Increasingly Competitive College Admissions: Much More Than You Wanted To Know

0: Introduction

This is from

Acceptance rates at top colleges have declined by about half over the past decade or so, raising concern about intensifying academic competition. The pressure of getting into a good university may even be leading to suicides at elite high schools.

Some people have dismissed the problem, saying that a misplaced focus on Harvard and Yale ignores that most colleges are easier to get into than ever. For example, from The Atlantic, Is College Really Harder To Get Into Than It Used To Be?:

If schools that were once considered “safeties” now have admissions rates as low as 20 or 30 percent, it appears tougher to get into college every spring. But “beneath the headlines and urban legends,” Jim Hull, senior policy analyst at the National School Board Association’s Center for Public Education, says their 2010 report shows that it was no more difficult for most students to get into college in 2004 than it was in 1992. While the Center plans to update the information in the next few years to reflect the past decade of applicants, students with the same SAT and GPA in the 90’s basically have an equal probability of getting into a similarly selective college today.

Their link to the report doesn’t work, so I can’t tell if this was ever true. But it doesn’t seem true today. From Pew:

The first graph shows that admission rates have decreased at 53% of colleges, and increased at only 31%. The second graph shows that the decreases were mostly at very selective schools, and the increases were mostly at less selective schools. We shouldn’t exaggerate the problem: three-quarters of US students go to non-selective colleges that accept most applicants, and there are more than enough of these for everyone. But if you are aiming for a competitive school – not just Harvard and Yale, but anywhere in the top few hundred institutions – the competition is getting harder.

This matches my impression of “facts on the ground”. In 2002, I was a senior at a California high school in a good neighborhood. Most of the kids in my class wanted to go to famous Ivy League universities, and considered University of California colleges their “safety schools”. The idea of going to Cal State (California’s middle- and lower- tier colleges) felt like some kind of colossal failure. But my mother just retired from teaching at a very similar school, and she says nowadays the same demographic of students would kill to get into a UC school, and many of them can’t even get into Cal States.

The stories I hear about this usually focus on how more people are going to college today than ever, but there’s still only one Harvard, so there’s increasing competition for the same number of spots.

As far as I can tell, this is false.

The college attendance rate is the same today as it was in 2005. If you’ve seen graphs that suggest the opposite, they were probably graphs of the total number of Americans with college degrees, which only proves that more people are getting degrees today than in the 1940s or whenever it was that the oldest generation still alive went to college.

(in fact, since the birth rate is declining, this means the absolute number of college-goers is going down).

I’ll go further. Harvard keeps building more dorms and hiring more professors, so there are the same number of Harvard spots per American today as there were ten years ago, twenty years ago, and all the way back to the 1800s:

I want to look into this further and investigate questions like:

– How did we get to this point? Have college admissions always been a big deal? Did George Washington have to freak out about getting into a good college? What about FDR? If not, why not?

– Is academia really more competitive now than in the past? On what time scale? At what levels of academia? Why is this happening? Will it stop?

– Is freaking out about college admissions the correct course of action?

1. A Harvard-Centric History Of College Admissions

For the first two centuries of American academia, there was no competition to get into college. Harvard admitted…

(Harvard is by far the best-documented college throughout most of this period, so I’ll be focusing on them. No, Ben Casselman, you shut up!)

…Harvard admitted anyone who was fluent in Latin and Greek. The 1642 Harvard admission requirements said:

When any schollar is able to read Tully [Cicero] or such like classicall Latine Authore ex tempore & make and speake true Latin in verse and prose, suo (ut auint) Marte, and decline perfectly the paradigmes of Nounes and Verbes in the Greek tongue, then may hee bee admitted into the Colledge, nor shall any claim admission before such qualifications.

Latin fluency sounds impressive to modern ears, like the sort of test that would limit admission to only the classiest of aristocrat-scholars. But knowledge of classical languages in early Massachussetts was shockingly high, even among barely-literate farmers. In 1647, in between starving and fighting off Indian attacks, the state passed a law that every town of at least 100 families must have a school that taught Latin and Greek (it was called The Old Deluder Satan Law, because Puritans). Even rural families without access to these schools often taught classical languages to their own children. Mary Baker Eddy, who grew up in early 19th-century rural New Hampshire, wrote that:

My father was taught to believe that my brain was too large for my body and so kept me much out of school, but I gained book-knowledge with far less labor than is usually requisite. At ten years of age I was as familiar with Lindley Murray’s Grammar as with the Westminster Catechism; and the latter I had to repeat every Sunday. My favorite studies were natural philosophy, logic, and moral science. From my brother Albert I received lessons in the ancient tongues, Hebrew, Greek, and Latin. My brother studied Hebrew during his college vacations.

By the standards of the time, Harvard admission requirements were tough but fair, and well within the reach of even poorer families. More important, they were only there to make sure students were prepared for the coursework (which was in Latin). They weren’t there to ration out a scarce supply of Harvard spots. In fact this post, summarizing Jerome Karabel’s Chosen, says that “there was no class size limit, because Harvard was trying to compete with Oxford and Cambridge for size”. They wanted as many students as they could get; their only limit was the number of qualified applicants.

These policies continued through the 19th century, with changes only in the specific subjects being tested. In the late 1700s they added some math; in the early 1800s they added some science. You can find a copy of the 1869 Harvard entrance exam here. It’s pretty hard – but it had an 88% pass rate (surely at least in part because you wouldn’t take it unless you were prepared) and everyone who passed was guaranteed a spot at Harvard. Some documents from Tufts around this time suggest their procedure was pretty similar. Some other documents suggest that if you went to a good high school, they assumed you were prepared and let you in without requiring the exams.

When did this happy situation end? Information on this topic is hard to find. I can’t give specific sources, but I get the impression that at the very end of the 19th century, there was a movement to standardize college admissions. At first this just meant make sure every college has the same qualification exams, so that one school isn’t asking about Latin and another about Greek. This culminated in the creation of the College Board in 1899, which administered an admission test that acted as a sort of great-great-grandfather of the SAT. Very gradually, so gradually that nobody at the time really remarked on it, this transitioned from making sure students were ready, to rationing out scarce spots. By about 1920, the transition was basically complete, so that nobody was surprised when people talked about “how colleges should decide who to accept” or questions like that. If you can find more on this transition, please contact me.

Acceptance was originally based entirely on your score on the qualifying exam. But by the 1920s, high-scorers on this exam were disproportionately Jewish. Although Jews were only about 2% of the US population, they were 21% of Harvard’s 1922 class (for more on why this might happen, read my post The Atomic Bomb Considered As Hungarian High School Science Fair Project). In order to arrest this trend, Harvard and other top colleges decided to switch from standardized testing to an easier-to-bias “holistics admissions” system that would let them implement a de facto Jewish quota.

Quota proponents not only denied being anti-Semitic but argued they were actually trying to fight anti-Semitism; if the student body became predominantly Jewish, this might inflame racial tensions against Jews. Harvard president Abbott Lowell, the quotas’ strongest proponent, said: “The anti-Semitic feeling among the students is increasing, and it grows in proportion to the increase in the number of Jews. If their number should become 40% of the student body, the race feeling would become intense”. Was he just trying to rationalize his anti-Semitism? I don’t think so. I doubt modern Harvard officials are anti-Asian in any kind of a hateful sense, but they enforce Asian quotas all the same. What would they say if you asked them why? Maybe that if a country full of whites, blacks, and Latinos had predominantly Asian elite colleges, that might make (as Lowell put it) “the race feeling become intense”. I see no reason to think that 1920s officials were thinking any differently than their modern counterparts.

Whatever the reasons, by the mid-1920s the Jewish quota was in place and Harvard had switched to holistic admissions. But Lowell and his contemporaries emphasized that the new policies were never meant to make Harvard selective. “It is neither feasible nor desirable to raise the standards of the College so high that none but brilliant scholars can enter…the standards ought never to be so high for serious and ambitious students of average intelligence.”

We’ll talk later about how this utopian dream of top-notch education for anyone with a foreskin failed. But before we get there, a more basic question: how come Harvard wasn’t overrun with applicants? If the academic requirements were within reach of most smart high-schoolers, how come there was no need to ration spots?

Below, I discuss a few possibilities in more depth.

1.1: Historical Tuition Fees

Were early American colleges so expensive that everyone except aristocrats was priced out?


(sources: 1, 2, 3, 4, 5, 6)

I find very conflicting accounts of colonial tuition prices. But after the Revolution, tuition stayed stable about about a third average median income until about 1990, when it increased to 1.5x median income. In other words: relative to income, historical tuition costs were about a fifth of what they are today. Some good universities seem to have not had tuition at all – Stanford had a $0 price tag for its first 35 years.

Even when tuition existed, historical accounts suggest it wasn’t especially burdensome for most college students, and record widespread effort to accommodate people who couldn’t pay.The first Harvard scholarship was granted in the 1640s. There are occasional scattered references to people showing up at Harvard without enough money to pay and being given jobs as servants to college officials or other students to help cover costs; in America, Ralph Waldo Emerson took advantage of this kind of program; in Britain, Isaac Newton did.

If you were a poor farmer who couldn’t get a scholarship and didn’t want to work as a servant, sometimes college were willing to accept alternative forms of payment. According to The Billfold:

Harvard tuition — which ran about fifty-five pounds for the four-year course of study — was paid the same way [in barter], most commonly in wheat and malt. The occasional New England father sent his son to Cambridge with parsnips, butter, and, regrettably for all, goat mutton. A 141-pound side of beef covered a year’s tuition.

1.2 Discrimination

Early colleges only admitted white men. Did this reduce the size of the applicant pool enough to give spots to all white men who applied?

I don’t think racial discrimination can explain much of the effect. Throughout the 19th century, America hovered around 85% white. New England, where most Harvard applicants originated, may have been 95% to 99% white – see eg this site which says Boston was 1.3% black in 1865; non-black minorities were probably a rounding error. So there’s not much room for racial discrimination to reduce the applicant pool.

The exclusion of women from colleges in the 1800s is less than generally believed:

(source: unprincipled sketchy attempt to combine this with this to get one measure that covers the entire period)

For every woman in college in 1890, there were about 1.3 men; this is no larger a gender gap than exists today, though in the reverse direction. How come you never hear about this? Many of the women were probably in teacher-training colleges or some other gendered institution; until the early 1900s, none of them were at Harvard. But after gender integration, the women’s colleges were usually annexed to the nearest men’s college, turning them into a single institution. Under these circumstances, it doesn’t seem that likely that integration had a huge effect on admissions selectivity. Also, admitting women can only double the size of the applicant pool, but 1800s college seemed much more than twice as easy to get into.

Overall I don’t think this was a major part of the difference either.

1.3: Lack Of Degree Requirement For Professional Schools

Nowadays college is competitive partly because people expect it to be their ticket to a good job. But in the 19th century, there was little financial benefit to a college degree.

Suppose you wanted to become a doctor. Most medical schools accepted students straight out of secondary school, without a college degree. In fact, most medical schools accepted all “applicants”, the same as Harvard. Like Harvard, there was sometimes a test to make sure you knew Greek and Latin (the important things for doctors!) but after that, you were in.

(This article has some great stories about colonial and antebellum US medical education. Anyone who wanted could open up a medical school; profit-motive incentivized them to accept everybody. Medical-schooling was so profitable that the bottleneck became patients; since there were no regulations requiring medical students to see patients, less scrupulous schools tended to skip this part. Dissection was a big part of the curriculum, but there were no refrigerators, so fresh corpses became a hot commodity. Grave robbing was a real problem, sparking small-scale wars between medical schools and their local towns. “In at least 2 instances, the locals actually raided the school to obtain a body. In 1 case, the school building was destroyed by fire, and in another, 2 people, a student and a professor, were killed.” There were no requirements for how long medical schools should last, so some were as short as nine months. But there were also no requirements for who could call themselves a doctors, so students would sometimes stay until they got bored, then drop out and start practicing anyway. Tuition was about $100 per year, plus cost of living and various hidden fees; by my estimates that’s about half as much (as percent of an average doctor’s salary) as medical school tuition today. This situation continued until the Gilded Age, when medical schools started professionalizing themselves a little more.)

Or suppose you wanted to be a lawyer. The typical method was called “reading law”, which meant you read some law textbooks, served an apprenticeship with a practicing lawyer, and then started calling yourself a lawyer (in some states you also needed a letter from a court testifying to your “good moral character”). Honestly the part where you apprenticed with an practicing lawyer was more like a good idea than a requirement. It’s not completely clear to me that you needed to do anything other than read enough law textbooks to feel comfortable lawyering, and then go lawyer. Most lawyers did not have a college degree.

Abraham Lincoln, a lawyer himself, advised a law student:

If you are absolutely determined to make a lawyer of yourself the thing is more than half done already. It is a small matter whether you read with any one or not. I did not read with any one. Get the books and read and study them in their every feature, and that is the main thing. It is no consequence to be in a large town while you are reading. I read at New Salem, which never had three hundred people in it. The books and your capacity for understanding them are just the same in all places.

