27 May 2009

Supreme Court: Ricci v. DeStefano

The Objectivist
Affirmative Action Unmasked
Dunkirk-Fredonia Observer
May 25, 2009

This past April, the Supreme Court case heard Ricci v. DeStefano. This case is interesting because it highlights the tradeoffs involved in affirmative action and, outside of education, likely signals its slow death.

In 2003, the City of New Haven, Connecticut took steps to fill the captain and lieutenant positions in its fire department. New Haven’s law and regulations required that hiring and promotions be based solely on merit as determined by a competitive exam. New Haven hired a firm that specialized in employment-related tests, particularly ones relating to public safety. The test was approved by independent experts. 118 people took the test, including 27 black candidates. None of the black candidates did well enough to qualify for the 15 promotions, although one or two Hispanics did. On a side note, New Haven had already tried to mitigate the effects of the exam by giving the written exam get 60% of the weight to get promoted and the oral exam get 40%.

After the exam, the Civil Service Board would normally certify the list of those eligible for promotion. However, a local minister with close ties to the mayor made it clear that because of the racial disparity, he opposed certification. City officials tried to impugn the exam, but the firm that designed it stood by it and offered to perform a post-exam study that would have validated the test. The city blocked the study, almost undoubtedly because they knew the test would have been validated. They did so even though their contract with the firm required them to pay for it.

New Haven then refused to certify the exam on the basis that it would face lawsuits relating to Title VII of the 1964 Civil Rights Act. Some of the firefighters denied promotion sued the city and several individuals on the basis that it violated their rights under Title VII and the Equal Protection Clause. The Equal Protection Clause prevents state agencies from denying people equal protection under the law. Title VII prohibits employers from discriminating in hiring or employment conditions on the basis of race, ethnicity, sex, etc The District Court and the Second Circuit ruled against the firefighters. It held that New Haven’s concern about being sued was reason enough to throw out the test results.

As George Will points out, the lead plaintiff, Ricci, presents an appearance problem for New Haven. To prepare for the exam, Will points out, he quit his second job, bought more than $1,000 worth of books the city recommended, paid to have them put on audiotapes (he is dyslexic), and took practice tests and interviews. He studied hard, sometimes as much as 13 hours a day, and got the 6th highest score.

New Haven was concerned with part of the law that bans employment tests and practices that have an adverse impact on members of one race. In fact, the federal government (specifically, the Equal Employment Opportunity Commission) has guidelines that hold that evidence of adverse impact occurs when the selection or promotion rate for one group is less than 80% of the group with the highest rate. As Steve Sailor points out, this means that if 50% of whites pass a test, then 40% of more of the minority groups must pass the test. If this doesn’t happen, then unless it can validate the test, the employer will be found to have discriminated. If the employer validates the test, then it will not be found to have discriminated unless those suing can show that there was an equally valid and less discriminatory test.

The Supreme Court should have an easy time with this case. The Individual Rights Foundation, Cato Institute, and others point out that the city blocked the post-validation study because it would likely validate the test. The evidence that there was an alternative, less-discriminatory test is laughable. In oral argument before the Supreme Court, New Haven had to admit that their evidence for such a less discriminatory test rested on the claim made by a competitor firm who hadn’t even seen the test New Haven actually gave.

As a policy matter, the argument against affirmative action, and against blocking such exams, is straightforward. Such policies severely harm people and this harm is probably not outweighed by the policy’s benefits.

Psychology professor Kevin Murphy of Pennsylvania State University points out that general intelligence is the single best predictor of job performance, but also the one most likely to have a substantial adverse impact on several minority groups. Murphy and others further point out that once general intelligence has been measured, the testing of more specific abilities may well add little in predicting who will perform well.

To see this, consider the following insights from Steve Sailor. Sailor, points out that as a group, black students on average score at the 18th percentile of white students on the graduate admissions test and at 20% on the medical admissions test. To admit large number of black students, standards are dropped, which then produces frustrating results. Using 2004 data from Richard Sanders of UCLA Law School, Sailor points out that 53% of black law students who enter law school fail to become lawyers versus 24% of whites. 40% of black law school graduates never pass the bar versus 15% for whites. Similarly, a 1994 study of medical boards published it The Journal of American Medical Association found that only 44% of blacks passed the medical boards for the first time, versus 84% for whites and 79% for Asians. Note that none of these observations depends on whether the differences are genetic.

Hiring less meritorious people harms people. In one study by economist John Lott, between 1987 and 1990, various policies, including affirmative action, decreased white male officers by 6% (6,912) and increased the number of Black male officers by 950 (5%) in the 189 cities Lott studied. He concluded that these policies produced 1,145 more murders and 30 more rapes. Similarly, a 1998 study in The Journal of the American Medical Association, 71% of newly licensed physicians prescribed potentially inappropriate medication and inappropriate medication is the 6th leading cause of death in the U.S. Admitting less able medical students likely exacerbates this problem. A similar pattern is likely true of firefighters. When you hire less talented people in life-and-death jobs, some people who would otherwise escape injury, get hurt and killed.

University of Maryland law professor Sherrilyn Ifill points out that New Haven has a history of discriminating in its fire department. She points out that black firefighters have repeatedly sued the city for discrimination in hiring and promotion and won, most recently in 2004. However, New Haven didn’t even try to defend its policy in terms of compensating for past discrimination because it knew that one cannot compensate one person for what was done to another. Nor did it rest its argument on the value of diversity, a common justification in educational contexts. So it is not clear what reason it thought outweighed the increased risk to its citizens of hiring less competent firefighters. However, even if could have cited one, it is unlikely to outweigh the greater risks to its citizens.

New Haven’s weak legal argument and even weaker policy argument will likely result in the Supreme Court reinstating the test results. The case highlights the real tradeoffs in death and injury that results from affirmative action.

There is an important addendum to this case. The Obama administration backed New Haven. President Obama then nominated Sonia Sotomayor to be a Supreme Court Justice. She served on the appellate court that summarily dismissed the firefighters’ claim. Obama and Sotomayer will act to preserve race preferences no matter how shaky the case and in so doing are pitting some racial groups against others. Wherefore by their fruits, ye shall know them.


The Objectivist said...

Note it surprises me that IQ tests outpredict performance when compared to job-specific tests. Nevertheless this seems to be the general pattern and hence the more intelligence-focused the test, the more likely it predicts job performance.

When the job involves firefighting, medicine, or education, a less talented class of people can be expected to produce additional deaths, injuries, or worse education.

The counterbalancing benefits of diversity are merely speculative and hence should not be given much weight. This is even more likely because there are other speculative harms.

The Objectivist said...

Note if New Haven has a history of discriminating against African-Americans, then it should be made to pay damages to all African-American applicants, plaintiffs, or some other group.

This just redistributes wealth from wrongdoers to innocent parties, rather than to those discrminated against.

The Objectivist said...

Note Steve Sailor claims that the 40% attached to oral argument is an attempt to game the system rather than an attempt to pick out the best candidate. This is an interesting claim.

Anonymous said...

好秘书 我爱皮肤 中国公文网

Arizona foreclosures said...

You can do most of this in software for free. Try Jeskola Buzz. Its is a VST capable modular software music studio. Texas

jaso said...

Yes, but can you afford the massive financial and technical outlays implicit in the iPhone studio system?
payday loan

jaso said...

how amazing Bob Dylan is.

call center