Should Harvard admit more legacy students?

That is the question behind my latest Bloomberg column, and basically the answer is yes.  Schools could scale up, using legacy admissions as a further source of finance.  But why don’t they?:

So why don’t top schools do more to expand their reach? No one doubts that they could find many more qualified students to admit. But there are two problems, both of which we should be willing to live with. First, expanding the size of top schools would lower faculty standards on the research side. That said, teaching quality is unlikely to suffer, as Harvard doesn’t select for the very best teachers. In any case, Harvard’s best researchers could continue their highly productive efforts without missing a beat. Second, administrators would face headaches and potential reputational liabilities from the new initiatives. But that is true in any kind of startup endeavor, and it isn’t a reason to remain stuck in the past.

The actual constraint on how big top schools could grow is how many eligible donors they can find and cultivate, if only through admitting their children. One question is how many such donors there are period, but in an age of high income inequality it seems America’s top schools have hardly tapped out this pool. Legacies make up a sixth of undergraduates at the University of Pennsylvania. A more unfortunate reality is that some donors might limit their support if say Princeton offered them and their children a less tony and exclusive experience. If that attitude can be overcome, America’s top schools could grow a great deal larger and more diverse.

Do read the whole thing.


This article doesn't even mention how legacy admits expand the network of opportunities for all students. The universities could use extra funding from alumni parents, but the presence of legacies in the school helps many of the meritocratic-ally admitted students create or capture enough wealth to become large donors.

if they are looking purely for more money why not just auction off 10% of the class seats and save the trouble of begging for donations after admissions? Chinese princelings, Saudi Royalty, various despot's children, the rich and stupid from all over the world could finance Harvard's expansion into Allston. They could buy up everything all the way to Watertown mall and make bank.

I sort of agree with this. I think the auction method would be a bit too tawdry and probably not optimal. Perhaps the DeBeers model makes the most sense. DeBeers makes a selection of diamonds they think is appropriate for you, and you are offered this "sight" for a certain price. Take it or leave it -- no negotiation. If you leave it, you won't be invited back.

This way, they could be highly individualized. This guy could almost make it through regular admissions, so he wouldn't drag down our standards. His parents are a couple of doctors, so let's set it at $5M. On the other hand, this other guy would struggle to make a C average, but his dad is a Russian oligarch who wouldn't blink at $50M. Let's set this one at the full $50M.

Interesting. But once most H students are Russian oligarch children, what is the value of a H diploma?

Who said most? Andre was saying 10%. I think that's a bit high. How many Russian oligarch's children are there? 20 a year at $50M a pop would be serious money that even Harvard would notice. You could free ride all the diversity students for that kind of money.

To what extent is that going on now? I feel certain that if you have enough money, you can find a way to get your mediocre student into Harvard.

The value of the Harvard degree is partly about rational astrology (nobody ever got fired for hiring a Harvard grad) and partly because Harvard students tend to be really smart and driven. That means you can let a few not-so-impressive students in, without undermining the whole school's reputation too much. So you get legacy kids and affirmative action kids and athletes and probably all kinds of other groups. (I don't know, but I suspect that the children/grandchildren of very powerful people are more likely to get in than comparable middle-class children.)

But for this to work, you kind-of need to pretend it's not happening. If it's clear that Joe only got into Harvard because his dad built the new chemistry lab, or that Jane only got in to make the racial numbers look better, then there's an incentive for the outside world to discount those peoples' Harvard credential.

'but in an age of high income inequality it seems America’s top schools have hardly tapped out this pool'

To think that before the age of high income inequality, many of America's best universities did not rely on donors.

They should more students period, not perpetuate more inequality by admitting more legacy students at the expense of first-generation college students.

The argument, though, is that it's not "at the expense of" the non-rich students. The additional legacy students would pay for themselves and even allow elite institutions to subsidize more non-rich students.

This argument is open to criticism, including that it is further privileging an elite group. But if you want Harvard et. al. To admit more students, a serious proposal needs to consider how to pay for that.

I think certain merits of this are being filtered out. Elite schools haven't grown in decades for a reason: clearly an instinctual sense of the ideal size to act as a soft aristocratic device. Schools are actually competitive to remain small because an ivy league that doubled in size would lose their value to the wealthy and actually lose both top donators and some motivation to donate among the rest.