Levi Woodbury, the 30th US Supreme Court Justice (appointed 1846), was the first to attend any kind of formal law school. James Byrnes, the 81st Supreme Court Justice (appointed 1941), was the last not to attend law school. It’s apparently still technically possible in four states (including California) to become a lawyer by reading law, but it’s rare and not very encouraged.

The ease of entering these professions helps explain why there was no oversupply of Harvard applicants. But then why wasn’t there an oversupply of doctors and lawyers? We tend to imagine that of course you need strict medical school admissions, because some kind of unspecified catastrophe would happen if any qualified person who wanted could become a doctor. Did these open-door policies create a glut of professionals?

No. There were fewer doctors and lawyers per capita than there are now.

Did it drive down salaries for these professions?

I don’t have great numbers on lawyer salaries, but based on this chart from 1797 Britain and this chart from 1900s America, I get the impression that throughout this period lawyers made about 3-5x as much as unskilled laborers, 3-4x as much as clerks and teachers, and about the same as doctors. This seems to match successful modern lawyers, and probably exceed average modern lawyers. This may because unskilled laborers now earn a minimum wage and teachers have unions, but in any case the 19th-century premium to a law degree seems to have been at least as high and probably higher.

The same seems true of doctor salaries. The paper above estimates physician salaries at $600 per year, during a time when agricultural laborers might have been making $100 and clerks and teachers twice that.

I conclude that letting any qualified person become a doctor or a lawyer, without gatekeeping, did not result in a glut of doctors and lawyers, and did not drive down salaries for those professions beyond levels we would find reasonable today.

1.4: Conclusions

So why weren’t there gluts of would-be college students, doctors, and lawyers? I can’t find any single smoking gun, but here are some possibilities.

Throughout this period, between 60% and 80% of Americans were farmers. Unless you were wealthy or urban, the question of “what career do you want in order to actualize your potential” didn’t come up. You were either going to be a farmer, or else you had some specific non-farm pathway in mind that you could pursue directly instead of getting a college degree to “keep your options open”.

Since rural children were expected to work on the farm, there was no protracted period of educational unproductivity. There was no assumption that your kids weren’t going to be earning anything until age 18 and so you might as well protract their unproductivity until age 22. That meant that paying to send your child to Boston or wherever, and to support him in a big-city lifestyle for four years, was actually a much bigger deal than the tuition itself. This article claims that in 1816, tuition itself was only about 10% of the expenses involved in sending a child to college (granted, poor people pinching pennies could get by for much less than the hypothetical well-off student analyzed here, but I think the principle still holds).

Another limiting factor may have been that there was ample opportunity outside of college and the professions, in almost every area. Twelve US presidents, including George Washington, did not go to college. Benjamin Franklin, everyone’s model of an early American polymath genius, did not go to college. Of the ten richest people in American history (mostly 19th-century industrialists), as far as I can tell only two of them went to college. Aside from the obvious race and gender discrimination, the 19th century was a lot closer to real meritocracy than today’s credentialist fake meritocracy; people responded rationally by ignoring credentials and doing meritorious things.

2. How Did The Zero-Competition Regime Transition To The Clusterf**k We Have Today?

Here is a graph of Harvard admission rates over time, based mostly on these data:

During the early part of the 1900s, Harvard was still in the 19th-century equilibrium of admitting most qualified non-Jewish applicants. Around 1940, the admission rate dropped from 95% to 25%. Most sources I read attribute this to the GI Bill, a well-intentioned piece of legislation that encouraged returning WWII veterans to get a college education. So many vets took the government up on the offer that Harvard was overwhelmed for the first time in its history.

But this isn’t the whole story.

You’ve seen this before – this is percent of Americans (by gender) to graduate college. It’s sorted by birth cohort, which means 1920 on the x-axis corresponds to the people who were in college in the 1940s – eg our GIs. The GI Bill is visible on this graph – around 1920, there is a spike in attendance for men but not women, which is the pattern we would predict from GIs. But it only takes college graduation rate from 10% to 15% (compared to its current 40%). And after the GI Bill, the college graduation rate starts dropping again – as we would expect of a one-time shock from a one-time war. And between 1955 and 1960, Harvard admissions rebound to about 40% of applicants.

The big spike in college attendance rates – and a corresponding dip in Harvard admission percentage – takes place in the 1938 to 1952 birth cohort. Why are all these people suddenly going to college? They’re dodging the draft. A big part of the increase in college admissions was people taking advantage of the college loophole to escape getting sent to Vietnam.

Again, this is a one-time shock, and mostly applies to men. So how come we see a quadrupling of college graduation during this period affecting men and women alike?

A standard narrative says that work has gotten more difficult over the past century, and so workers need more education. I’ve always found this hard to believe. In other countries, students still go to medical school and law school without a separate college degree first. Programming is a classic example of a high-skilled complicated modern profession, but many programmers dropped out of college, many others didn’t attend at all, and many programming “boot camps” are opening up offering to teach programming skills outside the context of a college education. And in many of the jobs that do require college education, the education is irrelevant to their work. Both of my adult jobs – as an English teacher and as a doctor – required me to have a college degree in order to apply. But my college education was relevant to neither (I’m a philosophy major). The degree requirement seemed like more of a class barrier / signaling mechanism than an assertion that only people who knew philosophy could make good teachers and doctors. I realize I’m making a strong claim here, and I don’t have space to justify it fully – for more on this, read my Against Tulip Subsidies and SSC Gives A Graduation Speech – or better yet, Bryan Caplan’s The Case Against Education.

If increasing need for skills didn’t cause increasing college attendance, what did? Again, this is based off of idiosyncratic beliefs I don’t have the space to justify (again, read Caplan) but it could be a sort of self-reinforcing signaling cycle. Once the number of people in college reached a certain level, it led to a well-known social expectation that intelligent and conscientious men would have college degrees, which made college a sign of intelligence and, conversely, not having been to college a sign of stupidity. If only 10% of smart/hard-working people have been to college, not having a college degree doesn’t mean someone isn’t smart/hard-working; if 90% of smart/hard-working people have been to college, not having a college degree might call their intelligence and work ethic into question. This cycle meant that after the shocks of the mid-1900s, there was a strong expectation of a degree in the knowledge professions, which forced women and later generations of men to continue going to college to keep up. The government’s decision to provide an endless stream of supposedly-free college loans exacerbated the problem and sabotaged the only natural roadblock that could have stopped it.

At the same time, several factors were coming together to discourage hunch-based “I like the cut of his jib” style hiring practices. Community ties were becoming weaker, so hirers typically wouldn’t have social contacts with potential hirees. Family businesses whose owners could hire based on hunches were giving way to large corporations where interviewers would have to justify their hiring decisions to higher-ups. Increasing concern about racism was raising awareness that hunch-based hiring tended to discriminate against minorities, and the advent of the discrimination lawsuit encouraged hiring based on objective criteria so you could prove you rejected someone for reasons other than race. The Supreme Court decision Griggs v. Duke may or may not have played a role by making it legally risky for corporations to give prospective hires aptitude tests. All of this created a “perfect storm” where employers needed some kind of objective criteria to evaluate potential new hires, and all the old criteria weren’t cutting it anymore. The rise of the college degree as a signal for intelligence, and the increased sorting of people by college selectivity, fit into this space perfectly.

Once society established that knowledge-worker jobs needed college degrees, the simultaneous rises in automation, globalization, and inequality made knowledge-worker jobs increasingly necessary to earn a living, completed the process.

If my story were true, this would suggest college attendance would not have risen so quickly in other countries that didn’t have these specific factors. I don’t have great cross-country data, but here’s what I can find:

College attendance in the UK supposedly remained very low until a 1992 act designed to encourage it, but it looks like part of that is just them reclassifying some other schools as colleges. I don’t know how it really compared to the US and I welcome information from British readers who know more than I do about this. Through the rest of the world, college attendance lagged North America by a long time, but the continent-wide categories probably combine countries at different levels of economic development. I don’t really know about this one.

Moving on: the graphs in the Introduction show that college attendance has been stable since about 2005. Why did the rise stop? These articles point out a few relevant trends.

First, the economy is usually to blame for this kind of thing. There was a slight increase in attendance during the 2008 recession, and a slight decrease during the recent boom. But over the course of the cycle, it still seems like the increase in college attendance has slowed or stopped overall, in a way that wasn’t true of past business cycles.

Second, birth rates are decreasing, which means fewer college-aged kids. The national population is still increasing, mostly because of immigrants, but many immigrants are adults without much past education, so they’re not as significant a contribution to the college population.

Third, the price of college keeps going up. I’m surprised to hear this as a contribution to declining attendance, because I thought it was the glut of students that kept prices high, but maybe both factors affect each other.

Fourth, for-profit colleges are falling apart.

In some cases, the government has shut them down for being outright scams. In other cases, potential students have wised up, realized they are outright scams, and stopped being interested in attending them. These colleges advertised to (some would say “preyed on”) people who weren’t able to get into other colleges, so their collapse looks like a fall in the college enrollment/graduation rate.

These are all potentially relevant, but they seem kind of weak to me: the sort of thing that explains the year-to-year trend, but not why the great secular movement in favor of more college has stopped.

Maybe it’s just reached a natural ceiling. Seventy percent of high school graduates are now going to college. The remaining 30% may disproportionately include people with serious socioeconomic or health problems that make going to college very hard for them.

Also, keep in mind that only about 60% of college students graduate in anywhere near the expected amount of time. Some economists have come up with rational-college-avoidance models where people who don’t expect to be able to graduate from college don’t waste their money trying.

3. If Number Of Students Applying To College Has Been Constant Or Declining Over The Past Ten Years, Why Are Admissions To Top Colleges So Much More Competitive?

To review: over the past ten years, the number of US students applying to college has gone down (the number applying to four-year private colleges has stayed about the same). But Harvard’s acceptance rates have decreased by half, with similar cuts across other top schools, and more modest cuts across most good and moderately-good colleges. There’s also a perception of much greater pressure on students to have perfect academic records before applying. Why?

3.1: Could the issue be increasing number of international students?

This would neatly match the evidence of constant US numbers vs. increasing selectivity.

Harvard equivocates between a few different definitions of “international student”, but I think it’s comparing apples to apples when it says the Class of 2013 was 10% foreign citizens and the Class of 2022 is 12%. These two classes bound the time period we’re worrying about, and this doesn’t seem like a big change. Also, across all US colleges international student enrollments seem to be dropping, not increasing. Some of this may have to do with strict Trump administration visa policies, or with international perceptions of increasing US hostility to foreigners.

Since fewer international students are applying in general, and even top schools show only a trivial increase, this probably isn’t it.

3.2: Could the issue be more race-conscious admission policies?

Might top colleges be intensifying affirmative action and their preference for minorities and the poor, thus making things harder for the sort of upper-class white people who write news articles about the state of college admissions? Conversely, might colleges by relaxing their restrictions on high-achieving Asians, with the same result?

This matches the rhetoric colleges have been putting out lately, but there is not a lot of signs it’s really happening. Harvard obsessively chronicles the race of its student body, and the class of 2010 and class of 2022 have the same racial composition. The New York Times finds that whites are actually better represented at colleges (compared to their percent of the US population) than they were 35 years ago, although Asians are the real winners.

The Times doesn’t explain why this is happening. It may be due to weakening affirmative action, including bans by several states. Or it may be because of a large influx of uneducated Mexican immigrants who will need a few more generations of assimilation before their families attend college at the same rate as whites or previous generations of Latinos.

What about Asians? There was a large increase in Asian admissions, but it was mostly before this period. The Ivy League probably has some kind of unofficial Asian quota which has been pretty stable over the past decade. Although the Asian population continues to grow, and their academic achievement continues to increase, this probably just increases intra-Asian competition rather than affecting people of other races.

3.3: Could the issue be increasing number of applications per student?

Here’s an interesting fact – even though no more Americans or foreigners are applying to colleges today vs. ten years ago, Harvard is receiving twice as many applications – from about 20,000 to more than 40,000. How can this be?

The average college student is sending out many more applications.

I am not Harvard material. But when I was looking at colleges, my mother pressured me to apply to Harvard. “Come on!” she said. “It will just take a few hours! And who knows? They might accept you! You’ll never get in if you don’t try!”

Harvard did not accept me. But my mother’s strategy is growing in popularity. Part of this might be genuine egalitarianism. Maybe something has gone very right, and the average American really does believe he or she has a shot at the Ivy League. But part of it may also be a cynical ploy by colleges to improve their rankings in US News and other similar college guides. These rankings are partly based on how “selective” they are, ie what percent of students they turn away. If they encourage unqualified candidates to apply, they can turn those unqualified candidates away, and then they appear more “selective” and their ranking goes up.

