Attending an Ivy League school results in higher income

A few days ago I blogged about the NY Times article which gave very bad advice to prospective college students—that they should go to a less prestigious college that’s a better “fit.”

But in the comments to my post, people wondered if more prestigious schools really conferred any benefit.

A research paper by Stacy Berg Dale and Alan B. Krueger (1998) (link to pdf file) confirms that people who attended more prestigious schools earn more money. They looked only at students who were accepted to multiple colleges, so they were able to determine what happens if a student is accepted to a better school but attends a lesser school. From page 24:

Based on the straightforward regression results in column 1, men who attend the most competitive colleges [according to Barron's 1982 ratings] earn 23 percent more than men who attend very competitive colleges, other variables in the equation being equal.

23 percent is quite a bit of money, it’s almost like getting two college degrees instead of one!

They also discovered that there was a benefit to attending a more expensive school. The more expensive tuition resulted in a lifetime internal rate of return of 20% for men and 25% for women.

THE MOST MIS-CITED STUDY EVER?

Whenever this study has been cited, it has always been for the exact opposite of its actual conclusion. This typical article states that the study “dropped a bomb on the notion of elite-college attendance as essential to success later in life.”

A third finding of the study was that when colleges were rated based on average SAT score, students who attended a school with a lower average SAT score didn’t earn any less money. Everyone used this finding to say it doesn’t matter what school a student attends. But what it really says is that the average SAT score of a school is unimportant, what’s important is how highly “ranked” it is. I suspect that in many cases, when a student attended a school with a lower average SAT score, they did so because the school with the lower score was actually the more prestigious school.

This demonstrates a persistent bias in which the media only reports what people want to hear instead of reporting the truth. Parents want to know that they didn’t harm their kid by sending him to a state school instead of a more prestigious private school. Unfortunately, the reality is that sending your kid to a state school instead of the best private school he can get into does irreparable harm to his future career.

STUDENTS WITH HIGHER SAT SCORES EARN LESS MONEY

The regression analysis in the Dale & Krueger study had a coefficient for the person’s SAT score and a second for the square of the SAT score. Based on these two coefficients, earnings peaks at an SAT score of 1100. People who have an SAT score higher than 1100 earn less money.

I would find it hard to believe if I hadn’t discovered the same thing myself. Seeing the same result in a completely different dataset confirms that I didn’t do anything wrong.

It seems that the only benefit of high intelligence is that it gets you into a better college and graduate school. After you get your degrees, high intelligence is of no benefit in the labor market.

ATHLETES EARN MORE MONEY

The Dale & Krueger regression analysis also included a variable indicating if the person was an athlete. Those who were athletes earned more money. This also confirms my own findings from the General Social Survey.

CONCLUSION

A kid who gets accepted to Harvard because of his athletic ability, even though his SAT score is lower than most other Harvard students', and attends Harvard, will likely earn far more money over the course of his life than an unathletic kid with a perfect SAT score who attends a state school.

THOUGHTS ON THE IMPORTANCE OF COLLEGE

The more I study the subject, the more I come to the conclusion of the immense importance of a college degree and a graduate degree on a person’s future earnings. It’s almost as if a person’s whole life course is set when he walks out of college and begins working at his first job.

Comments

[bbartlog]

Actually, I'd qualify your conclusion and note that a *well-chosen* degree is an excellent investment. I'm also curious whether the returns to the 'investment' that are calculated for tuition include the opportunity cost of attending college...

[Peter]

One possibility: a higher percentage of the students who attend top-rated colleges come from affluent, well-connected families, and these connections are what lead to career success.

[Tim Lundeen]

Don't confuse average results with individual ones. Anyone can go to a lesser ranked college, or not go to college at all, and do very well. Still, the expected value for the higher-ranked college degree makes it a good investment.

BTW, with regard to the relationship between SAT scores (presumably a proxy for IQ or g) and income. I suspect that a lot of higher-IQ people are relatively high systematizers and have relatively low empathy. In order to be successful financially, you need to market your ideas and/or products, and work effectively with a team to make things happen. So the people who have a higher empathy level will do better on the business side, and they make more money than pure product guys (typically).