Not only does every legacy added above the current size reduce the elite signal among legacies as their own average scholastic merit ticks down, it also reduces it by enabling multiple more deserving non rich students whose own average merit won't budge - so it's a double hit to the aristocratic signal.
I'm personally way more interested by the idea of legacy donations creating opportunities for others of varied backgrounds while also funding elite research but the point for those more focused on the advantage to the soft aristocracy is that these schools don't grow in size for a reason. The new legacies would benefit while the old ones would not, except in realizing a greater sharing of the wealth and power provided in their schools to the society they live in and also benefit from.

"A more unfortunate reality is that some donors might limit their support if say Princeton offered them and their children a less tony and exclusive experience."

This point seems under-emphasized to me. Isn't this the main issue? Top schools' "customers" are alumni donors, not students. Making schools less exclusive runs counter to the interests of alumni.

'Top schools’ “customers” are alumni donors, not students': an interesting insight. Will the current antics of spineless or deluded administrators change that attitude? Will rich men choose to donate to Princesston? Except, obviously, for buying a place for their unmeritorious offspring.

Princeton: Mr. Dearieme? I called because your daughter is having some difficulties.
Dad: Oh?
Princeton: Yes. She had a breakdown after getting marks on a rather minor exam. She did ok, but then cried for three days, accused the professor of sexism, and is now carrying a stuffed panda around as a mark of solidarity with the oppressed.
Dad: Yes, she likes her stuffed toys.
Princeton: She also has accused half the male population on campus of rape, threw a bag of urine at an administrator.
Dad: I love her expressiveness.
Princeton: Her behavior is getting destructive and distracting to other students. We are suggesting that she go home for a couple months and come back when she is ready.
Dad: Oh.
Princeton: Could we arrange for you to pick her up.
Dad: Hmm. Since you are on the phone, you might be able to answer a question. What would be the most suitable amount for a donation? $1 million or two? And who would I write the check to?
Dad: Are you there?
Princeton: Yes, yes. Two would be fine. I think we might have found a group that would help your daughter fit in.

Elite colleges don't compete that hard for top students. They used to have a formal cartel called the Overlap Group to adjudicate which college got which applicant. From the New York Times:

Published: September 3, 1992

A Federal judge ruled yesterday that the Massachusetts Institute of Technology had violated antitrust laws by participating in a decades-long conspiracy with other elite universities to fix the amount of student financial aid packages.

Louis C. Bechtle, Chief Judge of the Federal District Court for the Eastern District of Pennsylvania, clearly and pointedly rejected M.I.T.'s argument that it was engaged in charity, not conspiracy, when it shared confidential financial information on students with other universities.

In a ruling that punctured the prestige of some of America's most respected academic institutions, Judge Bechtle called the activities of the Overlap Group -- the name used by M.I.T. and the eight Ivy League colleges and universities -- "plainly anti-competitive." ...

In May 1991, to avoid a trial, the eight Ivy League institutions signed a consent decree with the Government, agreeing to discontinue sharing information about prospective students to determine how much they and their families should pay for education. M.I.T. chose to fight the case in court.

In his ruling, Judge Bechtle wrote: "M.I.T.'s attempt to disassociate the Overlap process from the commercial aspects of higher education is pure sophistry. No reasonable person could conclude that the Ivy Overlap agreements did not suppress competition."

Although I personally agreed with Bechtle, his decision was reversed on appeal, and the consent decree signed by the other Ivies was modified.

This approach makes sense only if a few conditions are met, most of which seem unlikely. Counterintuitive ideas are interesting, but sometimes they're counterintuitive because they don't have much chance of success.

1) Remove the few schools that already have giant endowments. They have enough money to expand admissions right now if they wanted. This includes Harvard, Stanford and at least a couple others.

2) Competitive schools that could actually use the money probably don't want to expand admissions because it undermines the perceived exclusivity of the degree, regardless of whether the quality of education actually changes.

3) Schools would have to be confident this would increase donations in the near-term. A lot of wealthy alums already give their alma maters lots of money, either because they want their names on stuff (even just a newsletter) or they think they have nowhere better to put it. Would a marginal increase in the chance their kids are accepted to the school really move the needle much? I'll grant that as a long-term strategy this could help, once the new graduates age into giving stages of life. But in that case why not just select for rich kids in general, regardless of legacy status? Maybe a little gauche, but we're talking money here.

You'll notice that the scoring of college entrance exams has gotten easier. Back in 1987, only 9 high school students in America scored a perfect 1600 on the SAT:

Nowadays, hundreds of students score 1600, and thousands score within one or two errors of a perfect score.

