provided it by Freeman. Rather, the expert “cherry picked” the data for inclusion, suggesting he was manipulating the underlying data to reach a predetermined result.
Next, the court noted that the analysis did not address decisions made over the time frame at issue. Finally, the court pointed out that the expert had hand picked additional data to add to the database to further manipulate the results in favor of the EEOC’s position calling this “an egregious example of scientific dishonesty.”
The EEOC tried to save its case by relying on national data contained in the reports. The district court quickly rejected this attempt noting that the generic national data did not reflect the applicant pool. After noting several other flaws in the information, the district court excoriated the EEOC noting, “By bringing actions of this nature, the EEOC has placed many employers in the “Hobson’s choice” of ignoring criminal history and credit background, thus exposing themselves to potential liability for criminal and fraudulent acts committed by employees, on the one hand, or incurring the wrath of the EEOC for having utilized information deemed fundamental by most employers.”
The most compelling aspect of the decision is not that it broke any new ground on the issues of the use of statistics in disparate-impact cases or in the use of criminal-background checks. Rather, it was the district court’s concluding that the EEOC brought a massive case against an employer using an expert who cooked the books to prove a fact that was not true.
State Attorneys General Weigh In
Shortly before the district court’s Freeman decision came down, nine state attorneys general wrote a letter to the EEOC protesting its position on criminal-background checks. The letter was sent in response to the EEOC’s suits against Dollar General and BMW calling those actions “gross federal overreaching.” The letter went on to attack the premise of the EEOC’s position that business necessity of disqualifying someone because of criminal history can only be established through an individualized assessment.
The attorneys general pointed out that for some positions, such as police officers, any criminal conviction is a rational basis for disqualification. They also noted that adding discretionary factors to criminal-background screening creates more opportunity for race discrimination in the process than a nondiscretionary bright-line rule. Finally, they called out the EEOC’s criminal-background check agenda as an attempt to create a new protected category, convicted criminals.
The attorneys general had another compelling reason to challenge the EEOC. Many states have laws that prohibit hiring individuals with criminal history for certain state positions. The EEOC’s guidance purports to invalidate any such laws. While it is doubtful that the state attorneys general have grounds to sue the EEOC over its position, it will be interesting to watch whether the EEOC is willing to file suit against a state for refusing to hire an individual pursuant to one of these laws.
Relief On The Way?
At least one state legislature has now taken note of the Hobson’s choice referred to by the district court’s opinion. The Texas Legislature recently passed a bill to protect employers who are willing to give applicants with a criminal record a second chance. The new law, which takes effect Sept. 1, 2013, limits the liability of employers who hire applicants with a criminal record. The law provides that a “cause of action may not be brought against an employer, general contractor, premises owner, or other third party solely for negligently hiring or failing to adequately supervise an employee, based on evidence that the employee has been convicted of an offense." Whether other states will follow suit is unclear.
The EEOC’s agenda does not consider the potential liabilities and risks retailers face when hiring individuals with criminal histories. Under tort principles in many states, an employer’s failure to perform criminal-background checks is negligence. Retailers cannot suspend their processes waiting for clarity. Our advice: spend time on the front end to develop a background-check policy that will balance the risks of discrimination claims against the true business necessity.