[mvpy]

I dont think the affluent, well-connectedness is the prime thing here; though, no doubt, thats a factor too. After all, whats the liklihood that daddy knows of my ideal job, in my field etc.
My view is that the Ivys open a door to the first great job; and thats prob the most important step - after that, you you're "in", so to speak. Noteworthy too is the fact the most paths of study are largely irrelevant to your career. Even in business oriented courses, its really the on the job training thats key. So, when you've got thousands upon thousands of students who can do the job, better to stay safe and go with school rankings. The schooling is just the signal.

Whats more, many students who attend these schools either get some kind of gov aid (either directly via state schools like Berkeley, or via gov loans, the NSF etc). My view is that state sponsorship for schools like this is a de facto subsidy to business; namely, since it makes their recruiting job far, far easier. This coupled with the fact that the middle class attend college makes the entire thing highly regressive.

[Dennis Mangan]

I think my failure to take typing in high school ruined me.

[tc]

Berg & Krueger have a more recent version of the paper in QJE:

http://www.davidson.edu/academic/economics/foley/324/Berg%20Dale_Krueger%20QJE%20(2002).pdf

It seems to be somewhat different in its conclusions. They don't seem to report the effects of the Barron's categories directly, but it sounds like they don't find an effect anymore. Higher tuition of the college does have an effect on future income.

[nobody]

Huh? Shouldn't this post be qualified by stating that the value of the degrees depends greatly on major as previous posts have pointed out. Without that preface, the conclusion drawn from post is misleading.

HS gets $65/hr doing ASP.NET not because he went to the "right school". There are plenty of people with Harvard degrees that don't make that kind of money.

There are two problem here.

1) Most people are ordinary. They lack the edge to distiguish themselves from the rest of the pack. This is a sorting problem that is solved by prestigiuos degrees but it does not apply to (A) famous people who have proven themselves in their field. (B) True talent.

If HS was the lawyer who successfully handle multiple history making cases, chances are that he gets a shot at securing a tenure professorship sometimes down the road at Harvard Law even if he graduated from Arizona State. The fame will get him in the door, and the talent lets him keep the job.

The truly talent is too hard to ignore. If you can consistantly throw a baseball at 110 mph, some major league scout will find you. If you know how prove the unified theory, Stanford/MIT/etc will hire you as full professor even if you never finished high school.

2) Most jobs are ordinary. They don't need the super high IQ employees because they don't add any value to the job. Perception is that many of these high IQ people think too highly of themselves and is hard to work with. Would you want a employee who constantly look down at you?

[spungen]

I think that people like to believe talent is more natural and inevitable than it really is. I call this the "Good Will Hunting" fallacy. In reality, whoever proves unified field theory will not only have graduated from an extremely good high school and college, his father and grandfather will have taught at MIT.

Most people with natural talent never refine it sufficiently to meet their potential because they lack the resources and guidance. A big part of becoming excellent is consistently observing excellence up close, IMHO.

[Half Sigma]

tc, thanks for the link, it seems that the newer verion of the paper is the older version of the paper with the finding inconsistent with what everyone wanted to read about were removed.

To everyone else who brings up college major, yes, I'm in complete agreement that major is extremely important.

[Rob]

I think this only applies to people who got in on merit. I got into Reed College for (geographic) diversity. I was not smart enough or hard-working enough for the school. I did take a serious major, chem, and then switched to biochem, and then failed out. After a few years of feeling too stupid to return to school, I went to a local, not-highly-rated-at-all school. Maybe only because I was more serious about school than my first try, I definately stood out more and had more opportunities than Reedies smarter than me, because they didn't stand out.

On SAT and income, maybe people are trading money for fulfillment?

[Superdestroyer]

HS,

I think "fit" is more important that you think. I will give you a case study.

A kid from suburban DC has 1400 on his SAT and good grades. The kid wants to go to medical school. The kid could apply for Georgetown and possibly get in. The kid will be in bottom half of his class and attending a school where over half the freshmen attended private schools. The kid will probably pay full tuition and probably not finish with a degree in science and could be on of the 10% who drop out instead of the 4% of Georgetown students who go to medical school.

http://www.princetonreview.com/college/research/profiles/admissions.asp?listing=1023791&ltid=1&intbucketid=

The same kid can easily be admitted to George Mason and probably get a Presidential Scholarship (free tuition and fees), get access to more professors and fewer TA's, special dorm space, and have a built in path to Virginia College of Medicine.
http://www.gmu.edu/departments/usp/vcumedicine.php

You would tell the kid to jump at
Georgetown University because of the name and probably finish with a degree in international studies and a huge debt load. I would tell the kid to take the George Mason deal and come out of college debt free and with admission to medical school.