The ultra-elite colleges like today's easier scoring because they don't have to compete as much over top students because it's harder to identify the true superstars when thousands of students score 1500 out of 1600 or higher. So HYPSM can admit legacies, donors' and celebrities' children and athletes, without getting into costly battles over the very best students.

A few years ago, Steven Pinker suggested that his college, Harvard, should specialize in admitting the absolute top of the top students in term of academic ability. But his recommendation disappeared almost without a trace because the top colleges have a very comfortable cartel at present.

Here's Pinker's article:

The Trouble With Harvard
The Ivy League is broken and only standardized tests can fix it

"The ultra-elite colleges like today’s easier scoring because they don’t have to compete as much over top students because it’s harder to identify the true superstars when thousands of students score 1500 out of 1600 or higher. "

Evidence? There's plenty of evidence that standardized tests identify better and worse students on average. 100 students with a 1200 average will do worse than 100 students with a 1500 average, for example. But is there evidence that doubling down on high test scores identifies the best of the best? Is 1600 really all that better than 1550?

To use an analogy, if you're recruiting for a football team standardized metrics of athletic ability would be things like running speed, endurance, muscle mass, max. weight a player can lift, etc. But you are using these as proxies for field performance, in actual play the guy who can run fast, hit with a lot of force and catch may end up performing horribly. Given two players who can, say, run very fast it is not a given the faster of the two will make for the better QB. At some point the simple metric that is easy to measure will see diminishing marginal returns. If you're trying to recruit the next Einstein, he might have a 1550 on his application as much as he might have a 1600.

Of course another problem is your assuming Havard is honest and correct that they just want the best of the best. I'm sure they'd be very happy to have more Nobel Prize winners in their alumni but economically I see a powerful counter motivation. Consider a rich person with a lackluster kid (Jared Kushner, Trump Jr. both fit that bill). Imagine if Harvard could demonstrate that they can take lackluster kids and after they graduate they nevertheless go onto powerful positions and high paying jobs. Guess what, rich people would pay a lot for that. So on another level Harvard would actually want some measure of stupid people and demonstrate that giving them the degree plus the alumni connections immunizes them from their stupidity. Recall the Emperor's New Clothes, the tailor didn't score the job by demonstrating superior craftsmanship.

Is 1600 really all that better than 1550?

I'm not sure how you missed an obvious point so badly. When 9 people score 1600, that's showing, basically, the peak of ability. When hundreds of people are there and thousands could have been there with one or two more answers correct, the test is simply too easy to be able to determine who is at the highest level of ability.

This is both economically and statistically significant. One student's 1550 is not very much different from a single student's 1600. But having a hundred people at 1600 matters a lot because it means that many of those are indistinguishable from scores just under 1600. You create a larger probability blob of top scoring students, hence debasing your enrollment. But for very top students the Ivies already use other methods of detecting the potential geniuses such as winning a national or international Math Olympiad. However it gives them cover for admitting lots more 1500s who are really 1450s just because they are legacy, Hollywood, sports stars, or AA.

"the test is simply too easy to be able to determine who is at the highest level of ability."

Here is where I am raising an objection. I agree if 9 people show up with a score of 1600 then we can use that test to narrow the entire population down into single digits of top test takers. If 100 people show up at 1600 then if you really want to get down to the top ten you should add more questions or make some questions harder or modify the test in some other way so the very top is occupied by less than ten people.

Here's my point. This test may be very if you're trying to filter your list of people to consider down from one million. Saying, for example, you will only consider people with a score of 1400 or higher may eliminate 950,000 applicants with a low cost of false negative (in other words the kid who will do great at your school but his score was 1399). What happens when you push the test harder? if now you say 50,000 people is too many to consider so maybe raising the bar to 1500 will reduce that down to just 5,000 people. But odds are you will be doing that at a higher rate of false negatives in the 45,000 people you reject out of hand. Push the test harder and you no doubt will suffer higher and higher error rates.

So if you have 100 people who scored a perfect score on the test and only 10 openings. You want to select the 10 best people but clearly if you do it randomly odds are you will not accomplish that. Do you modify the test to produce only 10 people at the top? Do you deploy some other criteria besides the test (interviews, essays, transcript reviews etc.)? What evidence do you have that even at this stage the test is so good that the first option still yields better results?

Note that sports recruiters do NOT do that. A recruiter might use some initial exclusion criteria like running speed to filter the initial mass of applicants but would never reduce the decision on who to put on a professional team to just selecting the top of some metric.

This is autism level missing the point. Jesus Christ, it's like all of you went to state school.

