Is Harvard Worth It? Conventional wisdom says yes. But with the price of a degree from America's most famous university and other elite private colleges now surpassing $125,000, many families--and a number of economists--aren't so sure. Here's a look at the evidence.
By Jeremy Kahn

(FORTUNE Magazine) – For America's high school seniors, April is the cruelest month. That's when colleges flood the postal system with news of who has won a place in next fall's freshman class. For more than a few families, a difficult decision will follow: Is it worth paying some $125,000 to give their child an education at an elite private college? Or would her future be just as bright if she went to a less expensive school?

These questions have no easy answers. Of course, that's not the impression you get from the $500-million-a-year college-admissions industry, with its magazine rankings, test prep courses, and guidebooks. Certainly many neurotic boomer parents--and their stressed-out, resume-building teenagers--assume that it is always better to choose Harvard over Big State U. because of Harvard's presumably superior educational environment, better alumni connections, and more lucrative on-campus recruiting opportunities.

It's true that big law firms, major teaching hospitals, and investment banks--heck, even the offices of FORTUNE--are stuffed with Ivy Leaguers. It's also true that if you want a career at what passes for the American establishment--Sullivan & Cromwell, McKinsey, Goldman Sachs--a gilt-edged diploma is a distinct advantage. Then again, there's plenty of anecdotal evidence that an elite education is hardly necessary. Steve Jobs dropped out of Reed. Jack Welch went to the University of Massachusetts, Warren Buffett to the University of Nebraska. The majority of top CEOs surveyed by FORTUNE in 1990 did not attend an elite college (though a disproportionate number did).

So what kind of return is there likely to be on that $125,000 investment? And how does it compare with the return on a less expensive but also less prestigious education? The academic evidence is murky. To start with the basics: College pays. On average, a person with an undergraduate degree now earns almost twice as much as someone with only a high school diploma, up from 1.5 times in 1975.

The economic literature on the payoff of graduating from an elite college, however, as opposed to any college, is far less conclusive. Several studies during the past decade found a connection between higher future earnings and attendance at a college with high SAT scores. Most of the research concluded that for each 100-point increase in the average SAT score, a graduate could expect a 3% to 7% increase in lifetime earnings.

But the studies compared schools, not people. You would expect graduates of selective schools--which attract successful students--to have successful careers. (It would be stunning if they didn't.) What such studies do not measure is how an individual's earnings are affected by the choice of college. If student X gets into, say, Amherst and Michigan State, and chooses to go to Michigan State, will X be shut out of Amherst-style earning potential? Other studies have tried to answer this question by looking at students with similar SAT scores who attended different kinds of schools. Researchers found that those who went to the more prestigious schools reported higher earnings.

But SAT scores are not everything. Admissions offices at elite schools include many other criteria in their decisions--grades, extracurricular activities, recommendations, essays, interviews. These factors may reveal abilities, like good communication skills, that are far more valuable in the workplace than a perfect 1600. Because economists have no data on these traits, they term them "unobserved." But they are hardly unimportant. Until recently, no one had tried to control for unobserved characteristics in measuring the effect of an elite education on earnings.

Then Alan Krueger, an economist at Princeton, and Stacy Berg Dale, a researcher at the Andrew W. Mellon Foundation, designed just such a study. In a widely publicized report, released by the National Bureau of Economic Research last year, they found no economic advantage in attending a selective college. Their research looked at the 1976 freshman class at 30 schools, ranging in selectivity (determined by average SAT scores) from Yale to Denison. The colleges were mostly private but included a few public universities: the University of Michigan, Ohio's Miami University, Penn State, and the University of North Carolina at Chapel Hill. Dale and Krueger compared the earnings of students who were admitted to the same colleges but made different choices. This ensured that they looked at similar individuals. In other words, because the students had been admitted to the same schools, they would have had equivalent SAT scores and "unobserved" traits.

Krueger and Dale concluded that smart, talented kids who attended less selective schools did just as well in their careers as their counterparts at elite colleges. There was no difference in average earnings. The same traits that made the students desirable candidates for admission to Yale--ambition, intelligence, wit--carried over to the workplace, where they were duly (and comparably) rewarded, even though they had turned down an elite education. Krueger says this is because the positive characteristics attributed to selective colleges were actually characteristics of the students, not of the schools. The advantages of Harvard, in other words, confer few benefits on the class slacker. Robert Zemsky, an education professor at the University of Pennsylvania, who believes a prestigious undergraduate degree does pay off, puts it this way: "It's like the brass ring on a merry-go-round. If you go to a high-priced, highly selective school, you have a better shot at the brass ring. But you have to grab it."