[SFG]

"I would tell the kid to take the George Mason deal and come out of college debt free and with admission to medical school."

Med school's kind of a weird example due to the relatively high value of a nonprestigious medical degree. (compared to, say, law or business).

[Russ]

I would tell the kid to take the George Mason route aswell. Something happens psychologically to people when they are in the bottom of their class no matter how hard they try. They become lazy because the reward is out of reach.

[Peter]

Med school's kind of a weird example due to the relatively high value of a nonprestigious medical degree. (compared to, say, law or business).

It's not that weird of an example. School prestige is relatively unimportant for fields with good job opportunities - pharmacy, electrical engineering, nursing, special-education teaching, and I'm sure many others. School prestige is vitally important in law not because of anything unusual about law school, but for the simple reason that it's a glutted field.

Business school is hard to categorize because many MBA students are somewhat older people with post-college work experience, often attending on a part-time basis while working, rather than 21- or 22-year-olds who go straight through from college.

[Half Sigma]

Notice that, with the exception of medicine, the non-glutted fields cited by Peter tend to be less presigious.

[dearieme]

"the relatively high value of a nonprestigious medical degree": fasacinating, this may be international. I've heard my daughter's friends discussing universities in Britain. For the Humanities, you should go to Oxford or Cambridge - not just more prestige, but far higher standards and more work expected - but for medicine it matters much less - all medical schools are viewed as pretty good, with high standards.

[Peter]

Notice that, with the exception of medicine, the non-glutted fields cited by Peter tend to be less presigious.

Does it matter? I'd rather make $60K as a "low prestige" nurse or pharmacist than $30K as a "high prestige" journalist.

[Half Sigma]

"Does it matter?"

Just an interesting observation. It surely matters to all the people who eschew practical lower prestige college majors.

[jult52]

Half Sigma: In the 2002 version of the Krueger/Dale paper which tc linked to, the authors state that "the effect on earnings of the average SAT score of the school the student attended is indistinguishable from zero" (top of p. 21 of the .pdf file). They sure seems to be endorsing the popular interpretation of their paper which you criticize. Also, see their conclusion on p. 33 of the .pdf file.

How do you respond?

(BTW, the link to the original paper in your original post did not work for me.)

[SFG]

"the effect on earnings of the average SAT score of the school the student attended is indistinguishable from zero"

Yup, that's exactly what HS said! SAT score doesn't matter, prestige does. If Harvard's SAT score dropped to 400 tomorrow, it would still be worth going there. Or, as in that old chestnut, it ain't what you know, it's who you know.

[tc]

I wonder how "nonlinear" the effects are. I can understand why everyone would want to go to Harvard if possible, but how many other places have as big a "name" as Harvard? Maybe a good state school would be comparable to a not-quite Ivy League private school.

[jult52]

SFG: But see p. 19 (2nd full paragraph of 2002 version), where the authors use the Barron's selectivity criteria. This paragraph is dense and not clearly written but it indicates to me that selectivity rankings different from SAT scores does not affect the result or the conclusion.

tc: good question. I don't like the constant references to Harvard either. Let's start using Haverford instead.

[superdestroyer]

Jult52

http://www.princetonreview.com/college/research/profiles/extracurriculars.asp?listing=1023306&ltid=1&intbucketid=

I looked up Haverford. 80% graduation rate (I guess 20% did not fit), top majors are History and English but only 3% to Law School. No real science, technology, or engineering. Looks like about a 1400 SAT score school. Tuition at $31,000 and also has no greek organizations. Looks like the kind of place that political interns come from but does not seem the place to become BIGLAW or InvestB. Might be the place if you want to become a college professor down the road or a Wonk.

However, I would tell virtually anyone to not major in history or english at a state university unless they are going into private school teaching.