Ivy League schools serve one function: filtering. If you attend it means you've made it past the filter. Believe me, the classes are not hard. The education is not better. Any smart kid can learn convolution integrals. We don't need Harvard for that.

The top paying careers ARE NOT based on how smart you are. It's credentials and salesmanship. I feel like I'm talking to a group of MIT Scott Alexanders. Grow the f up.

You think it's difficult to be an investment banker? Jesus Christ people they use Excel. They're not writing SQL queries and that's not difficult either.

You think it's difficult to be a management consultant ? You can be 22 out of Harvard with a PoliSci degree and make 110k. They're even dumber than iBankers.

The truth is that intelligence is one attribute among many. And anything over 110 is more than enough, after that it's diminishing marginal returns.

Credentials, salesmanship, pedigree, height, facial symmetry, etc. The world does not need that many smart people. It needs tall Harvard men who have square jaws and can sell pedigree to some other idiot.

The top paying careers ARE NOT based on how smart you are. It’s credentials and salesmanship. I feel like I’m talking to a group of MIT Scott Alexanders. Grow the f up.

Per the Bureau of Labor Statistics, the 30 most handsomely compensated occupations, encompassing 2.8% of the workforce are as follows (with mean annual cash compensation appended):

Anesthesiologists $269,600
Surgeons $252,910
Obstetricians and Gynecologists $234,310
Oral and Maxillofacial Surgeons $232,870
Orthodontists $228,780
Physicians and Surgeons, All Other $205,560
Internists, General $201,840
Family and General Practitioners $200,810
Psychiatrists $200,220
Chief Executives $194,350
Pediatricians, General $184,240
Dentists, General $173,860
Dentists, All Other Specialists $171,900
Prosthodontists $168,140
Nurse Anesthetists $164,030
Airline Pilots, Copilots, and Flight Engineers $152,770
Petroleum Engineers $147,030
Computer and Information Systems Managers $145,740
Marketing Managers $144,140
Podiatrists $144,110
Architectural and Engineering Managers $143,870
Lawyers $139,880
Financial Managers $139,720
Natural Sciences Managers $136,150
Sales Managers $135,090
Law Teachers, Postsecondary $134,530
Compensation and Benefits Managers $126,900
Health Specialties Teachers, Postsecondary $125,430
Public Relations and Fundraising Managers $123,360
Personal Financial Advisors $123,100

I really would not attempt practicing medicine without a vigorous intellect.

They’re even dumber than iBankers.

You're laying it on rather thick at this point.

This seems like mood affiliation in support of affirmative action whether consciously or not.

I really would not attempt practicing medicine without a vigorous intellect.

True but diminishing returns sets in. An I really would not attempt practicing medicine without a vigorous intellect. Take anesthesiologists, for example. If you get 100 of them I'm sure all of them are smart, but the 10 best ones are probably not at the top of the 100 because of their intellect but other factors such as experience.

Until 1995, an 800 on the SAT Verbal was extremely rare. That year the College Board added about 80 points to verbal scores, so that a student who would have scored 720 on the old verbal test now scored 800 just like a student who would have scored 800.

Part of what Pinker is complaining about is that Harvard now lacks the tools it once had to accurately identify the next Pinkers.

"A few years ago, Steven Pinker suggested that his college, Harvard, should specialize in admitting the absolute top of the top students in term of academic ability. But his recommendation disappeared almost without a trace because the top colleges have a very comfortable cartel at present."

So here I ask why? Who would this help? Some options:
1. Harvard.
2. The top student(s)
3. Society as a whole.

My thinking is:

1. I don't particularly care about Harvard. But as I pointed out elsewhere, Harvard may find it more in their interest to be a status maker rather than a collector of actual top talent.

2. What exactly happens to the top students now? Presumably they go to reasonably good schools, get good educations and go on to success. Given the rise of inequality in the US it seems hard to think top students are being deprived of their just rewards via some type of Atlas Shrugged scenario.

3. I'm not sure by what mechanism society as a whole would be better off by one institution scoring only the absolute top students versus a diversity of institutions having them. IMO this sounds like a recipe for fragility. You might argue that by having all the best minds in one place, centralized, they will come up with better ideas than if the best minds were decentralized. Charles Murray has, I think, made an interesting argument that the filtering of socio-levels by intelligence actually increases divisions in society and increases dysfunction. Someone who thinks only the top minds should be at Harvard probably would consider the idea of grouping all the top minds into a single business in the wider economy to be socialism.

Excellent comment and good questions.