While Krueger and Dale found that college selectivity did not affect earnings, it did make a significant difference to those from poorer backgrounds. A 200-point increase in the average SAT score of the college attended resulted in 7% greater earnings for students from families in the lowest fifth of income distribution. Krueger attributes this to the ability to network with a critical mass of affluent students and alumni. "The kid from the inner city won't have contacts at Goldman Sachs," says Krueger. Children from wealthier backgrounds are more likely to have a grasp on the upper rungs of the labor market.

Krueger and Dale's research, while intriguing, is not definitive. Critics have questioned their methodology, the limited number and range of schools evaluated, and their conclusions. Even Krueger finds it odd that the results seem to show that while there is no correlation between college selectivity and future income (except for poorer students), the more a college costs, the higher the earnings of its graduates.

Caroline Hoxby, an economist at Harvard specializing in the economics of education, has done research that challenges the Krueger-Dale findings. She placed several hundred schools in eight ranks based on the SAT scores of their students. She looked at students who entered these colleges in 1960, 1972, and 1982, then examined their earnings at age 32. Hoxby controlled for SAT scores by comparing students with similar scores from different colleges; she did not control for unobserved characteristics as Krueger and Dale tried to. Although she notes that "some of the apparent returns to graduating from a more selective college may actually be attributable to the self-selection of students who have low earnings potential into less competitive colleges," she doesn't think this explains the whole difference in income between graduates of elite and less elite colleges.

Using 1997-98 tuition figures, Hoxby concluded that a student who gave up a full scholarship at a Rank Three private college (average SATs: 90th percentile in verbal, 86th in math) to pay full price at a Rank One selective college (average SATs: 96th percentile in verbal, 93rd in math) earned back the difference in cost 3.4 times over his lifetime. Those who moved from paying average tuition at a Rank Three public college to paying average tuition at a Rank One private school earned back the difference in cost more than 30 times over. That remarkable statistic demonstrates the very real gap between what happens in classrooms at, say, the University of Florida and MIT. (Hoxby did not compare the difference in moving from a top public college to a top private one.)

Her findings are complemented by other statistical evidence. In their 1995 book, The Winner-Take-All Society, economist Robert Frank and public-policy professor Philip Cook documented the increasing concentration of the nation's top students at a few elite colleges. They noted that in 1989 fewer than 1% of the 1.1 million students who took the verbal SAT scored above 700. Yet 43.8% of them graduated from one of the 33 schools designated "most competitive" by Barron's. The concentration of talented students has had a predictable result on where companies go to recruit employees. Frank found the companies that recruit at Cornell, where he is a professor, made half their recruiting visits to the top 25 schools. Among companies Frank termed elite--those that conducted 70% or more of their interviews at the top 25 schools, and were either the largest firm in their industry or appeared on a list of the best places to work--the numbers were even more skewed.

These figures might have changed recently, as students at elite colleges test the dot-com world. (John Katzman, founder of the Princeton Review college-prep test course, calls the Internet "the old-boy network on steroids.") This shift opens opportunities at old-economy companies for students further down the educational food chain. Lehman Brothers, Merrill Lynch, and Wells Fargo are just a few of the firms that have begun recruiting at places like Indiana University's Kelley School of Business because they can't hire enough MBAs from Wharton, Harvard, and other traditional feeder schools.

Without a rigorous, large-scale, longitudinal study of students who spurn elite schools for less highly regarded ones--and given the high and rising cost of college (see box), this research begs to be done--there will be no clear answer to our $125,000 question. What can be said is (1) An elite education gives students--especially less affluent ones--better access to certain kinds of elite jobs; (2) There is no economic advantage to choosing an expensive, mediocre private school over a top public one; (3) Talented students everywhere rise to the top. So listen up, high school seniors: Your life is not over if you end up at the University of Oklahoma. In fact, if you want to be governor of Oklahoma when you grow up, you're probably better off graduating a Sooner.

Determining the payoff of an elite education also depends on how you calculate the return. Most studies use future income, the favorite yardstick of those dead souls called economists. But knowing the cost-benefit ratio of a purchase is not the same as knowing its value. For many people, the value of a college education is in friendships made (or forgone) and new roads taken (or not). In this sense, the economic arguments may miss the point. What parents really want to know when they ask if it's worth $125,000 to send a child to Princeton or Georgetown is not Is it worth it for the average student with x grades and y economic background? They want to know whether it's worth it for their little darling. And that's a question economists can't answer.