But increased application volume is mostly driven by an increasingly streamlined college admissions process, including the Common Application. I didn’t like my mother’s advice, because every college application I sent in required filling in new forms, telling them my whole life story all over again, and organizing all of it into another manila envelope with enclosed check. It was like paying taxes, except with essay questions. And there was a good chance you’d have to do it all over again for each institution you wanted to apply for. Now that’s all gone. 800 schools accept the Common Application, including the whole Ivy League. From the Times again:

Six college applications once seemed like a lot. Submitting eight was a mark of great ambition. For a growing number of increasingly anxious high school seniors, figures like that now sound like just a starting point…

For members of the class of 2015 who are looking at more competitive colleges, their overtaxed counselors say, 10 applications is now commonplace; 20 is taking on a familiar ring; even 30 is not beyond imagining. And why stop there? Brandon Kosatka, director of student services at the Thomas Jefferson School for Science and Technology in Alexandria, Va., recently worked with a student who wanted a spot in a music conservatory program. To find it, she applied to 56 colleges. A spokeswoman for Naviance, an online tool that many high school students and their counselors use to keep track of applications, said one current user’s “colleges I’m applying to” tab already included 60 institutions. Last year the record was 86, she said.

Does this mean increasing competitiveness is entirely an illusion? Suppose in the old days, each top student would apply to either Harvard or Yale. Now each top student applies to both Harvard and Yale, meaning that both colleges get twice as many applicants. Since each of them can only admit the same number of students, it looks like their application rate has been cut in half. But neither one has really become more competitive!

This can’t quite be it. After all, in the first case, Yale would expect 100% of accepted students to attend. In the second, Yale would know that about 50% of accepted students would choose Harvard instead, so it would have to accept twice as many students, and the acceptance rate per application wouldn’t change.

But if more people are following my mother’s strategy of applying to Harvard “just in case” even when you’re not Harvard material, then this could be an important factor. If the number of people who aren’t Harvard material but have mothers who imagine they are is twice as high as the number of people who are really Harvard material, then Harvard admissions will triple. If Harvard accepts these people, they will definitely go to Harvard, so there is no need for Harvard to increase its admission rate to compensate. Here there really is an illusion of increasing competition.

Finally, this process could increase sorting. Suppose that, for the first time in history, a Jewish mother had an accurate assessment of her son’s intellectual abilities, I really was Harvard material, and I was unfairly selling myself short. If the existence of a Common Application lets more people apply to Harvard “just in case”, and if the Harvard admissions committee is good at their job, then the best students will get more efficiently matched with the best institutions. In the past, Harvard might have been losing a lot of qualified applicants to unjustified pessimism; now all those people will apply and the competition will heat up.

And in the past, I think a lot of people, including really smart people, just went to the nearest halfway-decent state college to their house. Partly this was out of humility. Partly it was because people cared about family and community more. And partly it was because college wasn’t viewed as the be-all and end-all of your value as a human being and you had to get into the Ivy League or else your life was over. If all these people are now trying to get into Harvard, that will increase competition too.

Can we measure this?

This is the best I can do. It shows that over the past ten years, the number of students at public universities who come from in-state has dropped by 5%. This is probably related to sorting – people working on sorting themselves efficiently will go to the best school they can get into rather than just the closest one in their state. But it’s not a very dramatic difference. I suspect, though I can’t prove, that this is hiding a larger change at the very top of the distribution.

3.4: Could the issue be that students are just trying harder?

Imagine the exact same students applying to the exact same schools. But in 2009, they take it easy and start studying for their SATs the night before, and in 2019, they all have private tutors and are doing five extracurricular activities. College admissions will seem more competitive in 2019.

Any attempt to measure this will be confounded by reverse causation – increased effort might or might not cause increased selectivity, but increased selectivity will definitely cause increased effort. I’m not sure how to deal with this.

If studying harder improves SAT scores, these could be a proxy for how much effort students putting in. They changed the test in 2016 in a way that makes scores hard to compare, but we can at least compare scores from earlier years. Scores decline between 2005 and 2015 in both math and reading. This may be because more students are taking the SAT (1.5 million in 2008 vs. 2.1 million in 2018) so test-takers are a less selected population. This is kind of surprising given that college enrollment is stable or declining, but it could be that as part of pro-equality measures, schools are pressuring more low-achieving kids to take the SATs in order to “have a chance at college”, but those students don’t really end up attending. In support of this theory, scores are declining most quickly among blacks, Hispanics, and other poorer minority groups who may not have taken the SAT in earlier years; they are stable among whites, and increasing among Asians (increasing numbers of whom may be high-achieving Chinese immigrants). At least, this is the best guess I can come up with for why this pattern is happening. But it means SATs are useless as a measure of whether students are “trying harder”.

Why might students be trying harder? If there’s a ten year lag between things happening and common knowledge that the things have happened, the explosion of college attendance during the 1990s, with an ensuing increase in competitiveness, might have finally percolated down to the average student in the form of advice that getting into college is very hard and they should work to be more competitive. In addition, the Internet is exposing new generations of neurotic parents to messages that unless their child is perfect they will never get into college and probably die alone in a ditch.

Further, the decline of traditional criteria might be causing an increasing emphasis on extracurriculars, which take a harder toll on college students. Because of grade inflation, colleges are no longer counting high school grades as much as they used to; because meritocracy is passé, they’re no longer paying as much attention to the SAT. This implies increased emphasis on extracurriculars – things like student government, clubs, internships, charitable work, and the like. Despite popular misconceptions, the SAT is basically an IQ test, and doesn’t really reward obsessive freaking out and throwing money at the problem. But getting the right set of extracurriculars absolutely rewards obsessively freaking out and throwing money at the problem. Maybe twenty years ago, you just played the IQ lottery and hoped for the best, whereas now you work yourself ragged trying to become Vice-President of the Junior Strivers Club.

But all of this is just speculation; I really don’t know how to get good data on these subjects.

3.5: Are funding cuts reducing the number of college spots available?

Some people argue that cuts in public education are reducing the number of positions available at public universities, meaning the same number of students are competing for fewer spots. This source confirms large cuts in public funding:

These universities have tried to compensate by increasing tuition (or increasing the percent out-of-state students, who pay higher tuition). It looks like they’ve done this on a pretty much one-to-one basis, so that they’re actually getting more money per student now than they did when public funding was higher.

And from California:

It’s not clear that declining state support affected enrollment at all. Colleges just raised their prices by a lot.

In 2007, 2.8x as many students were in public universities compared to private ones. In 2017, the ratio was 2.9. If the problem were limited availability of public universities to absorb students, we might expect the percent of students at public universities to go down. This doesn’t seem to be happening.

Overall it doesn’t look like funding cuts to public universities mattered very much here.

3.6: Conclusions?

The clearest reason for increasing academic competition in the past ten years is the increasing number of applications per person, enabled by the online Common Application. This has doubled the number of applications sent to top colleges like Harvard despite the applicant pool staying the same size. Some of this apparent increased competition is a statistical illusion, but parts of it may be real due to increased sorting.

Other reasons may include increased common knowledge of intense competition making everyone compete more intensely, and decreased use of hard-to-game metrics like the SAT in favor of easy-to-game metrics like extracurriculars.

4. What Has Been Happening Beyond The College Level?

Competition is intensifying.

Between 2006 and 2016, the number of applicants to US medical schools increased by 35% (note change in number of applicants, not number of applications).

In a different statistic covering different years, the number of people enrolled at medical school increased 28% from 2002 to 2017. These two numbers aren’t directly comparable, but by eyeballing them we get the impression that the number of spots is increasing more slowly than the number of applicants, probably much more slowly.

As predicted, the MCAT (the med school version of the SAT) scores necessary for admission have been increasing over time.

This is also the impression I have been getting from doctors I know who work in the medical school and residency admissions process. I got to interview some aspiring residents a few years ago for a not-even-all-that-impressive program, and they were fricking terrifying.

Law schools keep great data on this (thanks, law schools!). US News just tells us outright that law schools are less competitive than in 2008, even at good programs. Here’s the graph:

And despite it feeling like lawyers are everywhere these days, law school attendance has really only grown at the same rate as the population since 1970 or so, and dropped over the past decade. This may be relating to word getting out that lawyer is no longer as lucrative a career as it used to be.

Unlike law schools, graduate school basically fails to keep any statistics whatsoever, and anything that might be happening at the graduate level is a total mystery. We know the number of PhDs granted:

…and that’s about it.

Part of what inspired me to write this post was listening to a famous scientist (can’t remember who) opine that back when he was a student in the 1940s, he kind of wandered into science, found a good position at a good lab, worked up the ranks to become a lab director, and ended up making great discoveries. He noted that this was unthinkable today – you have to be super-passionate to get into science grad school, and once you’re in you have to churn out grant proposals and be the best of the best to have any shot at one day having a lab of your own. I’ve heard many people say things like this, but I can’t find the evidence that would put it into perspective. If anyone knows more about the history of postgraduate education and work in the sciences, please let me know.

I’m also interested in this because it would further help explain undergraduate competition. If more people were gunning for med school and grad school, it would be more important to get into a top college in order to have a good chance of making it in. Since increasing inequality and returns to education have made advanced-degree jobs more valuable relative to bachelors-only jobs, this could explain another fraction of academic competitiveness. But aside from the medical school data, I can’t find evidence that this is really going on.

5. Is Freaking Out Over College Admissions Correct?

Dale and Krueger(2011) examine this question, using lifetime earnings as a dependent variable.

In general, they find no advantage from attending more selective colleges. Although Harvard students earn much more than University of Podunk students, this is entirely explained by Harvard only accepting the highest-ability people. Conditional on a given level of ability, people do not earn more money by going to more selective colleges.

A subgroup analysis did find that people who started out disadvantaged did gain from going to a selective college, even adjusted for pre-existing ability. Blacks, Latinos, and people from uneducated families all gained from selective college admission. The paper doesn’t speculate on why. One argument I’ve heard is that colleges, in addition to providing book-learning, help induct people into the upper class by teaching upper-class norms, speech patterns, etc, as well as by ensuring people will have an upper-class friend network. This may be irrelevant if you’re already in the upper class, but useful if you aren’t.

A second possibility might be that college degrees are a signal that help people overcome statistical discrimination. Studies have shown that requiring applicants share drug test results or criminal histories usually increases black applicants’ chances of getting hired. This is probably because biased employers assume the worst about blacks (that they’re all criminal drug addicts), and so letting black applicants prove that they’re not criminal drug addicts puts them on more equal footing with white/Asian people. In the same way, if employers start with an assumption of white/Asian competence and black/Latino incompetence, selective college attendance might not change their view of whites/Asians, but might represent a major update to their view of blacks/Latinos.

Dale and Krueger also find that the value of college did not increase during the period of their study (from 1976 to 1989).

Does this mean that at least whites and Asians can stop stressing out about what colleges they get into?

What if you want to go to medical or law school? I can’t find an equally rigorous study, but sites advising prospective doctors tell them that the college they went to matters less than you’d think. The same seems true for aspiring lawyers. As usual, there is no good data for graduate schools.

What if you want to be well-connected and important?

From here, the percent of members of Congress who went to Ivy League colleges over time, by party:

Only about 8% of Congresspeople went to Ivy League colleges, which feels shockingly low considering how elite they are in other ways. The trend is going up among Democrats but not Republicans. There is obviously a 40-50 year delay here and it will be a long time before we know how likely today’s college students are to get elected to Congress. But overall this looks encouraging.

On the other hand, presidents and Supreme Court Justices are overwhelmingly Ivy. Each of the last five presidents went to an Ivy League school (Clinton went to Georgetown for undergrad, but did his law degree at Yale). Every current Supreme Court justice except Clarence Thomas went to an Ivy for undergrad, and all of them including Thomas went to an Ivy for law school. But there’s no good way to control for whether this is because of pre-existing ability or because the schools helped them succeed.

Tech entrepreneurs generally went to excellent colleges. But here we do have a hint that this was just pre-existing ability: many of them dropped out, suggesting that neither the coursework nor the signaling value of a degree was very important to them. Bill Gates, Mark Zuckerberg, and Larry Ellison all dropped out of top schools; Elon Musk finished his undergrad, but dropped out of a Stanford PhD program after two days. This suggests that successful tech entrepreneurs come from the population of people smart enough to get into a good college, but don’t necessarily benefit from the college itself.

Overall, unless people come from a disadvantaged background, there’s surprisingly little evidence that going to a good college as an undergraduate is helpful in the long term – except possibly for a few positions like President or Supreme Court justice.

This doesn’t rule out that it’s important to go to a good institution for graduate school; see this paper. In many fields, a prestigious graduate school is almost an absolute requirement for becoming a professor. But there doesn’t seem to be an undergrad equivalent of this.

Digression: UC schools

I mentioned at the beginning the universal perception in California that UCs are much harder to get into. I know this is the perception everywhere, but it seems much worse in California. Sure, it’s anecdotal evidence, but the anecdotes all sound like this:

My friend’s daughter got 3.85 GPA, had 5 AP classes in high school, was on competitive swimming team, volunteered 100+ hours, was active in school activities, yet she got rejected by all 4 UCs that she applied to. And these were not even the highest tier of UCs, not Berkeley. She did not apply for more schools and thought that UC San Diego and UC Santa Cruz were her safe choices. The whole family is devastated.

The data seem to back this up. Dashed line is applications, dotted line is admissions, solid line is enrollments:

…but I don’t know how much of this is just more applications per person, like everywhere else.

Why should UC schools be hit especially hard? I assumed California’s population was growing faster than the rest of the country’s, but this doesn’t seem true: both California and the US as a whole grew 13% between 1990 and 2000, when the cohort attending college between 2008 and 2018 would have been born.