Who is worse off, and by how much, if a top student finds himself at Duke, say, instead of Harvard? The student? Well, he may be disappointed, but in pure quality of available education terms, is there any significant harm?

Could it be that other Duke students benefit by having a marginally smarter student body?

He may not be worse off.

Imagine we are talking about military officers. One option might be to take the top ten students in officer training and putting each in charge of a squad of ten. Or take the top ten and create a squad of elites with the absolute top student the commanding officer.

In terms of running the army, society seems to be better off with option 1 in that you now have 100 men lead by the 10 best officers. In terms of the opportunities for the best man, I suspect he would also prefer the first option. Leading a team of average people in competition with ten other teams lead by his former classmates would, I suspect, be a more positive and healthier competition than what I would imagine would be the case if the top ten were on the same team (very cutthroat IMO).

You also have terms of trade. Trade usually works for greater advantage when different types are trading with each other. Think of a football team, the top team has different types of players for different types of positions and that works to make for a better team. A team that consisted of simply selecting all the best college quarterbacks and forming them into a team would probably produce the worst team in the NFL.

Presumably, Pinker thinks his proposed reform would help Pinker, who is perhaps North America's top intellectual. So I gave it a listen.

By the way, it would be interesting to know Pinker's standardized test scores.

He went to undergrad in his native Canada, where college admissions are less frenetic.

he has "no skin in the game" as N Taleb would say... My personal experience I came from a 3rd tier university, to do a masters in a 2nd tier university and now work at a first tier university, so it is possible, though it depends on the field.

Pinker did his undergrad at McGill, which is obviously a fine college, but Canadian colleges aren't as obsessive about status as American colleges.

That's because the USA doesn't have a class structure, obs. Just ask any unreflective American.

We have classes. What we don't have are people who chuffer interminably about what does and does not make a gent, or who is and is not a gent, or who is or is not a Sloane. We do not obsess over accents, which are regional and not class-based in this country. Also, private boarding schools are the redoubts of a tiny minority and there isn't much cachet attached to them. Henry Fairlie spent the last 24 years of his life in the United States because he liked living in a matrix where British chickens*** didn't matter. Or, in his words, "There are classes in America. There is no class system".

And one other thing: the American chatterati is much more subtle than the British chatterati when they want to be snotty.

Pinker was accepted into McGill in 1973, this is before the PQ destroyed the province. At that time McGill was, by a good distance, the preeminent university in Canada. Going back that far, I would guess that going to a US university would be difficult for someone from middle classes.

However, it is a very large university compared to elite institutions in the US, so it can't be as selective.

Anyway, I agree it would be interesting to see how he scored in standardized tests. My guess would be quite well.

The auto-destruct button in Quebec was pushed in 1960, when Jean Lesage landed in the premier's chair.

A Professor at a first tier university? Not with a masters, correct? I mean you must have gotten a Ph.D. from somewhere along the way, correct?

where college admissions are less frenetic.

Admissions are 'frenetic' for a modest minority only. About 73% of all students attend in state and about 70% attend public institutions. Most states have only a small corps of private institutions with any cachet (4 or 5 on average and half that outside the northeast).

Do elite colleges such as Harvard lack resources? Harvard's endowment is nearly $38 billion. It's true as Cowen points out that Harvard has come to rely on donations (the endowment) rather than tuition (with tuition covering only a small part of Harvard's costs). That model brings into question Harvard's (and other elite college's) educational purpose (i.e., it's not for profit status under 501(c)(3)). Indeed, Harvard (and the other elites) function more as a for profit investment fund and tech incubator. Cowen's column in Bloomberg is a warning to Harvard and the other elites to reconsider their funding model

"A more unfortunate reality is that some donors might limit their support if say Princeton offered them and their children a less tony and exclusive experience."

Once you're in, you're in.

This ties into my skepticism about vouchers.

When Facebook first launched, it was only for Ivy League College students. Then it was for college students. Then it was basically for everyone. If you have a dynamite product, you don't restrict who can buy it, you sell it to anyone who has cash. Except, if part of your product is exclusivity then you can't. This is why there are countless McDonalds around the world but there was only one Studio 54.

If Harvard had a great product on its own merits, then there's no reason they wouldn't be franchising Harvard around the world, let alone inside the US. If, however, a part of Harvard's product is that there will only be so many people at any given time with a Harvard degree, then there you have the reason they can't simply admit more legacies and use their tuition dollars to increase slots for non-legacies.

