Moderation in principle is always a vice

Old Thomas Sowell, when he's on a roll, he rolls. From The Perversity of Diversity by Thomas Sowell in which he reviews Mismatch and in which he discusses academic mismatch and affirmative action.

Read the whole thing for an extraordinarily succinct criticism of the ethos of affirmative action. Pithiness abounds.

Anyone who follows public policy issues can easily think of policies that help one group at the expense of some other group. What is rarer, however, is a policy that on net balance harms all groups concerned, even if in very different ways. Affirmative action policies in the academic world can claim that rare distinction.

As always with Sowell it all comes down to "But what are the facts?" Emphasis added:

The empirical data presented in Mismatch shows that black students admitted to colleges and universities where the other students have higher academic qualifications do not graduate as often, graduate with much lower grades, and, when they start out trying to major in difficult subjects like mathematics, the natural sciences, engineering, or economics, they end up majoring in much easier subjects with much less of a payoff in terms of their careers in later life. Moreover, black students with very similar academic qualifications who attend predominantly black colleges succeed in graduating with degrees in the natural sciences, mathematics, engineering, and economics far more often. Nor is this simply a matter of their being granted college degrees while having less knowledge of their subjects. Predominantly black colleges are 17 of the top 21 colleges whose black graduates go on to receive Ph.D.s in scientific, mathematical, and technical fields.

17 of 21 - Wow!

And these numbers are even more astounding.

In short, black and other minority students seem to learn better at institutions where the other students are similar in academic qualifications. The same conclusion is implied in data on what happened after affirmative action in admissions was outlawed in the University of California system. When racial preferences were banned by the voters in California, there were dire predictions that this would mean the virtual disappearance of black and Hispanic students from the University of California system. What in fact happened was a 2% decline in their enrollment in the University of California system as a whole, but an increase in the number of black and Hispanic students graduating, including an increase of 55% in the number graduating in four years and an increase of 63% in the number graduating in four years with a grade point average of 3.5 or higher.

Instead of the predicted drastic decline in enrollment in the system as a whole, there was a drastic redistribution of black and Hispanic students within the University of California system. Their enrollment dropped at the two most elite campuses, Berkeley and UCLA—by 42% at the former and 33% at the latter. But their enrollment rose by 22% at the Irvine campus, 18% at the Santa Cruz campus, and 65% at the University of California at Riverside. After this redistribution, the number of black and Hispanic students who graduated with degrees in science, mathematics, and engineering "rose by nearly 50 percent," according to Sander and Taylor. The number of doctorates earned by black and Hispanic students in the system rose by about 20%.

Again, improvements of 55%, 50% and 63% in the graduation rate, STEM enrolment, and in academic achievement? Wow! That is a phenomenal success. That sounds like a win all around.

The other thing I like about Sowell is that he does challenge sloppy thinking.

The authors of Mismatch draw policy conclusions from their work. The most obvious conclusion might seem to be that group preferences in academic admissions should be ended. But Professor Sander and Mr. Taylor see that as a virtually impossible thing to achieve, and indeed see a beneficial role for "a race-neutral system of smaller, socioeconomic preferences" focused on increasing the enrollment of people from lower income backgrounds, whether such people are black or white or whatever. They apparently see a role for thoughtful statesmanship toward that end by the Supreme Court, which is now considering academic affirmative action issues yet again, after having tried judicial statesmanship before, in earlier decades.

Preferential treatment based on individuals from poorer socioeconomic backgrounds? I am absolutely inclined that direction. But then Sowell reproves me.

I could not disagree more with the distinguished authors of this outstanding study. It was precisely by trying to be judicious social engineers and statesmanlike legislators that Supreme Court Justices have left affirmative action a bleeding sore on the body politic that will not heal, but which only produced polarizing bitterness on all sides. The time is long overdue for them to carry out their judicial function and recognize that the 14th Amendment means what it plainly says about "equal protection of the laws."

He is right, there is a sharp tug between logic and sense of fairness. Thomas Paine had something to say about that in The Rights of Man.

Moderation in temper is always a virtue; but moderation in principle is always a vice.