The Worthless Ivy League?

We all "know" that going to college is essential for economic success. The more prestigious the college, the greater the success. It's better to attend Yale or Stanford than, say, Arizona State. People with the same raw abilities do better and earn more by graduating from an elite school. The bonus flows (it's said) from better connections, brighter "peers," tougher courses or superior professors. Among many parents, the terror that their children won't go to the "right" college has supported an explosion of guidebooks, counselors and tutoring companies to help students in the admissions race.

The trouble is that what everyone knows isn't true. Going to Harvard or Duke won't automatically produce a better job and higher pay. Graduates of these schools generally do well. But they do well because they're talented. Had they chosen colleges with lesser nameplates, they would (on average) have done just as well. The conclusion is that the Ivy League--a metaphor for all elite schools--has little comparative advantage. They may expose students to brilliant scholars and stimulating peers. But the schools don't make the students' success. Students create their own success; this makes the schools look good.

Evidence of this comes in a new study by Alan Krueger, an economist at Princeton, and Stacy Berg Dale, a researcher at the Andrew W. Mellon Foundation. Until now, scholarly studies had found that elite colleges lifted their graduates' incomes beyond their natural abilities. The bonus was about 3 percent to 7 percent for every 100 points of difference in SAT scores between schools. Suppose you go to Princeton and I go to Podunk; Princeton SAT scores average 100 points higher than Podunk's. After correcting for other influences (parents' income, race, gender, SAT scores, high-school rank), studies found that you would still earn a bit more. If I make $50,000, then you might make $53,500 (that's 7 percent).

But Dale and Krueger suspected that even this premium--not huge--might be a statistical quirk. The problem, they write, "is that students who attend more elite colleges may have greater earnings capacity regardless of where they attend school." Characteristics important for admission "may also be rewarded in the labor market." What might these be? Discipline. Imagination. Ambition. Perseverance. Maturity. Some exceptional ability. Admissions officers may detect these characteristics from interviews or course difficulty (different from grade average). But earlier studies didn't capture these factors.

To do so, Dale and Krueger examined the 1976 freshmen of 34 colleges. They ranged from Yale, Bryn Mawr and Swarthmore (highest in SAT scores) to Penn State and Denison University (lowest in scores). The SAT gap between top and bottom was about 200 points. Dale and Krueger knew which colleges had accepted and rejected these students as well as their future earnings. By 1995, male graduates with full-time jobs earned an average of $89,026; women earned $76,859.

Dale and Krueger then compared graduates who had been accepted and rejected by the same (or similar) colleges. The theory was that admissions officers were ranking personal qualities, from maturity to ambition. Students who fared similarly would possess similar strengths; then, Dale and Krueger compared the earnings of these students--regardless of where they went. There was no difference. Suppose that Princeton and Podunk accept you and me; but you go to Princeton and I go to Podunk. On average, we will still make the same. (The result held for blacks and whites, further weakening the case for race-based admission preferences. The only exception was poorer students, regardless of race; they gained slightly from an elite school.)