The Atlantic points out that, because of budget cuts, UC schools are admitting more out-of-state students (who have to pay higher tuition), lowering the number of spots available to Californians. But is this really that big an effect?

It looks like nonresidents went from 6% to 12% over the space of a decade. That shouldn’t screw things up so badly.

I’m really not sure about this. One possibility is that California’s schools are remarkably good. On’s list of best colleges, four of the top ten schools are UCs, plus you get to live in California instead of freezing to death in New England. Since the college admissions crisis is concentrated at the top schools, California has been hit especially hard.

I’m not satisfied with this explanation; let me know if you know more.

6. Conclusions

1. There is strong evidence for more competition for places at top colleges now than 10, 50, or 100 years ago. There is medium evidence that this is also true for upper-to-medium-tier colleges. It is still easy to get into medium-to-lower-tier colleges.

2. Until 1900, there was no competition for top colleges, medical schools, or law schools. A secular trend towards increasing admissions (increasing wealth + demand for skills?) plus two shocks from the GI Bill and the Vietnam draft led to a glut of applicants that overwhelmed schools and forced them to begin selecting applicants.

3. Changes up until ten years ago were because of a growing applicant pool, after which the applicant pool (both domestic and international) stopped growing and started shrinking. Increased competition since ten years ago does not involve applicant pool size.

4. Changes after ten years ago are less clear, but the most important factor is probably the ease of applying to more colleges. This causes an increase in applications-per-admission which is mostly illusory. However, part of it may be real if it means students are stratifying themselves by ability more effectively. There might also be increased competition just because students got themselves stuck in a high-competition equilibrium (ie an arms race), but in the absence of data this is just speculation.

5. Medical schools are getting harder to get into, but law schools are getting easier to get into. There is no good data for graduate schools.

6. All the hand-wringing about getting into good colleges is probably a waste of time, unless you are from a disadvantaged background. For most people, admission to a more selective college does not translate into a more lucrative career or a higher chance of admission to postgraduate education. There may be isolated exceptions at the very top, like for Supreme Court justices.

I became interested in this topic partly because there’s a widespread feeling, across the political spectrum, that everything is getting worse. I previously investigated one facet of this – that necessities are getting more expensive – and found it to be true. Another facet is the idea that everything is more competitive and harder to get into. My parents’ generation tells stories of slacking off in high school, not worrying about it too much, and knowing they’d get into a good college anyway. Millennials tell stories of an awful dog-eat-dog world where you can have perfect grades and SAT scores and hundreds of hours of extracurriculars and still get rejected from everywhere you dreamed of.

I don’t really have a strong conclusion here. At least until ten years ago, colleges were harder to get into because more people were able to (or felt pressured to) go to college. The past ten years are more complicated, but might be because of increased stratification by ability. Is that good or bad? I’m not sure. I still don’t feel like I have a great sense of what, if anything, went wrong, whether our parents’ rose-colored picture was accurate, or whether there’s anything short of reversing all progress towards egalitarianism that could take us back. I’m interested to get comments from people who understand this area better than I do.

This entry was posted in Uncategorized and tagged , , . Bookmark the permalink.

363 Responses to Increasingly Competitive College Admissions: Much More Than You Wanted To Know

  1. humbug1994 says:

    “My parents’ generation tells stories of slacking off in high school, not worrying about it too much, and knowing they’d get into a good college anyway. Millennials tell stories of an awful dog-eat-dog world where you can have perfect grades and SAT scores and hundreds of hours of extracurriculars and still get rejected from everywhere you dreamed of.“

    I can only offer a personal anecdote to combat the millenials’ sentiment. I’m a 24 year old (white male) who graduated from a state college in 2017 and I’m heading to an Ivy League school for my master’s degree this fall.

    I didn’t really try in my small-ish rural high school, did some extracurriculars like Science Olympiad because I was genuinely interested in it, not because I wanted to beef up my ECs, and played sports, because that’s what one did at my high school. I finished 11th out of ~165 students, got a 740M/710V/530W on the SAT, and headed off to a good state school (top ~100 university) because that’s all my parents and I were willing to pay for.

    I did poorly (2.75 GPA) in my first semester of college because I was used to my uncompetitive, rural, small-ish high school. By junior year I had learned how to effectively study and got straight As during the remainder of college, giving me a 3.55 at graduation. The only real ECs I had during college were as a TA for a tough course and as a research assistant for one semester (didn’t even come close to publishing anything). I took the GRE and got 167V(98%ile)/160M(74%ile)/4.5W(82%ile). As I mentioned, I’m heading to an Ivy League school for a STEM master’s this Fall.

    I guess this is all to say that you can still glide through most of school without killing yourself and get into an elite graduate school (just like yesteryear) so long as you have reasonably high standardized test scores. I’d wager that most people you hear about getting rejected even though they have a high GPA and a ton of ECs have shitty test scores. By design, very few people have high standardized scores. If you have high test scores, good grades, and strong ECs AND you get rejected… Well then you’re probably Asian. Not even kidding.

    I do consider myself very lucky for getting into the graduate school that I got into. I know college admissions can be a crapshoot. And I’m lucky that my low conscientiousness didn’t bite me. But I’m proof (n=1) that you don’t necessarily need to do anything over-the-top to get into a really good school.

  2. kwr says:

    Admission to elite colleges is not nearly as competitive as the data suggests or people believe. This has become a scenario where the free market isn’t very efficient because there are so many choices and too much information for consumers to digest. Unfortunately, the college rankings and the Common App aren’t bringing order to this chaos. They are just stirring up a mania that encourages trends like intensive parenting and drives demand for college counseling and test preparation services. All of these attempts to game the system then have to be identified and discounted by admissions officers. That’s an imperfect science that leads to imperfect outcomes that families compare to the apples to oranges rankings of colleges and reach all sorts of bad conclusions. The volume of bad conclusions is amplified in places like College Confidential because more and more of these competitive students are now found in the same high schools such that families can compare outcomes. These discussions always start with a mention of test scores, GPA, AP’s and other quantitative measures which have already been discussed here as not being that useful to admissions in discriminating between applicants at elite colleges. All of the reasons for that, like grade inflation, differences in resources, etc., have already been mentioned here.

    There aren’t more competitive applicants out there. There are just more applicants with high test scores and tons of extracurriculars. But those EC’s are the same EC’s as every other student. And too often they are academic related, the one quality where the student already has quantitative proof of their ability. No student is a better candidate because they studied harder for the SAT the second or third time they took it. A really bright student does not benefit from participating in Science Olympiad. That student should be off doing their own thing by high school except for cases where something like leadership is their thing and they are leading other students.

    In a sense, there are fewer competitive applicants out there. The college admissions mania is driving too many students to spend their time in high school unproductively. They miss out on activities that develop soft skills for example. They are basically gambling on admission. And if they aren’t admitted to that selective school, they now head to their safety school with fewer soft skills and a decreased chance of standing out at that college. They get less practice in high school accomplishing goals for which there are no lists. There is no list for how to invent Google or Facebook. There are too many other qualities to list here that also fail to appear in abundance on apps.

    In one of the other replies there was a mention that fencing gave a student better odds than other sports for college admission. That is a good analogy for the strategy many families take with admissions. The problem with that approach is that a coach can sniff out the athlete that isn’t really passionate about the sport. And in the Ivies, for example, that athlete can just quit after the first day of practice because they are not there on athletic scholarship. Most coaches can sniff out the players that aren’t there for the right reasons and those players don’t get offers. And that’s probably become a large part of what admissions officers are trying to do with applicants. The coach doesn’t make this decision based on what the player tells him; it’s based on what the player doesn’t tell him. In other words, it isn’t captured in any data.

    The data most people are familiar with isn’t giving us a good picture of what is happening in college admissions. And most people aren’t aware of data like the declining number of students or that those declines are steepest on the coasts. We need to start a new conversation that leads to some different priorities.

    • Of possible relevance to this whole thread …

      I interviewed four or five Harvard applicants as part of the admission process (I’m an Alum). None of them were intellectually impressive. I don’t think any of them were accepted, so that is evidence about who applies, not who gets in.

      • kwr says:

        Exactly. I have heard it said by a number of adcoms (and I’m fairly certain Harvard was one of them) that the first half of the class is easy to fill. The second half of the class is filled from a stack of applications that all look the same. So you can imagine how confusing the outcomes are to families trying to compare applicants based on quantitative measures of merit. Or families trying to decide whether Science Olympiad is valuable. Most of the time, they end up bench marking to students in the second half of the class because they assume from the numbers that those students were among the top students admitted.

  3. Owen George says:

    A few additional thoughts on the underlying reasons for college admissions at elite universities getting more competitive:
    1. Income segregation for the wealthy has been going up for the last few decades, especially post 2000. That means that a lot of kids with parents who care a lot about elite colleges are now going to the same high schools. That has to accelerate the competition over how many extra-curriculars, AP’s classes, and test prep classes each kid takes.
    2. If you look at Gunn High School in Palo Alto, (featured in the link about stressed out kids committing suicide in Scott’s post), the average SAT score has gone from 1288 in 2007 to 1369 in 2017. (Greenwich, CT, another epicenter of affluence segregation, also had an increase over the same time period, but a smaller one). So wealthier couples, with smarter, more pressured, and more prepped kids, have moved into these neighborhoods. At the very least, that kind of concentration has to increase anxiety and focus on college admissions.
    3. The Palo Alto effect might explain some of the focus on this issue in the bay area. The fact that the University of California system has had a 50 point increase in their average SAT’s 2001-2014 might also contribute.
    4. The wealthy are now spending much more on education, which means that they are now willing to send their kids to expensive camps, hire tutors, and do lots of things that up the competition for elite colleges. From an aeon article: “In fact, top 1 per cent spending on education has increased 3.5 times since 1996, while middle-income spending on education has remained flat over the same time period.”
    5. High school kids are spending a decreasing amount of time working for pay outside of school. So it’s not surprising that “upper- and middle-class students have become more active in school clubs and sports teams over the past four decades”

  4. naath says:

    I guess at least part of it is “how many people would attend Harvard if the could” I attended Cambridge (the original) and in 1600 most people in England never mind the world would likely never even think of traveling that far; these days Cambridge takes students from all over the world. I thought about (couldn’t afford, didn’t apply) Harvard, and thought nothing of applying to places all over the country. Where once your Harvard aspiring New England farmer was competing mostly with other New Englanders he is now competing just as much with the brightest young things of Beijing and Sidney because “flights to the US” are available for almost nothing compared to a Harvard education.

  5. suitengu says:

    A couple of typos: “about about” and “may because”.

  6. aesthesia says:

    If I remember correctly, the Dale and Kreuger paper is somewhat limited in its conclusions. Their sample was weighted toward the high end of the achievement spectrum, so it really says something like: conditional on being accepted to Harvard, there’s not much difference in lifetime earnings between actually attending Harvard and instead choosing to attend UC Berkeley. There aren’t a lot of students admitted to Harvard who instead choose to go to the University of Southern North Dakota at Hoople, so we don’t really know what happens to them.

    This is a somewhat personal issue for me: I read summaries of Dale and Kreuger’s earlier work when applying for college and decided not to bother applying to selective schools, and just went to a middling state university, thinking that going somewhere more selective wouldn’t make a difference in my future life. I’m no longer confident that was the right move. I believe I would have learned more and made better and stronger connections had I gone somewhere more difficult to get into.

  7. Quintus Fabius Minimus Cunctator says:

    I recently got accepted into Harvard and I think my being not-even-fluent in Latin was a major part of that.
    I don’t think becoming fluent in Latin is THAT much harder than becoming fluent in some living language, but I would guess you could look much more impressive in the Latin world (eg NJCL, NLE) by being 75% fluent in Latin, than in the French world (idk French contests) when 75% fluent in French.

    So if you want to get into Harvard, you could just try following the admissions criteria of the 1640s. You’d be better than me; I don’t know Greek at all, and my Latin poetry composition is sub-par, to say the least.

    (I’m even Asian, so…)

  8. futatsuiwa says:

    I wonder if the college admissions problem is actually in some way isomorphic to the online dating problem.

    The main differences are in parameters – there are much fewer recipients of affection, but they are able to accept many more requests for it.

    I can’t really flesh out that mapping at the moment, but my thoughts bring up these strategies from various games:

    – The “futatsuiwa applies to college game”, in which instead of shotgun applying to colleges, i signaled high interest in my school of choice (free tuition and my advisor had a pretty good contact in their admissions) and applied with the so-called “Early Decision” process.
    – The Jackbox game “Monster Seeking Monster” (because my dating experience isn’t robust enough to draw upon), which in each round gives you four opportunities to message the other players and subsequently attempt to match with one of the recipients. Messaging a single player with all of your messages means that your only choice is to pick that player in the following decision where everyone picks a player to go out with, which also signals the other player that you have committed to select them.

    In the future, I’d expect ever-increasing amounts of jumping through hoops and, as in the previous article on dating, dealing with pain, to show how the prospective student is SO dedicated to this school and totally aren’t applying to 50 other schools via the magic of the internet, or if they are it’s not serious and only because they were told to apply to backups.

