Anyway, Emily Bazelon returns to my good graces with this in her recent post on the recent Ricci decision:
Why all the competitive storytelling? The Supreme Court's ruling today will apply whenever cities try to base promotions (or, by extension, hirings) on a procedure that turns out to eliminate most or all of the minority candidates. This is called disparate impact, and Congress wrote it into Title VII in 1991. New Haven's test for promoting firefighters had such a disparate impact—that's one thing all the justices do agree on. The question is, What happens next? Can the city say, hey, we think there's another better way to make these promotions that won't leave us with a fire department led only or mostly by white people; now let us go figure that out? Until today, the answer seemed to be yes. Now the answer is clearly no. If a city in this position sees a disparate impact problem coming toward it like a right, it can only step out of the way if it has, as Linda explains, a "strong basis in evidence" for thinking that the test can't be defended—in other words, that the test it used is more job-related, and a better measure of performances, than the other measures of assessment it didn't use.
The city can't throw out its test even though the results allow for the promotions of no black firefighters because it would be making such a decision based on race or, as Kennedy writes, because "the City rejected the test results because the higher scoring candidates were white." He is treating the decision to throw out the test and start over as an absolute racial preference. Here is where the bad facts come in: Frank Ricci and the other white (and one Hispanic) firefighters who sued did what they were supposed to do. They studied for the test the city offered, and they scored the highest. But they didn't get the promotions they felt entitled to because no black firefighters scored as well. Justice Ginsburg points out that Ricci and his fellow plaintiffs "had no vested right to promotion." She's right. But to the majority, that doesn't really matter, because the majority focuses only on this test and this round of promotions in New Haven. And when you frame the case that way, Frank Ricci and his dashed hopes take up the whole screen. At which point, you think about justice for them and only them.
Ginsburg widens the lens. She goes back to the early 1970s, when African-Americans and Hispanics made up 30 percent of New Haven's population and only 3.6 of the city's 502 firefighters. This is when the black firefighters in New Haven started suing. Their efforts yielded much better representation among the rank-and-file in the department. But as Ginsburg says, not among the fire department's leadership: "The senior officer ranks (captain and higher) are nine percent African-American and nine percent Hispanic. Only one of the Departments' 21 fire captains is African American." (More from me about the history of New Haven's fire department here.) New Haven threw out the test results because it was trying to rectify that imbalance. To Ginsburg, this should surely be permissible—a city trying voluntarily to comply with Congress' 1991 mandate to address disparate impact.
And then, Ginsburg points out, the majority raises the bar for what New Haven must show to justify throwing out these test results. Kennedy dismisses as "stray facts" in the record the doubts raised about how the test was weighted—60 percent written, 40 percent oral—and the city's proposal to replace the test with an assessment center, which are designed to evaluate the particular skills needed for a job. But Ginsburg sees that the 60-40 weighting simply reflects the demands of the union—the same union that filed its own suit against the city in support of Frank Ricci. And she sees the merit of the assessment centers as an alternative measure. "Relying heavily on written tests to select fire officers is a questionable practice, to say the least," she writes. "Successful fire officers, the City's description of the position makes clear, must have the '[a]bility to lead personnel effectively, maintain discipline, promote harmony, exercise sound judgment, and cooperate with other officials.' These qualities are not well measured by written tests." No wonder, in Ginsburg's view, a 1996 study found that two-thirds of the cities surveyed were using assessment centers in making promotion decisions.
They make a good point that I've been (inadvertently) dancing around with my fellow discussants in other fora: the (reasonable) sympathy evinced for Ricci and his fellow white (and one hispanic) firefighters in this case is in conflict with the institutional history and less acute but nonetheless real plight of black firefighters in New Haven. And I would differ with my friend EssEee (Sean) who sees institutional racism as less of a problem than institutional discrimination against poverty; I guess part of my counterexample would be the various problems women have with equality of management positions, salary, and promotion (see i.e. Lilly Ledbetter), where the problems of disparity and inequality for women is, if not completely distinct from poverty, relatively so.