[spungen]

Peter, one quibble: A "high prestige journalist", like say an NYT reporter, makes more than $60K. There just aren't many of them. The people in those jobs went to either Ivies, or non-Ivy Eastern and midwestern private schools known for J-programs, like Northwestern. The $30k reporters, the unwashed 95 percent, did not.

Whether it's a glutted field depends where you come from. The Ivy person has a good shot at that top 5 percent; the Podunk State person, virtually none. Only the Podunk State student is better off in nursing or K-12 teaching. Assuming equal pay, most people would prefer NYT reporting.

[Peter]

I looked up Haverford. 80% graduation rate (I guess 20% did not fit), top majors are History and English but only 3% to Law School. No real science, technology, or engineering. Looks like about a 1400 SAT score school. Tuition at $31,000 and also has no greek organizations. Looks like the kind of place that political interns come from but does not seem the place to become BIGLAW or InvestB. Might be the place if you want to become a college professor down the road or a Wonk.

The college where I attended, Trinity in Connecticut, was similar to Haverford in many respects. It has long been popular among pre-meds, despite its reputation as a liberal arts college and the absence of a formal pre-med program. Medical school admissions committees supposedly look favorably upon applicants from smaller liberal arts college. In fact, as of about 15 years ago - I have no idea if it's changed - Trinity had a higher medical school acceptance rate than Yale!
It wouldn't surprise me, therefore, if Haverford attracts a number of pre-meds. Note that most colleges will offer enough science courses to satisfy pre-med requirements even if they're not science-oriented institutions.

[Superdestroyer]

peter,

Pharmacist can easily make much more than $60K. Starting pay at Wal-Greens, Rite-Aid, CVS would probably start around $80K and if you are willing to live in a small town (the $30K journalist kind of town), a pharmacist can make over $100K.

Nurses can also make much more by specializing into things likeNurse Anesthesist and make $100K.

I think the problem with comparing careers is that some careers have a normal distribution of wages (nurses, pharmacist, engineers) and some career fields (lawyers, journalist( have a log-normal distribution of wages. A full time pharmacist is never going to make millions but they can all find work and the pay is good without being great. A lawyer can make millions but many never find work and many do not make that much.

[Half Sigma]

"The Ivy person has a good shot at that top 5 percent; the Podunk State person, virtually none."

Because of an optimism bias, 90% of students think they're going to make the top 5% of their chosen profession.

[spungen]

Optimism, outdated information, unrealistic fictional portrayals, and flat-out lying by professors and flaks who profit off misguiding the young.

[Peter]

I think the problem with comparing careers is that some careers have a normal distribution of wages (nurses, pharmacist, engineers) and some career fields (lawyers, journalist( have a log-normal distribution of wages. A full time pharmacist is never going to make millions but they can all find work and the pay is good without being great. A lawyer can make millions but many never find work and many do not make that much.

Which brings up the question of why the careers with normal wage distributions aren't considered as prestigious as those with log-normal distributions. My guess is that people who choose the former type are seen as playing it safe, unwilling to take risks, while those who go into log-normal careers are admired for their courage. Of course, it's also likely that many people don't realize just how risky high-prestiage careers can be, for example there's still a popular perception that going to law school is a sure ticket to affluence.

[Peter]

Optimism, outdated information, unrealistic fictional portrayals, and flat-out lying by professors and flaks who profit off misguiding the young.

Hmmm, how about a realistic fictional portrayal ... here's a idea, let's have a TV series about some poor schmuck who's five years out of (a nonprestigious) law school and is trying desparately to eke out a lower-middle-class living while paying off student loans that'll choke an elephant? Somehow, I don't think the TV networks will be interested.

[spungen]

*tsk* Peter, I wasn't clamoring for shows about the dreary, homely and poor. Just pointing out that American fiction confuses kids by portraying people who start in a disadvantaged position, and work their way up to the big time. Makes for a comforting backstory. But in real life, the successful Madison Avenue exec does not have coal-miner parents, or even government worker parents, and we should not give their kids the idea that if they study hard, behave, and are plucky, they can take Manhattan. Or even live in Manhattan.

The unwritten rule in reporting is that every personality profile must include something a successful person "overcame." Sometimes the contortions are painful, but it's what people want to hear. To the point where Bill Gates gets held up as an example of an underdog dropout near-loser who turned his life around. No wonder there's an optimism bias.

