ARB Resolves a 20 Year Dispute Involving Racially Motivated Hiring at Bank of America

In April 2016, the Department of Labor’s Administrative Review Board (ARB) settled a twenty-year dispute in Office of Federal Contract Compliance Programs, United States Department of Labor v. Bank of America.  Judge Luis A. Corchado authored the ARB opinion affirming an Administrative Law Judge’s (ALJ) August 2004 ruling regarding discriminatory hiring practices allegedly used by Bank of America (BOA) to exclude African American applicants in 1993. But relying on statistical analysis, Judge Corchado reversed the ALJ’s ruling that BOA utilized similar discriminatory hiring practices between 2002 and 2005. Judge Corchado’s opinion defined the contours of legally persuasive statistical analysis.The ARB cited a long-established principle regarding the role of statistical analysis in employment discrimination cases: “[S]tatistical evidence may be used to rule out chance.” Bank of America, ARB No. 13-099, LJ No. 1997-OFC-016 slip op. at 13 (ARB April 21, 2016)(Corchado L.). Courts have consistently considered disparities exceeding two standard deviations to be significant.  And the more extreme the statistical disparity, the less additional evidence a Plaintiff need present to prove that racial discrimination  caused the variation.

With this in mind, the ARB noted that the ALJ’s opinion regarding BOA’s 1993 hiring practices was based on the availability of a variety of statistically-based evidence:

(1) Multiple statistical analyses showing significant standard deviations occurred at multiple stages of the hiring process. Even when examining BOA’s own expert’s statistical analysis, the numbers “rarely” fell below the two standard deviations watermark. In other words, the deviation regarding BOA’s hiring practices as they applied to African American applications was rarely less than legally “significant.”

(2) The lack of standards for utilizing the RC (Credit Check Failure) Code. BOA was unable to offer any evidence or explanation to rebut OFCCP’s assertion that the RC disqualifying code fell more heavily on African American applicants.

(3) Unexplained disparate use of the RH (Work Schedule Conflict) code. BOA was again unable to offer an alternative explanation for why African American applicants’ availability to work so disparately disqualified them from positions at BOA.

The ARB contrasted this wealth of evidence against what was available to support OFCCP’s allegations that BOA continued similar discriminatory hiring practices between 2002 and 2005: a 4.0 statistical disparity that was applied to the four-year period as a whole. The ARB explained that “without more evidence, one bottom line standard deviation of 4.0 for four years with minor shortfalls in two of those years is not enough in this particular case to prove a pattern or practice of intentional racial discrimination.” The ARB thus could not affirm the ALJ’s ruling that BOA continued its discriminatory hiring practices in the second timeframe and reversed this portion of the ALJ’s decision.

The ARB’s decision fundamentally holds that racial disparity cannot be sufficiently proven with a singular statistical analysis demonstrating a legally significant deviation; rather, parties must utilize various studies and analysis.

Tagged: Administrative Review Board (ARB), Enforcement Bodies