  9. jaimeastorga2000 says:

    The subreddit is right; this is your best post since “Considerations on Cost Disease” (2017).

    Part of what inspired me to write this post was listening to a famous scientist (can’t remember who) opine that back when he was a student in the 1940s, he kind of wandered into science, found a good position at a good lab, worked up the ranks to become a lab director, and ended up making great discoveries. He noted that this was unthinkable today – you have to be super-passionate to get into science grad school, and once you’re in you have to churn out grant proposals and be the best of the best to have any shot at one day having a lab of your own.

    You are thinking of James Watson. From “The id of the basic income” (2014) by Athrelon:

    But until very recently, they had another option. Even after the death of the old aristocracy, there was a pretty guaranteed way to live a low-key life of the mind; you could join academia. It’s only quite recently that academia as a whole was so difficult and Darwinian to try to break into. James Watson was a mediocre college student until he read What is Life, got excited, and decided to go to grad school in Indiana to work on it. Today he wouldn’t stand a chance against students who had spent undergrad getting good grades, working with the right mentors, and generally jumping through the right hoops. Academia is now a rat race like any other, and if you want that career, you have to apply your nose firmly to the grindstone and not spend too much time looking around.

  10. alexhutcheson says:

    I think this analysis is excellent, but I don’t think the Common App explains the increase in applications-per-admission. There’s another significant factor that causes students to apply to dozens of schools: No one can accurately forecast what a given school will cost anymore.

    At some point (I believe in the late ’00s, but could be wrong), elite schools started to extend their financial aid programs to include students from middle-class families. Here’s a press release from Harvard in 2007. Yale and Princeton followed quickly, and most other elite schools seem to have done similar things, albeit with more constrained resources.

    Prior to this change, a student from a family with middle-class income or above could know with reasonable certainty what a given college would cost: they would expect to pay the tuition, fees, etc. listed on the brochures. After this change, they would have no idea until after they were admitted. They might get a generous financial aid package that brings the cost down to the price of their local state university, or they might get nothing except loans. The systems used to determine financial aid packages are opaque and not well-publicized, so the outcome is unpredictable.

    In 2009, I applied to a broad sample of 14 schools on the east coast. They were a mix of “elite” private and flagship public universities. My parents were comfortably middle-class. I was lucky enough to get into most of them, and so I had the opportunity to compare financial offers. In maybe 1 case out of 12, I would have had to pay the full sticker price. The rest would have been heavily “discounted”, but the discount varied widely between schools, from ~5% off the total cost of attendance, to 60% off, to 100% off (full ride). For the private universities the “discount” came from a mix of “need-based aid” and merit scholarships, while for the public universities it was exclusively from merit scholarships.

    In this system, you don’t know what the financial offer is until you get in, there are possible windfalls from getting a generous financial offer, and it’s difficult to predict in advance what the financial offer will be until you get in. The incentives here are obvious: student who are conscious about the cost of their education have a strong incentive to “play the lottery” by applying to as many schools as is feasible for them, while biasing towards schools that are known to provide generous financial packages.

  11. DragonMilk says:

    Top schools actually have very generous grants and need-blind admissions

  12. Manx says:

    The obvious solutions seems to be some form of capping the number of applications. You could do this for the ‘common application’ if the schools involved agreed to it. This could benefit less prestigious schools, because people will always save a slot or two for their ‘safeties’ and it could benefit more prestigious schools by cutting down the number of applications they need to process. People will be less likely to ‘waste’ their application spot on a school they have next-to-no chance of getting into. Part of the issue might be schools want to look more selective, so getting a bunch of applications from people without a shot might be good for them. In which case, they’re not going to agree to this.

    Another option would be having a rank system like residency programs, however that is complicated by colleges offering money to sweeten the deal after the application is received, making ranking the program before knowing those things very difficult.

  13. Anthony says:

    The debt thing is a big change from when I went to college, but it’s driven by policy more than by parents.

    The government encouraged student loans because those would get paid back while subsidizing tuition at the student end or the college end wouldn’t (directly), and since they were paid back, the had very little budget cost. With a relatively significant college premium, people were willing to take on fairly large amounts of debt (and were not very likely to default). This allowed colleges to spend more, driving up the sticker price, making loans more necessary for students who weren’t poor enough to get direct subsidies or rich enough to afford the higher price.

    Unfortunately, this is a politically very difficult problem to unwind. Even though the best policy would be to stop subsidizing student loans entirely and make them dischargeable in bankruptcy, which would hugely limit the availability of student loans, that’s never going to happen.

  14. Steve Sailer says:

    Robert Heinlein’s 1950s juvenile novels are full of scenes where people ask the hero what he’s going to do for college and he says, “I dunno, I guess the day after Labor Day I’ll drive down to State and sign up for some classes.” (That was pretty much what UCLA was like when Heinlein went there for awhile in the 1930s.) But in the end of “Have Space Suit, Will Travel,” the hero gets accepted into MIT when the head of the CIA, or whatever, pulls some strings. So, like it says in The Bell Curve, attitudes toward going off to an elite national college were changing in the 1950s.

    • brmic says:

      Err, ‘The Bell Curve’ is clearly culture war territory and referencing it doesn’t actually add anything to your post, yet you saw the need to do it twice in the span of a single comment thread and within two hours. Care to explain why?

      • SamChevre says:

        I think you haven’t read “The Bell Curve”; it’s central thesis is that the kind of elite we have changed in the 1950’s in a way that made elite national colleges central.

      • Steve Sailer says:

        Because the 1994 “Bell Curve” consists of 745 pages of time-tested quarter-century old empirical data that is obviously relevant to college admissions questions, such as whether Americans in the 1950s became more amenable to moving far away for college.

      • broblawsky says:

        I think he’s referring to the classist first part of The Bell Curve, which claims that American society has been divided into a “cognitive elite” and a lower-IQ “everyone else”, rather than the racist second part.

        • Steve Sailer says:

          If you want to understand the way the world works in the 21st century, you should read “The Bell Curve” from the 20th century.

  15. jumpinjacksplash says:

    I think what people may be missing is that time is a thing:

    In a world where there are 10 places at Harvard every year but 9 applicants, everyone who can read Latin gets in, and no-one bothers to study harder than what it takes them to meet the minimum requirement.

    In a world where there are 10 places at Harvard but 11 applicants, the worst Latin-reader doesn’t get in. But someone who realises they’re the worst Latin readers will practice more, and someone else will be the worst. That person will then practice enough to make someone else be the worse, who then practices a bit more. Given enough time, the standard keeps rising as everybody needs to stay ahead of everybody else; eventually everyone hits the ceiling of Latin literacy and starts taking up the violin, then competes on violin-skills until they all have to start climbing Everest.

    The only equilibrium is where everyone exerts maximum effort on top of their abilities. It takes a long time for this to become even remotely necessary (I still don’t think we’re close to being there). But a world where the first 18 years of your life consists of maximum effort to get into college is a dystopian nightmare, hence as we trend towards it everything gets constantly worse.

    This also explains the anecdotal people who expect to swim into Berkeley but get rejected from UCLA: their parents think they’ve massively overshot what’s required, but they’ve failed to realise how far down the slope we’ve slid.

    • Aapje says:

      This development may be moderated by a lack of knowledge. The person who could have competed, but didn’t because they falsely believed that their effort was good enough or alternatively, are unaware of how they can reach the threshold, keeps the threshold lower than if they maximized their chances.

      A major cause for the rapid increase in standards might be a relative lack of uninformed people, due to the Internet. In the past it would take many years or even generations for people to figure out that the standards increased. In many cases, people may have aimed for the standards of the past.

      Today, people may be far more aware of the standards of today. The rapid increase in standards may have made people aware of this increase, which makes it logical to start extrapolating and aiming for the standards of tomorrow. When people do that, the feedback loop really accelerates.

    • johan_larson says:

      Maybe the answer is to care a little less. It’s not particularly difficult to go to college in the US. It’s not even very difficult to go to a college where there are capable profs and where you can learn a lot if you study diligently. What’s difficult is going to a place that is actually prestigious. The reason people are fighting so very very hard is mostly about prestige. And apparently that prestige doesn’t translate to money, as Scott mentions in his article (section 5.). Perhaps it’s time to wonder whether the rewards offered are proportionate to the demands being made.

    • raj says:

      Normally you think of these things as a race to the bottom, this sounds more like a race to the top. Not clear to me that that is a bad thing.

      The mental health of those with overly competitive all-or-nothing attitudes may be at risk. But quite frankly it seems to me that “young people today” are getting more lazy and complacent, not less. (myself included). I don’t think even harvard is turning away violin playing mountain climbers just yet.

      • johan_larson says:

        They are saying no to some really accomplished young people.

        A couple of weeks after speaking to Taylor, I drove to the campus of Williams College in Massachusetts to meet Michael Wang, a student there, at a café near the main student center. Over the past several years, Wang has served as a vocal poster child for alleged discrimination against Asians in college admissions. He scored a perfect 36 on the ACT entrance exam, placed third in a national piano contest and first in California for a math competition, competed in national debate tournaments as a finalist, graduated second in a class of more than 1,000 students, and sang in the choir at Obama’s 2009 inauguration. Yet out of the seven Ivy League schools to which he applied, only the University of Pennsylvania accepted him, which he holds as proof of rampant racism in the admissions process.

        • John Schilling says:

          I suspect math competition and classical musical performance may have crossed the Goodhart line, in that they are the specific things smart but dull people are forced to do by their Tiger Mom parents to “prove” that they are Very Smart and Broadly Talented. Possibly also debate, but less confident of that.

          If so, Wang looks academically brilliant but otherwise uninteresting. I don’t want to speak for Ivy League admissions officers, but I’d rather have the kid with the 35 ACT who tinkers with old cars or spent the year after college backpacking around the world or produces a webcomic, presuming demonstrated excellence in those fields. The bit where there aren’t unambiguous metrics for excellence in those fields is both a blessing and a curse in this regard.

          • meh says:

            or spent the year after college backpacking around the world

            I’m sure much of the admissions process could be fixed if they could predict the future.

          • johan_larson says:

            Heh. I can just see some poor college admissions counselor struggling to explain to some ABC youth and his FOB parents how the American concepts of “cool” and “edgy” apply to the college admissions process and just what sort of precisely calibrated mix of conformity and rebelliousness should be cultivated for optimal effect.

            Yes, yes, playing music is good. But not classical music. That fits rather too neatly into a stereotype of you as the obedient son. You want something more rootsy and American. Learn to play blues. And by all means look for competitions. That part is fine.

            As for sport, you put down soccer. That’s again too conventional. It’s so terribly earnestly suburban. You need something edgier, with a whiff of danger. Find yourself a boxing gym. And make sure you get plenty of pictures of yourself with the black fighters there.

          • Clutzy says:

            Sports I think appear to be underrated here/by parents. Look at the admissions scandal. Athletics is the backdoor to getting in. A lot of elite institutions dont have elite athletics programs. If you can plausibly be a walk on for the Harvard girls basketball team you essentially get a pretty big bump.

            Plus you need to exercise anyways, so why not do it in an enjoyable way?

          • meh says:

            These guys say you should pick fencing:

            yeah, i mean what 17 year old is interesting? Certainly singing for the president, and wining a math competition puts you as more interesting than almost all 17 year olds. And you can’t tiger mom your way to first in a math competition… maybe scoring high, but you need some talent to win the thing. And these are all merit based things the kid did… not ‘my parents were rich enough to buy me classic cars and pay for me to backpack around the world’

          • Clutzy says:

            That transcript is really cute. Its amazing how deluded most athletes are about how much of it is genetics. Without fail every elite athlete talks about how hard they work (despite it mostly being impossible to dedicate more than 3-4 hours to your craft just physically).

            But the fencing thing is pretty smart. But I would argue something that trends against his idea: fencing is likely to attract mostly the children of smart/successful people who are also likely to be smart, based on genes and environment. OTOH, there are lots of sports where participation is pretty high, but most of them aren’t all that smart. Or at least participants aren’t almost all from the smart-person crowd.

            Basketball, football, soccer, etc all have so many participants that there are probably better players that are smart enough to get into school than you (Jeremy Lin at Harvard for instance), but other more niche sports like golf and tennis, are more popular for rich people, thus you are also likely competing with other Harvard-adjacent players. A sport like wrestling, however, isn’t a high socioeconomic sport, but also has smaller participation. My brother was a good student, but not at the level of me or my sister, but he got admitted to some pretty elite schools based on the fact that he was a 2x state placer in wrestling, including runner up his sr year, and 2x winner of a premier midwest midseason wrestling tournament.

          • Nornagest says:

            I fenced competitively in college, and I did pretty well, but if I’d started in high school and tried to parley it into an athletic scholarship they would have, first, laughed, and then told me to fuck off. The school didn’t even pay for our equipment.

            I’m sure it would have been an okay extracurricular (about all I had there was a smattering of the usual bullshit and a brown belt in $OBSCURE_MARTIAL_ART, the more fool I), but I can’t imagine it being much better than, say, math team.