In relation to vouchers, keep in mind the US higher education system is built on the premises of voucher advocates. Students and parents go to whatever college they want using the money they get from gov't against tuition...a more expensive school that isn't better means their aid covers less so tada 'competition'. Does it make great schools? Actually no it makes schools that cater to their customers and one of the things customers want is exclusivity. A parent would see the value in a system of great education for , say, 20% of the population with 1% being given special elite status....provided their kids are in that 1%. As a society, though, we'd be better off if, say, we could pay for great education for 70% of the population and not contribute towards any 'elite premiums' that are by definition a zero sum game. Put it in more simple terms, suppose gov't established a universal $7500/year voucher. I would predict the long run equilibrium result would simply be every school would just raise their tuition by $7500/year.

"This is why there are countless McDonalds around the world but there was only one Studio 54."

I hope you aren't too bitter about our recent decision, Boonton. You got quite a few votes based partly on your persistence in trying to get in but even my vote couldn't put you over the top.

This ties into my skepticism about vouchers.

Gassers gotta gas.

Just privatize everything. Dum dums can die in a gutter.

Fortunately, that never happens now....

Surely there is a sort of Laffer curve for legacy admissions. If you admitted every legacy applicant, donations would go down.

teaching quality is unlikely to suffer, as Harvard doesn’t select for the very best teachers.

I don't see why this is a plausible claim at all. If anything, the best teachers are likely to be the least harmed by higher teaching loads (unless what they are especially good at is small seminars, but that's surely not true of everyone.) As a merely good teacher, I'll note that adding more students very quickly starts to cut into my teaching ability, or else requires the use of various sorts of helpers - TAs, adjuncts, graders, etc., most of whom are likely no better, and probably worse, teachers. The over-all proposal may be a good one, but this claim seems to me to be obviously wrong, at least if we take teaching seriously.

I'm pretty sure that the idea is that Harvard could expand its faculty as much as it needed to cover any additional load without suffering any diminution in quality of teaching. If the quality of their teaching isn't that high to begin with, the pool they can draw additional at-least-as-competent teachers from is very large.

>But that is true in any kind of startup endeavor

Oh! I didn't realize Harvard was a startup. You're very informative, Ty.

As has been explained to you before, exclusivity is about 95% of what Harvard is selling. Good luck telling them to water it down.

Every organization has to be a startup these days, even if they don't. How else can they be cool?

The purpose of a startup is to find product-market fit. Harvard has obviously done that, so they should now be executing on their formula, right? Nope. Why? Because TC doesn't like their market and, therefore, they need to serve another.

Indeed, there seems to be a lot of grumbling in the Harvard community about the very existence of the Extension School.

I've seen charlatans of various sorts calling themselves Harvard graduates when all they did was spend some money on a month-long certificate course. It's pretty reasonable to be annoyed at the brand-dilution potential of this.

"all they did was spend some money on a month-long certificate course."

How dare they!

Everyone else had to spend some money on a four year long certificate course.

Whether or not a school ultimately serves more students and becomes more diverse by admitting more legacy students depends on what they do with the additional donor revenue, doesn't it?

Harvard could probably double the number of legacies it admits with only a slight drop-off in overall quality of its student body. Presumably this would increase donor revenue. If it enlarged its intake of legacy students *and reduced its intake of non-legacy students by the same amount* and then used the additional donor revenue to reduce the *top-line sticker price* of its tuition, that would do nothing to serve more students and would likely *reduce* diversity.

Alternately, it could double the number of legacies students it admits, keep admitting the same amount of non-legacy students (i.e. it would have a slightly larger incoming class), then use the additional donor revenue to fund need-based aid to reduce the bill of the least-wealthy incoming students.

This would achieve Tyler's ends of serving more students and increasing (socioeconomic, at least) diversity. But it requires more than just admitting more legacy students.

One more point on revenue: legacy students likely skew more wealthy than the average applicant, so tuition-per-student is going to be higher for legacy students than for the average student.

Tyler's kids must be approaching college age?

More likely shirt-tails. His wife has a daughter, though.

I'm told the University of Chicago gives little advantages to legacies. Of course alumni of the 70's and 80's tended not to send their children to the school. While the academics were excellent, the overall experience for those former students was less positive. However a few successful graduates have been generous donors.

I was also told that the Business School at Chicago is less interested in admitting students who might be future donors but are very interested in how they can market graduates to employers. They admit students that their customers, future employers, want to buy. Connections can play a role i.e. an applicant with connections to employers who heavily recruit at the school can have an advantage. In addition, employers who recruit on campus have many current alumni employed. This can lead to a concentration in some fields or geographic locations. The cycle can be positive for all involved.