[spungen]

Peter, many people don't even know about the good jobs with normal wage distribution, unless someone in their family does it. I had no idea how much pharmacists make.

Most people don't think of law as risky at all, but a stable boring way to make money. Most people don't even think of journalism as risky, just as vaguely lower-paid until you "work your way up."

The main reason people don't want to be nurses or special ed teachers is probably not lack of prestige, but not everyone has what it takes to care for the sick or troubled.

I also suspect guidance counselors and schoolteachers are a big part of the problem, because kids get their career ideas from them. They only know about a handful of jobs (mainly the "helping professions"), except what they see on TV, and nothing about business.

[Jack]

The optimal SAT is 1100?? This score is way too low for Ivies, unless you are a stud hockey or football player, and likely even then. If Ivy League grads make more than state school grads, shouldn't the optimal SAT be like 1400? Almost no one goes to an Ivy with an 1100, or even a 1250. Although if the 1100 is before recentering, it could be similar to an 1150-1200 now. As a high-IQ person in a job which I am overly bright for, I am curious what kind of jobs HS would recommend for the very bright. I was thinking law, but, well, you know...

[Douglas Knight]

nobody,
Most jobs are ordinary. They don't need the super high IQ employees because they don't add any value to the job.

In almost all jobs, IQ predicts performance better than anything else, often better than tests designed for the job. Of course there are diminishing returns.

[anonradguy]

"As a high-IQ person in a job which I am overly bright for, I am curious what kind of jobs HS would recommend for the very bright. I was thinking law, but, well, you know..."

Study hard for the LSAT and see where you get in. If it is not a top 14 law school, don't go.

If you don't mind working really hard for 10 years, you could try for medical school.

There is the personality issue, which hits you at the point of med school admissions and (I assume) law firm job interviewss. I was unable to get into a top-tier med school despite being junior phi beta kappa from an ivy, but did get into a mediocre one. I then worked hard again, got AOA (the med school equiv. of phi beta kappa...about top 15%, probably =coif), good medical board score, published a paper or two ...and wound up with a mediocre radiology residency!

But I'll still probably be able to pull $200K...if they don't ship my job to India. You could still try for anesthesia or one of the surgical subspecialties that are harder to outsource, you can crack 200K in one of those if you can sneak in. Something to think about. Good money, even if interviewing at Hopkins and winding up at a state school, TWICE, is kind of demoralizing.

If you are in fact charming, you might try for business, but be aware your high IQ is not going to get you anywhere.

[michael vassar]

I disagree strongly about high IQ not helping in business. It won't help you to get in, get customers, or stay in, but it will help you to recognize opportunities well, manage well, etc.

[jult52]

Half-Sigma: I'm hoping you'll get back to me on my very specific comments on the interpretation of the Dale/Krueger paper. The paper is dense and complicated but it looks to me like you've misinterpreted it.

[Steve Clymer]

Half-Sigma:

Apparently one of the people who misinterpreted the study was the author himself. Here's the opening bit of an article he wrote for the NYT in April, 2000, stating in quite clear terms exactly the popular impression that attribute to the media.

In fact, the accompanying chart in the Times actually shows that selecting a "Highly Selective" school over a "Moderately Selective" school results in an income PENALTY of $1,100.

What gives? Is Krueger misinterpreting his own study?

-Steve Clymer

http://partners.nytimes.com/library/financial/columns/042700grads-econoscene.html#offsite
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
New York Times April 27, 2000
Better Pay for a Better College? Not Really
By ALAN B. KRUEGER

Your son or daughter has just been accepted to both the University of Pennsylvania and to Penn State. The deadline for decision is May 1. Where should he or she go?

Many factors should be considered, of course, but lots of parents and students are particularly interested in the potential economic payoff from higher education. Until recently, there was a consensus among economists that students who attend more selective colleges -- ones with tougher admissions standards -- land better paying jobs as a result. Having smart, motivated classmates and a prestigious degree were thought to enhance learning and give students access to job networks.

But is it true?