          • Clutzy says:


            From my EXP Math team has little impact in college admissions. They have the SAT/ACT math for that. I wont 2 statewide competitions in Math and Chemistry and it wasn’t worth anything as far as I can tell because they get the same info from my SAT, ACT, and AP scores in those subjects. OTOH the fact that I was all conference in wrestling and soccer was very interesting to alumni interviewers at a few elite colleges that I go into, but ended up not attending due to the scholarships I got for pure academics elsewhere.

          • Nornagest says:

            Right, I’m saying it’s basically equivalent to any other UMC-sounding time-waster activity. I believe you about wrestling, but that’s, first, a much bigger sport in the US though not on the level of football/basketball, and, second, not UMC coded.

          • Clutzy says:

            UMC coded?

          • Nornagest says:

            Upper middle class. Bougie. SWPL.

          • meh says:

            I wont 2 statewide competitions in Math and Chemistry and it wasn’t worth anything as far as I can tell because they get the same info from my SAT, ACT, and AP scores in those subjects.

            I guess it depends on the competition, but the SAT cap of ‘perfect score’ makes it hard to compare kids at the top end.

          • Clutzy says:

            I guess it depends on the competition, but the SAT cap of ‘perfect score’ makes it hard to compare kids at the top end.

            I agree (obviously because it would have benefited me), but the people who do admissions clearly (at least somewhat clearly) do not. The spelling bee, which is the most high profile of such events, obviously recognizes this because they give kids a scholarship/money regardless of where you attend because they know that Harvard/Yale won’t just accept winners. Otherwise winning the Bee would be rewarded with admission and full ride to any university of the person’s choice. Which would be deserved because that is an insane achievement.

          • meh says:

            I would guess the spelling bee is higher profile because of popular appeal, not because it is a better predictor of success. I’m also skeptical that it wouldn’t help your application, and distinguish you from other 800 verbal scorers.

  16. Steve Sailer says:

    I wrote a column back in 2013 documenting that famous colleges were not expanding their freshmen class sizes at anywhere close to population growth over the last generation:

    Consider the growth rate of Harvard, the world’s richest university. The number of undergraduates in its class of 1986 was 1,722. After a quarter of a century, during which the US population grew by 75 million, Harvard’s class of 2011 was 1,726: an increase of four.

    This is not to say that Harvard isn’t expanding: Faculty and grad students are up, and non-teaching staff skyrocketed.

    Similarly, Yale’s undergraduate student body has been the same size since 1978. Five years ago, the second-richest college announced a proposal for adding a couple of dormitories, but construction won’t proceed until another $300 million is raised.

    In 2010, MIT unveiled plans to expand undergrad enrollment by six percent, which would only get it back to where it was in the 1990s.

    Among the most prominent colleges, Princeton is the only one over the last generation to have actually succeeded in boosting enrollment (and that by only about ten percent) after it opened the Whitman residential college in 2007. …

    Perhaps the defining activity of American life since the 1960s has been elites conspiring to become more elite.

    I presume that there have been some increases in the size of the student body at the most famous colleges since I wrote six years ago: Stanford, for example, with its 8,000 acre campus and vast endowment, appears to have grown a few percent.

    But still …

    • Aapje says:

      Perhaps it is very hard to run a very large university?

      The better question might be why the number of Ivy League universities didn’t grow…

      • Steve Sailer says:

        My impression is that the people running the moderately large elite universities like Harvard and Yale are, on the whole, very good at doing what the people who select them want them to do. But, unlike in most businesses where the Board wants CEOs to grow the business, the boards of trustees of famous colleges seldom care much if the presidents grow the undergraduate student body. Growth is for the Arizona States of the world, not for the superstar colleges.

        • faoiseam says:

          My impression is that the people running the moderately large elite universities like Harvard and Yale are, on the whole, very good at doing what the people who select them want them to do.

          In my experience, the Presidents of universities often act in ways the trustees did not expect. You get a perfectly reasonable candidate, and after a few years they go on some crazy quest to change the school. This happened to USC, as far as I can tell, though I am not associate with it. The Trustees liked USC as it was, a university for spoiled children. The President decided to focus on diversity and raising the colleges academic profile. Olivia Jade was on the yacht of the Chairman of the Board of Trustees when the scandal broke, which suggests the trustees still like the old model better.

          • Steve Sailer says:

            The Chairman of the Board of Trustees of USC is Rick Caruso, the best real estate developer in Southern California. It’s incredibly hard to get anything built in SoCal in the last 50 years, but Caruso, more than anybody else, can get it done.

  17. physticuffs says:

    …he kind of wandered into science, found a good position at a good lab, worked up the ranks to become a lab director, and ended up making great discoveries. He noted that this was unthinkable today – you have to be super-passionate to get into science grad school, and once you’re in you have to churn out grant proposals and be the best of the best to have any shot at one day having a lab of your own.

    I’m a first-year grad student now so I can’t speak to attempting anything beyond…just getting into the program yet, but this definitely isn’t my impression. I went to a quite good, but not very famous, liberal arts college. Everyone in my physics program basically assumed that if they felt like going to grad school, they could get in. Maybe not to their first choice, but definitely to at least one good-enough program. (I bet part of this is because state universities tend to be “better” at the graduate level than the undergrad level, so the availability of programs that are really good is higher.) We were encouraged to be super passionate about science to want to do science grad school in the first place, but that was mostly due to the high time investment/low pay of Ph.D. programs, not because we had to be much better than everyone else. I know plenty of grad students here who are significantly less passionate and invested than the above quote makes it seem like you have to be; some people seem to be in this program out of a sense of “eh, seems like a good next step”. And it was definitely easier than getting a job right out of undergrad. I applied to seven grad schools and was accepted by three. This is a way better rate than what friends have described for job applications.

    Anecdotally, I now work in a lab run by a professor who’s well-known in the field. (A talk he gave was recently put on the TED website, for one thing.) The lab is relatively competitive to get into at the moment and he had his pick of top candidates applying to our university, but it wasn’t that way four years ago when he started the lab–his first four students were guys who sort of stumbled into it by some combination of coincidental interest and being just-barely-admitted to the program. They built a majorly successful research team. Wandering into a good lab is still viable, it’s just that you don’t always know what the good labs are in advance.

  18. meh says:

    I posted upthread that Scott’s Common Ap conclusion is contradicted by the pew article he links:

    They claim:

    The expansion of the Common Application, which makes it easier for students to apply to multiple schools, doesn’t appear to be behind the increase in application volume. The Common App, as it’s called, is accepted by nearly 800 colleges and universities in the United States and several dozen overseas. Of the 1,364 institutions in our sample, 729 accept the Common App along with (or in some cases instead of) their own application forms; the other 635 use their own forms. Although one might suspect that the ease of applying to multiple schools via the Common App would result in stronger growth in application volume among those schools, there was almost no difference in 2002-2017 growth rates between the schools that used the Common App and those that didn’t.

    But is their reasoning sound? Isn’t it possible the common ap increases applications uniformly among all schools, not just ones using the common ap?

    Consider if pre-common ap, I am willing to fill out 10 applications, so I apply to 10 schools. Now with the common ap, say half of them accept it, so I only need to fill out 6 applications. I am still willing to fill out 4 more applications, so I look at my 11th choice. If they take the common ap, I apply to them for free. If not, I fill out an application, and am still willing to fill out 3 more applications. Does this lead to a similar increase in applications for both common ap and non common ap schools?

  19. haxen says:

    I feel like the number of international students is probably higher than shown here. Would like to see whether those Harvard Gazette numbers match other tertiary institutions.

    Here in Australia, tertiary study for overseas students is a big business, with total international student numbers over 50% in some cases. From a purely numbers perspective, the change in opportunity for East and South Asian nationals over the last 30-40 years means that the candidate base changes from some fraction of 25M Australians to one with some similar fraction of 2B Indian and Chinese students (these two countries, especially the latter, make up the majority of overseas students). This is not even considering that 28% of those 25M Australians were overseas born, and a reasonable proportion of those used university as a migration strategy. Australian universities have a ‘decent’ second-tier reputation (though on average relatively woeful teaching compared to top-tier universities in US/UK – I think they get good outcomes by selecting capable students and forcing them to figure out how to learn themselves) and this is enough for a very high demand. I would expect a higher demand for second-tier US universities based on its reputation, and more motivation for those institutions to increase supply vs say Harvard or any top-tier university.

    These changes aren’t necessarily a negative, but it does show there are significant downsides to moving away from a community-ties-focused approach to one with global competitiveness. Perhaps some of the hyper-competition in education that has been around for at least a generation in Korea and India is necessarily migrating to the west as a result.

    To generalise a bit, the competition could be the outcome of a more efficient and more global meritocracy, and for me makes a strong case against pure meritocracy (in the same way one might argue a pure free market has undesirable side-effects that it can’t fix itself) and at least some recognition that focusing on competitive meritocracy at the expense of community ties has bigger problems than it would seem at first.

    • Steve Sailer says:

      Right, the rise of Tiger Mothers, both domestic and international, has had a huge impact on the culture of American college admissions.

      • haxen says:

        Assuming Tiger Mothers just means more competitive students from one cohort, then yes, I suppose that that leads other cohorts to be competitive to match.
        Though to me it’s mostly a numbers game, coupled with lack of equivalent local university/opportunity supply and relatively open immigration to the west. If there were (US Pop)x100 martians, then the top 1% of martians would have the opportunity to get almost all places at US universities if they all chose to, unless there were restrictions on martian intake or the martians built/found better facilities elsewhere.
        So then I’d argue that the competitiveness is then just a result of necessity – limited opportunities for a large/growing number of people, most located outside their country. Better local facilities and opportunities in large ‘developing’ countries would benefit everyone.
        English as the lingua franca also has some confounding effects here I guess.

        • Steve Sailer says:

          To be born an Early Baby Boomer in the USA was to be born a Prince Among Men. For example, 3 Presidents of the United States were born in the year 1946, a year in which the cohort born that year went through life with relatively few people 15 or fewer years older, but a huge number or people 17 or fewer years younger.

  20. Worley says:

    I read some academic study which noted in passing that over the past decades, students have become much more willing to travel distances to their colleges. One consequence is that college student bodies have become much narrower in terms of academic ability, and much broader in the geographic spread of their students. One consequence of this will be that the best schools will contain a lot fewer mediocre students even if the distribution of students overall remains the same.

    • Douglas Knight says:

      I read a similar paper. Was it this? Another paper by one of the authors?

    • Steve Sailer says:

      When my wife’s uncle from the West Side of Chicago won a scholarship to MIT around 1950, the local newspaper ran a picture of him getting on the train to go off to college. Apparently, Local Boy Leaves Home for College was considered news at the time.

      “The Bell Curve” has a lot on changes in the rate of kids going far away from home for college, which was rare before Mid-Century. Perhaps WWII was a “nationalizing” phenomenon. GIs would have old war buddies from other parts of the country, so the idea of applying to college in far-off New England would seem less outlandish to guys who’d taken government paid trips to, say, Okinawa.

    • Steve Sailer says:

      The Mormon flagship college BYU is one of the increasingly rare schools with a student body with a wide dispersion of test scores. Others in the same camp include military schools like The Citadel and VMU.

  21. jeangoodwin says:

    Labaree’s amusing synthesis of the literature, _A Perfect Mess: The Unlikely Ascendency of American Higher Education_ (2017) has a somewhat parallel analysis, esp. in chap 3.

    On the side of the colleges, there is a constant incentive to become more “prestigious,” achieved by rejecting more applicants, taking on the patina of “liberal” as opposed to “professional” education, adding “research,” etc.

    On the side of the applicants, he thinks that by around 1910, college became the central method for passing middle-class status to on to one’s kids. As more people got high school diplomas, college became a necessary distinction. With the increasing scale of businesses, small shopkeepers and tradesmen could no longer expect to hand down their business. White-collar jobs in the larger-scale enterprises became the standard support for the middle class, with college being the ticket of entry. Employers cooperated, and not only because of the signaling function of college; success in college showed that the man could maneuver within hierarchies, be loyal to an organization, work somewhat independently, and get along with colleagues.

    Anyhow, if both producers and consumers of college are heavily invested in being better than everyone else, then it’s no surprise that everyone in the system is going to think that the system is becoming more selective. That is the key good the system produces–distinction.

  22. faoiseam says:

    The College Board has lots of data on AP exams since 1996. They track how many people take exams, in what grade, by race and result, which is enough to get a sense of how people did. They also keep track of AP scholars. In 2002, 1738 people were National AP scholars, (8 APs with score 4 or more) enough to get you into HYPSM. This increased to 7k by 2005, 16k by 2010 and now is 35k, not enough to get you into a top 20 school. By this measure, college entry has got more competitive.

    The number of schools offering APs has increased in that period, from 10k in 2000, to 17k in 2010, to 20k now. Looking at individual subjects, the two most taken are AP Lit, and AP Calc. 5s in lit increased 10 fold from 2000 (6978/105k) to now (61k/580k). For AP Calc BC in 2000 we have 5k 5s from 15k exams, in 2010 39k 5s, from 79k exams, and now we have 56k 5s from 139k exams. As these are the most common exams, this shows that while the number of schools barely changed in the last 10 years, the number of 5s has doubled.

