It’s hard not to feel sorry for the residents of Memphis, Tennessee. Depending on which source you consult, its violent crime rate hovers between three and four times the national average, and various publications describe it as one of the top 10 most violent cities in the United States. As we’ve commented previously, its fire department has been embroiled in Title VII litigation regarding promotion practices for over 30 years. And now its police department, whose primary task should be reducing that crime rate, is still fighting over its promotion practices after 40 years of litigation. This dispute, if a person, would now be old enough to qualify for protection under the ADEA.

All of these cases stem from the tests and procedures used to make decisions as to promotions.

Most recently, in Johnson v. City of Memphis, Case Nos. 13-5452/5454 (6th Cir. Oct. 27, 2014), the city created a promotional process for police officers in 1996 that was itself in response to what was then roughly 20 years of litigation over its promotional practices. With the assistance of an expert, it relied heavily on a video “high-fidelity” law enforcement role-play exercise (50 percent), with consideration of a written test (20 percent), performance evaluations (20 percent), and seniority (10 percent). 

The video test turned out to be inordinately time-consuming and expensive and was marred by “coaching” of some candidates. In 2000, it was replaced by a simpler “low-fidelity” test that had no role-playing. To make matters worse, the answers to the video test were leaked, causing the city to ditch the video test entirely. These changes prompted the next lawsuit that, not surprisingly, resulted in the invalidation of the process entirely. Back to the drawing board.

The city hired yet another expert who prepared a new procedure under the court’s review. The new test measured components such as “investigative logic,” job knowledge, grammar and clarity, and another low-fidelity test. The city added up to 10 points to the score for seniority pursuant to an agreement with the union. The results, however, had an unmistakable disparate impact on minorities. Whites were promoted at the rate of 73.3 percent and African-Americans at only 31.4 percent.

Given the clear disparate impact, the case focused on the validity of the promotional criteria and the existence of alternatives that would not have a disparate impact. The plaintiffs proposed alternatives such as reinstating the high-fidelity video test and adding a highly subjective factor of “integrity and conscientiousness.”

As an aside, the fact that testing was taking on a life of its own is itself problematic. The performance evaluations – perhaps the most important factor in the private sector – were never given more than 20 percent weight in the process. One can only wonder why officers were not being evaluated on the things that apparently mattered the most. Further, proposed factors such as “integrity and conscientiousness” are clearly make-weights and had they themselves resulted in a disparate impact they might have proven difficult to defend.

The district court tried the matter in 2005, and various delays prevented the case from reaching the court of appeals until 2010. Even then, the issue was limited to preliminary injunctive relief and the court of appeals expressed concern over the undue delay, directing that a final judgment be entered within six months. It wasn’t. The plaintiffs had to move for mandamus relief in 2013, and only then did the district court issue its final order, concluding that the 2002 process violated Title VII because of its disparate impact (undisputed) and the existence of less discriminatory alternatives (contested).

The Sixth Circuit’s opinion details the problems with the district court’s decision and ultimately reversed on the most significant issues. In summary, however, the problem was that the plaintiffs’ proposals could not be shown to be much better than what the city was already doing, and those proposals presented their own difficulties as proven by the past misconduct on the various video tests, excessive subjectivity, and apples-to-oranges comparisons of the results of various components. Ultimately, the court found that the city’s tests were reasonable and job-related and that the plaintiffs had failed to propose a sufficiently comprehensive process that would render any less discriminatory results.

The court reversed the finding that the program violated Title VII, and remanded on the issue of attorney fees.

This latest chapter proves the difficulties for both sides in Title VII disparate impact litigation. The city, for its part, hired experts to create the tests and the process to promote police officers, but still found itself embroiled in many years of litigation. The plaintiffs, for their part, spent over $1 million in fees, but still could not come up with a better alternative. Both sides used or proposed cumbersome processes to make what should have been routine promotion decisions, and probably half of an average police officer’s career passed before either side had a final decision from the courts.

The bottom line: Testing, although often well-intentioned and designed to select the best candidate, can itself be a source of costly and protracted class litigation.