I don't care much if my children go to the same schools that I attended. Schools and faculty change over time. Student bodies change. Expecting that my child's experience would be like mine is a bit delusional.

I send money every year to my former schools, but I'm not sure why. Others gave so that I could attend. I am, I suppose, thanking those mostly dead white guys,

Rather then thinking what benefit legacies can offer the school, I would ask what value does an involved alumni offer graduates of the school. Assuming that a school with more legacies has a more involved alumni (think Notre Dame, UNC, etc) the more value it generates for applicants.

Chicago is the best college in America, without a doubt. It's not for everyone, though, as the demands on the students would have overwhelmed me at that age (I'm old).

The hoopla regarding the quality of education at elite institutions can be quickly dispelled

If you go online

And watch some of the MOOCs offered by these schools

And compare them to the offerings of other schools.

Elitism is just tribal loyalty.

Didn't Stanford explore putting a campus in NYC at some point?

Real estate (bricks and mortar) expansion is hard and expensive in desirable locations for professors.....however if the idea is to expand student reach less glamorous locations are the right idea.

Any suggestions for locations? Lets say a Hispano magnet Harvard in San Antonio Texas and an African American magnet Harvard in Oakland California?

Do you really think that would happen?

Why not a new "Harvard" funded by some billionaire money in some neutral location without a great school?

The idea that the elite colleges are curently run for the students is kind of a polite fiction. Thay are run for the professors and the students are tolerated. Daddy and Mommy pay for the babysitting and the family comes away with a very valuable piece of paper. For the upper middle class (including professors) this is a good game. Not so sure it is right approach for boosting the less economically well off. Im pretty sure many of them have culture shock as it is whne they see what goes on at these schools.

Daddy and Mommy pay for the babysitting

There isn't any babysitting. The faculty loathe fraternities and wish to destroy them. That aside, institutional policy allows (even promotes) grossness in large quantities.

The students have to complete and turn in their work for that 'valuable piece of paper' and (at the place I know best) most qualify for work-study jobs (though there are more qualified than there are jobs).

"I'm pretty sure many of them have culture shock as it is whne they see what goes on at these schools"
But then they learn to enjoy studying and things better, they may even get ahead in life.

"Or Harvard could open a branch for part-time study, and for Harvard certificates, but based in and serving one of the poorer neighborhoods of Boston."

The jokes write themselves.

What's missing from this essay is any actual evidence of the role of legacy admissions in donations to a school. Tyler, you assume they are responsible for significant funding, and that more legacy admissions would create additional funding, but you don't provide any evidence for either point. Thus, this article is just speculative about that. What's more persuasive, in my view, is your view (which you have better addressed elsewhere) that elite schools should educate many more students.

Why should they 'educate many more students'? There is plenty of opportunity for the aspirant student outside the ambo of private research universities and fancy liberal arts colleges.

Harvard is very cash flush, so it's hard to see the point of this.

The schools for which legacy admissions might be consequential number perhaps 200-odd and encompass perhaps 10% of the FTE student population at baccalaureate-granting institutions. At the place I know best, a faculty member on the relevant committee reported about 15 years ago that about 8% of the students admitted each year were legacies or were admitted for reasons of 'political relations'. Given the correlation in academic performance between generations, it's a passable guess that the share for which the status consideration was decisive would be more along the lines of 3%. You're really not doing these youths any favors. It's a small scale problem, however.

Can we please just drone strike Harvard and Yale and forget about it?

Here's a suggestion: impose a light federal regulatory regime on private higher education (which would, to be sure, include some elements the institutions in question would hate). Then end the grant money flows, end the loan guarantees, and end the super-duper creditor protection on student loans. Let them be a**holes on their own time and their own dime.

+1 to whomever this really is

Isn't the reproductive rate of Harvard alumni maybe just a little bit lower than say, the general population of Lagos, Nigeria? Left to their own devices won't Harvard alumni pretty much allow their line to disappear through birth control? The elite/upper class belief that reproduction is wrong means that inevitably they'll become even more outnumbered by the proles than they are now, a dangerous situation. Increased legacy admissions will address this issue, especially in the financial arena, but won't solve it. Harvard grads will become proportionally more exclusive yet, as the commoners come to their available wits, will also become less influential.

If Harvard was like a normal business, they would have franchise outlets all over the country by now. That might degrade standards somewhat for Harvard, but it would probably also raise them for the students being served. Of course it would also diminish the signaling value of a Harvard degree, but that wouldn't be a bad thing overall. Surely it would be more profitable to expand than to remain exclusive. It would be better for nearly everyone if the best educational institutions replicated their model across the country.