    In general, things have gotten twice as hard since 2010 (as measured by AP exams), and 5 times 3005, and 10 to 20 times since 2000. The AP curriculum has stayed pretty stable, so that does not explain the change. The number of schools has been pretty steady since 2010, though it did double since 2000. This does not explain all of a factor of 2, as the 10k schools that offered APs in 2000 were the better schools anyway.

    This makes me question why the US does not used AP exams for college admission. If, as I estimate, the number of people who get n 5s doubles as n increases by 1, this suggests that four 4s and four 5s should be enough to get into HYPSM. The UCs should requires 8 4s. or 7 4s and 1 5, etc.

    This could be done by computer when AP results come out, delaying admission until the Summer of senior year. It would be really nice to have a clear bar that was required for entry, as opposed to the current opaque system.

    • BBA says:

      It would be un-American. We don’t want Asian-style test prep death marches, which high school would turn into if APs became the sole determinant of college admissions. Nor do we want European-style hard tracking – we’re not going to tell any student who fails to make honors/AP track in every subject that they can’t go to college. We have illusions of equality to uphold!

      And then there’s the other elephant in the room – the reason why college admissions went opaque here to begin with.

      • faoiseam says:

        Right now we do have Asian style test prep death matches, we just have them graded in a non-standard way by local high school teachers, subject to favoritism, as opposed to being blind graded. And we have the nightmare that is extracurriculars, which, as far as I can see, only has the benefit of teaching people how to lie about their achievements at an early age.

        I would suggest that 8 APs is a reasonable maximum to grade people on. College track kids take one AP in sophomore year, and five or more in Junior and Senior year. This has plenty of room to not take all APs.

        This would not apply to all colleges, just the selective ones, and naturally, as colleges got less selective the number of APs would drop.

        I agree that this suggestion would have problems in that it is not gameable enough by people who want to control who gets into college. Its fairness is a weakness in that sense.

        • johan_larson says:

          Another issue is that the really elite schools don’t want students who are only good at academics. They are looking to enroll the future leaders of society, not just a bunch of grinds and geeks. It makes sense that they look for signs of leadership, and that’s not going to be evident purely in grades.

          But in most circumstances, your proposal of using AP courses sort of like the Brits use A-level exams has a lot of merit. In particular, the grading is out of the hands of the teachers, so there is less opportunity for favoritism. AP courses are also more demanding than what most high schools teach in senior year, so making AP courses the de facto senior-year courses would raise expectations on average. The one issue I see is opportunity; not all schools offer AP courses and many offer only a few.

          • faoiseam says:

            not all schools offer AP courses and many offer only a few.

            20,000 schools offer APs. Eight states (Arkansas, Iowa, Louisiana, Mississippi, West Virginia etc.) have laws making it mandatory to offer at least 4 APs. 14 others make offering APs or dual enrollment mandatory. Most other states are considerably better than those states. Random schools nowadays offer lots of APs.

            They are looking to enroll the future leaders of society, not just a bunch of grinds and geeks.

            Oxford and Cambridge do ok just on grades. I don’t believe there is a way to determine that a child at age 17 is going to be a future leader that does not rely on either grades or money/social status. I am fairly sure that all the holistic admissions an non-probative on this count. Yes, if you are a Duke, (or the US equivalent) or the scion of a billionaire, you have good chances, but no essay or extracurriculars can predict future leadership. (I would guess)

          • AlphaGamma says:

            Oxford and Cambridge do ok just on grades

            Plus interview, which is more important- it’s certainly not uncommon for one candidate to be accepted after interview while another from a similar school with the same or better grades is rejected.

        • Peffern says:


          As I commented earlier, high school is a death march, it’s just a more opaque death march that doesn’t produce productive students at the end of it, and is much more easy to cheat.

    • Steve Sailer says:

      I favor colleges weighting Advanced Placement exam results more heavily in their admissions decisions on the grounds that since test prep appears inevitable, we might as well be encouraging students to test prep in real subjects like chemistry and American history, in which they might actually learn something from all their test prepping.

    • OriginalSeeing says:

      When I had exposure to AP exams back in 2005ish, the exams were pretty wonky. Things may be different now, but back then each test had drastically different difficulty levels. There was one AP exam (physics?) that only one or two students were able to get a 3 on while tons of us got 4s and 5s on calculus and literature. (We may also have had a trash physics teacher.)

      Also, the high school I was at was intentionally pushing all students to take tons of AP tests just so that the high school could point at the number of successful AP test takers they had.

      Increasing the number of standardized tests up and up to the point where each student has to take 10+ exams (SAT and probably at least 9 AP exams to get that precious 5) sounds horrible.

      • faoiseam says:

        Increasing the number of standardized tests up and up to the point where each student has to take 10+ exams (SAT and probably at least 9 AP exams to get that precious 5) sounds horrible.

        I would have expected all schools to have finals in all their subjects. The AP exam replaces the final that the student would have taken anyway.

        Some AP classes are harder than others, for some students. The rates of scores at various levels are published. The exams are meant to line up with the corresponding class in a college, and they do track that quite closely.

        In my idea world, admission to college would be based solely on the results of your AP exams, so you would find out that college you got into when the AP results came out at the end of your Senior year. This would remove all the bullship extracurriculars, all the constant pressure of continuous assessment, all the bias involved in written recommendations, and the silliness of admitting people based on the SAT, an exam based on material the children learned in Middle school (25% of Harvard has completed multivariate calc/linear algebra or beyond).

        Basing admission on a (or two, including Junior year) one week exam would free up children to actually have childhoods, rather than need to obsessively worry about how things will look on a college application.

        • Steve Sailer says:

          Right. The Advanced Placement exams are close to being the ideal for college admissions, they just need fine-tuning to make them more feasible for college admissions.

          I must point out, however, that Charles Murray’s view is that the SAT Subject tests are superior to the SAT for the purposes of College Admissions, but he won’t go to my position the the AP tests are potentially superior to the SAT Subject tests.

          Of course, somebody will probably get angry that I cited Charles Murray as disagreeing with me and complain to Scott that I’m engaging in Culture Warism by pointing out Murray’s dissidence from my view.

  23. meh says:

    The pew article you link has a different conclusion about the common ap

    The expansion of the Common Application, which makes it easier for students to apply to multiple schools, doesn’t appear to be behind the increase in application volume. The Common App, as it’s called, is accepted by nearly 800 colleges and universities in the United States and several dozen overseas. Of the 1,364 institutions in our sample, 729 accept the Common App along with (or in some cases instead of) their own application forms; the other 635 use their own forms. Although one might suspect that the ease of applying to multiple schools via the Common App would result in stronger growth in application volume among those schools, there was almost no difference in 2002-2017 growth rates between the schools that used the Common App and those that didn’t.

    • Ghillie Dhu says:

      Confounder: what drives a school to start accepting the Common App? Perhaps it does increase application volume, and is only adopted by schools with otherwise-falling inflow.

      • BBA says:

        Here’s a wild guess – the widespread adoption of the Common App coincides with applications shifting from mostly on paper to mostly online. Rather than hire software people to build custom admissions software, many colleges went for the cheaper option of using the preexisting Common App.

  24. marc200 says:

    First, I think I have your answer on the University of California system and why competition has gotten so much harder there. International admits have increased from 3% in 2009 to 18% in 2018. That’s a big increase! International enrollments aren’t up quite as much, but still increased by a lot, from 2.2% in 2009 to 13.4% in 2018. It looks like a chunk of that took place at the expense of white students, whose share of admits dropped from 34% to 21% over the same period. Data is here:

    Also like to take this opportunity to renew my plea from the book review thread for Scott to review “Medical Nihilism” by the philosopher Jacob Stegenga. It’s a very rich book with many SSC-relevant and medical practice relevant themes, and I’d be interested to see Scott’s view. Here is a review of the book by Richard Smith, the former editor of the British Medical Journal:

    Smith says that in his experience the best doctors tend to be instinctive “medical nihilists” in Stegenga’s sense.

    Amazon link to the book:

    (No, I have no connection whatsoever to the author, just liked the book and love Scott’s reviews, particularly on medical and philosophy of science topics). Thanks!

    • Steve Sailer says:

      California got over its budget crisis in part by admitting a lot of foreign students to pay full fare at University of California campuses.

      One of the repeated lessons of my lifetime has been that the old 1960s idea that Americans were wealthier than the rest of the world is less and less true. For example, when I applied to college in 1975, a large majority of the East Asians in the world were slopping pigs in Red China or being beaten for wearing glasses in Cambodia or the like. But today, the number of East Asians who can afford to pay list price at famous American colleges is likely greater than the number of Americans who can afford that.

  25. secondcityscientist says:

    One thing I’ve been thinking about recently has been the conditions for kids born in the 70s, after the peak of the baby boom (graph here, fig 2, some further back figures here that may be less reliable). In the 70s, around 20% fewer babies were being born compared to the 50s, around 3.2M births/year compared to a peak around 4.2M/year. So all of the infrastructure that kids need, especially schools, would be over-provided for those kids. Does that mean that in the 90s when these kids applied to college, those schools had to be less selective? Most 18 year old kids now would be the children of 70s kids – is that the comparison that is being made?

    I notice that most of Scott’s data doesn’t include the 90s. That’s the time I’d expect to have been “easier” to get into a top university, not necessarily the early 00s. Total US births appeared to have a local peak in 1990, so I would expect that 2008 would be peak difficulty-of-college-admission but there wasn’t much decline after that peak, so it’s probably been relatively constant since then.

  26. Reasoner says:

    increased common knowledge of intense competition making everyone compete more intensely

    I think this phenomenon is underrated in general. Behavior lags incentives, and people don’t instantly converge on the Nash equilibrium. This is part of why trend extrapolation works: if there’s a trend in some direction, it’s evidence that the Nash equilibrium lies in that direction, and the trend will continue until the Nash equilibrium is reached. This makes me very concerned about increasing US political polarization.

    Dale and Krueger(2011) examine this question, using lifetime earnings as a dependent variable.

    Some criticism. (I’m confused too. I think maybe it’s one of those multicollinearity in regression analysis things, where there are 2 closely related variables where one happens to get a much stronger weight due to quirks in the data, and reporters covered 1 of the 2 closely related variables while ignoring the other one?)

    This suggests that successful tech entrepreneurs come from the population of people smart enough to get into a good college, but don’t necessarily benefit from the college itself.

    I think I saw a study contradicting this conclusion, but I’m too lazy to dig it up.

    • Scott Alexander says:

      The study I’m citing is from 2011, so it can’t be the one Robin was criticizing in 2009 (I would look directly, but his link is broken)

  27. Jacob says:

    FYI the American Institute of Physics collects of bunch of education-related data you may be interested in. Generally covers STEM fields, at both undergrad and grad levels.

  28. albertborrow says:

    I went through the modern college admissions process last year. Here’s what I experienced:

    Every single adviser, teacher, and parent I talked to said that I needed some kind of extracurricular activity, like community service or club work, in order to have any chance of getting into a good college. I was encouraged to stay in Boy Scouts because it meant a better chance at admissions. I stopped attending because I found it boring. I was encouraged to break my back to participate in the National Honor Society, because my grades were barely high enough to qualify. I (wisely) didn’t participate because I had no chance of getting the scholarship. I was encouraged to join Link Crew, to tutor other students, to do literally anything, because it would set me apart from the other names and SAT scores sent in.

    If I had followed their advice, I would have snapped. I couldn’t do it, so I didn’t. I nearly did snap taking all of the AP classes I did – I entered college with enough credits to start a year and a half ahead. Even with that kind of advantage, I still have to attend eight semesters worth of classes, because the school structured the prerequisites for each class in such a way that I literally couldn’t take all of my required classes in two or three. The only way I can leverage my AP credits to my advantage without paying extra for summer courses is to enter the BS/MS program.

    Whatever the reason for college being the way it is, it needs to stop. I wanted a college – a structured learning environment where people would finally fucking teach at my speed for once – and I got an overpriced high school. I wanted a place where I could focus on the kind of knowledge relevant to my future career, and I got a place where I have to take two semesters of physics in order to get a software engineering degree. Just in case I happen to be one of the three future software engineers working on the LHC. (And every time I suggest that the philosophy majors should have to take high level courses, they look at me funny. Philosophy majors would probably get more relevant material out of it than I do!) It’s so enormously frustrating. I pushed myself through high school telling myself that there was only a couple more years before I would finally start learning what I needed to in order to do what I want, and it turns out that was just a lie.

    I live in a world where I would have had a better quality of life overall if I had dropped out of high school and went to a coding boot-camp at the age of sixteen, instead of studying for my SAT. Don’t go to college, kids.

    • A Definite Beta Guy says:

      My sister-in-law went all-in to become valedictorian and loaded up with a bunch of extracurriculars to get into an Ivy, and she absolutely did snap. Then she became a hipster. She had a pretty horrible relationship with her family until pretty recently.

      Looking at the rest of my in-laws, yeah, they pretty much all snapped pretty right and proper. Tiger Mom….Tiger Mom…Tiger Mom…wait, you’re downing 8 cups of coffee a day and crying on the floor every night? You just quit your job and wandered off to Colorado? You retreated into manga and anime and failed out of school?