A study that I conducted with Stacy Dale of the Andrew W. Mellon Foundation, "Estimating the Payoff to Attending a More Selective College" (available online at http:// papers.nber.org/), has unintentionally undermined this consensus.
[...]

[Half Sigma]

In the NY Times article, he writes "College selectivity is based on the average College Board score of the freshman class."

He is cherry picking the means to determine the rating of the college in order to get the results that everyone wants to read about.

[jult52]

Half-Sigma:

1) You're ignoring the point I made in my Aug 4 12:50 post. Please respond. If I am right, you're misinterpreting the study.

2) Does anyone really believe that a college's prestige exists independent of the avg SAT score of its entering student?

[Steve Clymer]

So you do think the author is misinterpreting his own study! Interesting, I don't think so.

In fact, it appears to me that the study actually uses all sorts of convoluted syntax and selective emphasis to avoid the obvious conclusion detailed in the NYT article. For example, it highlights the advantage of selective schools for students with low parental income with no mention of the corresponding penalty that must exist for students with high parental income.

In any case, you can hardly fault the media for misinterpreting the study if the author is doing the same.

-Steve

[Half Sigma]

I have to confess it could take me hours to figure out everything Krueger is talking about, but it seems relatively clear that in the later paper he simply left out the finding that prestige, independent of average SAT, is correlated with future earnings.

A person is better off going to a school with the same prestige but lower average SAT because that person will get higher class rank against easier competition.

[Half Sigma]

And yes, bogus schools with high average SAT scores like Illinois Institute of Technology or Davidson College won't help you as much as a more prestigious school.

[Rob]

Why are they bogus schools? I've often thought that a school could climb crazy high by giving lots of very bright kids scholarships and forcing everyone to play sports.

Have any schools tried something like that?

[Li]

I can't believe that many years after graduating from college and understanding that people take various paying jobs for a variety of reasons, you all are still arguing about colleges and SAT scores. There are so many fields to make good money in, outside of your obsession with credentialism, that I am having a hard time believing you live in the same country I do.

I also think you should resign from the Half Sigma Club. It might do you some good by lowering your hat size and perhaps put a dent in your aforementioned devotion to credentialism. Perhaps you should also get a substantive hobby, and interact with those whose talents are substantial, real, and unrecognized through standardized testing and college graduate programs, like musicians, singers, actors, painters, novelists and playrights, golfers, yachtsmen, etc. I think you get the picture.

Or you can continue to compare diplomas and 1978 SAT scores. Have a great week!

[superdestroyer]

Li,

I believe that HS really started the credentialism idea when he lived in Washington, DC. If you look at Money Magazine's most educated cities, you find Arlington, Virginia at the top of the list. In a place like DC, if you do not have credentials, you cannot find a job.

[jult52]

Li: You're completely missing the point. A lot of readers are actively thinking about whether it's worth it to pay in excess of $40k annually for a "prestige degree" for their children. This paper from Krueger and Dale is a very interesting piece of evidence in this debate and is worth understanding correctly.

[SFG]

"Why are they bogus schools? I've often thought that a school could climb crazy high by giving lots of very bright kids scholarships and forcing everyone to play sports.
Have any schools tried something like that?"

Dunno about the sports, but as I recall the U of Chicago didn't engage in the anti-semitic quotas of the 20s and wound up with a lot of Nobel laureates and a steep decline in prestige.

Still, they're considered top-tier nowadays, so it must have kind of worked.

[spungen]

"interact with those whose talents are substantial, real, and unrecognized through standardized testing and college graduate programs, like musicians, singers, actors, painters, novelists and playrights, golfers, yachtsmen, etc."

Yeah, if I were in a position to hang with "yachtsmen" I might not give a hoot about inequality, either.

As for the rest, I know too damn many -- all parental leeches. Especially the "actors ... novelists and play[w]rights." Yeesh, they're what got me pissed in the first place.

[The Engineer]

Why is IIT a "bogus" school? Because people thing that you went to the India Institute of Technology when you say that you went to IIT?

I'm an IIT grad (the Illinois one). I might as well have gone to the one in India. All the students and most of the profs were Indian.

Still curious as to why IIT is "bogus". It is not a high prestige school, that's for sure. I got a graduate degree there because they have the best distance learning program I've found. It was easy to get a graduate degree in engineering from there.