Hazel, they don't have a product or a process they can market extensively. They've got their human capital and their endowment and their physical plant as is.

It is a routine agency issue for administration. They can be highly paid and high status as things are. For them any expansion is a risk.

Not true for state university regents, see below.

"Of course it would also diminish the signaling value of a Harvard degree, but that wouldn’t be a bad thing overall."

Street lights should have the same color for stop and go to reduce their signaling value.

Some signals are useful. Some aren't.


It must be embarrassing living in a country where apots at the most prestigious universities can be sold and bought because meritocracy has no place whatsoever...

Just say no.

I think we need more and better state schools, from trade schools to elite universities. Fifty states can do much more for innovation, productivity, and future wealth, than expanded Ivies.

Yes it's the lack of schools. As we can clearly see, if we push everyone to college everyone will be rich! Magik books!

Reminds me of the 30 Rock sketch about buying tuxedos for the homeless.

It's very obvious that, if nothing else, schools are failing to teach signaling theory.

But hey, it is a massive transfer of wealth from blue collar Americans to the education industry. So democrats come out ahead.

A reader would have spotted "trade schools" and recognized something a bit different than "college for everyone."

FWIW, I watched UCI go from being a lesser UC to .. ranked 39 nationally?

It can be done. More states should go for it.

And all it took was discrimination against Asians for years at UCLA and UC Berkeley. Magically UCSD and UCI rose in the rankings.

Is that your silver bullet? Make it illegal for Asians to go to good schools to "spread the (iq) wealth around?"

UCI's history is more complicated than that. It helped to have OC millionaires fund institutes and endow chairs, and in turn for that to attract talent.

As far as "the Asian problem," I think elite public schools should be colorblind and favor test scores, but the flip side is that a state should have a good string of less elite schools for all qualified candidates.

How many mechanical engineering slots should there be over all? Enough for all willing and able to do the work.

This is not the same as accepting anyone and then dropping them, with debt, sophomore year.

UCI has been heavily Asian for thirty years or more, and its rankings have held steady in the 40s for all that time.

Meanwhile, Berkeley was ranked fifth in America back in 1983, but plummeted to the low teens when US News changed its methodology. And since it's become increasingly Asian after CA ended AA in 1997, it's dropped another ten points. UCLA, which has also become much more Asian, has seen no increase in its rankings.

Most top public universities hover between twentieth and fiftieth regardless of their demographics.

My memories go back to the 70's, fwiw.

Lately, UC Irvine has been heavily recruiting Hispanics and its Asian percentage has dropped sharply.

UC Irvine will never live down its being the birth place of the Sherwood Rowland/Mario Molina computer modeling study that produced the bogus CFC/ozone hole and subsequently the Montreal Protocol.

I have a scientific degree, and can read science with fair literary. It is increasingly hard for me to understand how anyone with scientific literary can stay "right" or "Republican."

There is simply no respect for science, and calling anyone on science leads immediately to "that is just your religion," "we don't need priests" or other b.s.

I trust the Smithsonian as a rational and accessible source:

"Each year during ozone hole season, scientists from around the world track the depletion of the ozone above Antarctica using balloons, satellites and computer models. They have found that the ozone hole is actually getting smaller: Scientists estimate that if the Montreal Protocol had never been implemented, the hole would have grown by 40 percent by 2013. Instead, the hole is expected to completely heal by 2050."

Read more:

Feel free to demonstrate why scientists were abandoned by the right, and not his versa.

But apparently I do not catch when my tablet turns "literacy" to "literary."

So the essential argument is that Harvard should sell more acceptances, because it could do good things with the money.

Would it? Maybe. Would the same amount of money directed elsewhere accomplish more? Who knows.

Also missing, as so often from these lovely theoretical arguments, is any information as to how much of the money Harvard gets from donations is intended to buy admission, and how much is just philanthropy. In other words, what is the actual financial benefit to the university from the (ahem) marginal legacy admitee?

I just can't believe what I just read and it's penned by prof Cowen? What a terrible idea! But if you are already well established in this country and want to continue the status quo for your off springs, why not? Legacy is the affirmative action for the privileged, and predominantly, white.

Oh dear, and what will we all do if Harvard doesn't do everything right? If their practices are somehow suboptimal, rather than spilling ink over how to fix poor Harvard, we should ignore them and look to Universities who are doing it in a way more to our liking.

Comments for this post are closed