      How could this happen?!

      This is something I am really, really, really hoping to avoid in any future family of mine.

      • Tarpitz says:

        Is it possible that your in-laws simply have snap-prone genes? My schizophrenic mother, snap-prone self and full siblings, and seemingly mentally hale half-siblings say hi…

    • DinoNerd says:

      Yowch! You have my sympathy.

      In the long run, some of those weird college requirements may prove useful; I know my classmates were concerned about “relevance” (e.g. computer languages actually used in real jobs), but when I got out I found I had a huge advantage over people who just knew how to code – I knew the theory, and could pick up a new language in a day or two – they had much more trouble just “picking up” the theory. But I graduated in 1978, so your mileage will vary.

      OTOH, I also liked college a lot. I was finally out of most of the things I had hated about high school. And in the mid 70s, where I was, there were lots of options for doing what I wanted – which for me was to take courses on everything rather than to focus on future career success.

  29. ec429 says:

    I previously investigated one facet of this – that necessities are getting more expensive – and found it to be true.

    “Necessities” seems an odd choice of word here. Looking at the second graph from the cost disease post, for instance, food prices have gone down in real terms, while shelter isn’t that far above the CPI. (Yes, the housing market is broken, and we can talk about why that is, but it’s not “dectupled prices” broken.) And the only way that the medical line counts as ‘necessities’ is by conflating “healthcare” with “healthcare that satisfies an ever-more rapacious bureaucracy and is supplied by members of a guild with a legal monopoly”. Now, sure, the presence of a legal mandate to buy “healthcare” makes it in some sense “necessary”, but I don’t think that’s what is usually meant by “necessities”. As for a university education being “necessary”, I think you yourself have argued the case against quite effectively (and not only is it not necessary, it’s not sufficient either). You even link to those in this post!

    My parents’ generation tells stories of slacking off in high school, not worrying about it too much, and knowing they’d get into a good college anyway. Millennials tell stories of an awful dog-eat-dog world where you can have perfect grades and SAT scores and hundreds of hours of extracurriculars and still get rejected from everywhere you dreamed of.

    I wonder if that’s because the selection function determining what kind of people you hear stories from differs by generation. It certainly doesn’t match the stories I hear (or the one I lived — b. 1991, which apparently makes me a millennial. Although some parts of the post suggest you’re actually talking about Gen Z; uni applicants haven’t been millennials for about five years now). Or maybe the US is just uniquely bad at this?

    Anyway, my pet theory (for what it’s worth) is that the strategies for gaming the easy-to-game metrics have gone from the secret weapon of the ‘elite’ (scare quotes because I’m using the term very loosely. But cf “the sort of upper-class white people who write news articles about the state of college admissions”) to common knowledge among anyone with the kind of educational and parental background to be even thinking of top-tier colleges; consequently, the ‘elite’ now find it much harder to get their kids into Hahvard, and write news articles about it. And maybe the reason things didn’t look the same in the UK was because Cambridge don’t care how many Strivers Clubs you presided over vice in (they just want to see your STEP scores) so the ‘elite’ never had an edge to lose. Alternatively, they had an even better strategy: convince everyone else that the top universities cared about extra-curricular crap, so all the upwardly mobile challengers would waste their effort on ineffective Striving. (Brits have been playing the game of class-system for a lot longer, so one-level deceptions just don’t cut it 😉

    This does also lead me to wonder, why didn’t grade inflation just lead to Harvard coming up with some equivalent of STEP? If it is just down to “meritocracy is passé”, then that supports the “US is uniquely bad” position; ‘elites’ bought themselves a few years of getting their dumb kids into Harvard (exaggeration) at the cost of now everyone has to club strivers just to get back to where they were. (“It takes all the running you can do, just to stay in the same place”, as the Red Queen would say.) Why the US fell into that particular Molochian trap when it did, I don’t know.

    • Steve Sailer says:

      My general assumption is that Harvard is pretty good at picking out which applicants are most likely to make big donations to Harvard in a half century and/or surface in the national news within 20 or 30 years. For example, who is in the news lately? How about Pete Buttigieg? I’ll bet he went to Harvard. Okay, I will go look him up on Wikipedia. …

      Yup, Pete Buttigieg went to Harvard.

      • habu71 says:

        How do you think they predict this? What about Pete Buttigeg screamed “This guy is going to be famous”?

        • rlms says:

          If you’re willing to have a high false positive rate, a high true positive rate isn’t that difficult.

        • Edward Scizorhands says:

          Hustle predicts a lot. When I see college mates in the newspaper or magazines, I remember them as people who kept on going that extra mile. (Lots of extra-milers just ended up with very nice but not nationally-famous careers, so, like rlms says, accept some false positives.)

        • habu71 says:

          I would argue that, by most definitions, it is indeed possible to have both a high true positive and a high false positive rate. One can get many true positive results by accepting many students, but the rates will be inversely proportionate to each other.
          I agree with Steve Sailer that it does indeed appear that harvard has a higher than average rate of selecting people who will be in the limelight in future years, but I am greatly confused on exactly what piece(s) of information they are using to predict such a thing. All of the indicators I would think to use to predict such things would be hit very, very hard at the level of student Harvard selects from and thus would likely not be very useful.
          Other qualities, such as “hustle”, might be predictive of future success and fame, but seem very difficult to reliably measure and quantify for an entire applicant class. Intelligence and wealth are much easier to measure, but their predictive value likely suffers greatly from restriction of range. To use Steve’s example, what qualities let Harvard predict Buttigeg just from his application?

          • faoiseam says:

            To use Steve’s example, what qualities let Harvard predict Buttigeg just from his application?

            He had he usual qualifications, valedictorian etc., and he also won the JFK Profiles in Courage essay competition, which is awarded in Boston, where he met the Kennedys. This seems like the thing that could put you over the edge, and signal that you are hoping to be a leader of men. Of previous winners. Daud Shad is at Yale, Ben Wolman is at Columbia, Margo Balboni at UNC Chapel Hill, Maia Gottlieb was Penn then Harvard, Ben Loffredo was Yale, Will Schmidley was Minnesota, Stephanie Dziczek was Yale. I can only check the people with weird names.

            Too many Yalies for it to be normal.

          • Steve Sailer says:

            There is a book, “Choosing Elites,” by the social scientist Robert Klitgaard of the RAND Institute and one-time president of Claremont McKenna college on the lessons Harvard learned in the 1970s while he was working on Harvard admissions staff as a data modeler:


            It has been a long time since I read it so I don’t remember the details. But one insight I took away from it is that Harvard is Very Serious about choosing the best applicants (“best” being defined as being in the best interests of Harvard) and that it expends a lot of brainpower on studying how to do it right.

          • Steve Sailer says:

            Here’s Buttigeg’s Wikipedia entry:

            “Buttigieg was born in South Bend, Indiana, to Jennifer Anne (née Montgomery) and Joseph Buttigieg. His father, an immigrant from Ħamrun, Malta, initially studied to be a Jesuit priest before immigrating to the United States and embarking on a secular career as professor of literature at The University of Notre Dame at South Bend.[12][13] The surname Buttigieg is Semitic; “tiġieġ” meaning “poultry” in Maltese.[14] His mother was a professor at Notre Dame for 29 years.[15]

            “In 2000 Buttigieg was valedictorian of his high school senior class at St. Joseph High School in South Bend.[16] That year he was the recipient of a first prize for the JFK Profiles in Courage Essay Contest awarded by the John F. Kennedy Library in Boston. He traveled to Boston to accept the award and met Caroline Kennedy and other members of President Kennedy’s family. Buttigieg had written about the integrity and political courage demonstrated by U.S. Congressman Bernie Sanders of Vermont, one of only two independent members of Congress.[17][18]

            “Buttigieg attended Harvard College, majoring in history and literature.[19] While at Harvard he was president of the Harvard Institute of Politics Student Advisory Committee and worked on the Institute’s annual study of youth attitudes on politics.[20][21] He wrote his undergraduate thesis on the influence of puritanism on U.S. foreign policy as reflected in Graham Greene’s novel The Quiet American.[22]

            “Upon graduating from Harvard in 2005, Buttigieg was awarded a Rhodes Scholarship and in 2007 he received first-class honors in philosophy, politics and economics from Pembroke College, Oxford (later promoted to a Master of Arts (Oxon) per tradition).”

            That’s pretty impressive.

            Personally, however, I don’t think that’s as impressive of an artisanally-crafted resume as that of Rep. Seth Moulton (D-MA), who is expected to announce his candidacy for President shortly. Moulton is a fellow Harvard grad (and a physics major, which is more intimidating than Butiegeg’s humanities major). who served not one but four tours in combat zones and was in the Marines rather than the Naval Reserve.

            But Moulton has a wife and just became a father last year, so he doesn’t seem terribly Current Year.

      • meh says:

        My general assumption is that Harvard is pretty good at picking out which applicants are most likely to make big donations to Harvard in a half century

        Maybe picking them out is rather easy, they just mostly want to go to Harvard. Which schools were Pete Buttigieg rejected from?

  30. jonabar says:

    1. Even they don’t show up that much in enrollment figures, I think applications by international students are rising. Enrollments have almost tripled in 25 years, and that doesn’t cover how many students apply but don’t get accepted or ultimately don’t attend due to cost reasons. There is a huge pool of potential applicants abroad that is not yet plugged into the efficient sorting mechanism you describe, and at the same time studying abroad is seen as a big deal and often (in my personal experience) parents will urge unqualified kids to apply to name-brand schools because they don’t understand how the system works and don’t realize their child has no chance of getting in.

    1a. The effect of international applications should have first impacted grad schools, med schools, etc. (though not law schools, because those degrees don’t transfer as well), and is now filtering into undergrad colleges.

    2. I think you’re right that it’s mostly the quality of the top-tier UC schools that is impacting the UC system (also, if international applicants are having a big impact, these top UC schools are better known to, say, Chinese parents than say, WashU or Williams or whatever). For the lower-tier schools, the problem might also be that UC has a separate application system that applies to all schools, so that if you’re gunning for UCLA or Berkeley from out-of-state you might be tempted to tack on a few extra UC schools just in case, since you’re already filling out their complicated application.

  31. dark orchid says:

    College attendance in the UK supposedly remained very low until a 1992 act designed to encourage it, but it looks like part of that is just them reclassifying some other schools as colleges. I don’t know how it really compared to the US and I welcome information from British readers who know more than I do about this.

    Happy to oblige! The first bit is as you guessed – before 1992 we had universities and polytechnics. The government wanted more people to go to university, so they declared polytechnics to be universities in the Further and Higher Education Act (1992), coining the phrase post-1992 university (a.k.a. New University).

    Since the introduction of tuition fees and basically converting universities to businesses over the last decade, enrollment figures have gone up (my own institution more than doubled its student intake) but most of that is what universities politely refer to as “high margin students” which means that although you can only charge a home or EU student £9250 a year, you can charge an overseas student £20k+ a year, which is where we make the money that keeps us afloat these days.

  32. Lasagna says:

    This was a great article as usual (thanks Scott!). That said, I really, really wish this were true, but I’m not sure it is:

    6. All the hand-wringing about getting into good colleges is probably a waste of time, unless you are from a disadvantaged background. For most people, admission to a more selective college does not translate into a more lucrative career or a higher chance of admission to postgraduate education. There may be isolated exceptions at the very top, like for Supreme Court justices.

    I graduated from Columbia Law School in 2001 (the same year Scott was a junior in high school. Get off my lawn). I’ve been eating out on that degree ever since, and I’m not a Supreme Court justice.

    I know plenty of successful, prosperous attorneys that didn’t graduate from a top school. But I also know lots of desperate, broke people with useless JDs that went to “lesser” schools. I know no unsuccessful attorneys from top schools.

    This plays out in a lot of different ways. For example: the recession devastated the legal industry in New York. Just reduced it to rubble. People were losing their jobs left and right. Not everyone who survived in the NY market during that period was from a top school, but pretty much everyone from a top school survived – we took different jobs, we switched careers, whatever. My colleagues who had to leave the area to find work and cheaper places to live were all from good, but not sexy, law schools.

    Now that I have kids of my own, I’m desperate not to engage in this sort of “Ivy league or death!” stuff. I really would love to consider not pushing college at all, but at LEAST I’d like them to take on as little debt as possible – go to a local school, live at home, that sort of thing. But I’m not sure I see it working out like you describe.

    • Scott Alexander says:

      I think law school matters more than undergrad.

      • Lasagna says:

        I was just about to edit my comment along those lines. Sorry – I didn’t catch that you weren’t including law schools in this analysis.

    • BBA says:

      I graduated from a Top 14 law school in 2010 with lousy grades – coming from a math/comp sci background, it took me way too long to figure out how to take law school exams. I just answered the questions, mistake #1.

      I interviewed with a couple dozen law firms in NYC and southern California, was passed over by every single one of them, ended up taking a non-legal job doing stuff I was qualified for before I entered law school, and never looked back. In retrospect I obviously would’ve hated working as a lawyer…but I was young and naive in 2007 and had dollar signs in my eyes.

      What I’m wondering is, why did no firm hire me? Was it me, or was it the recession? Or a bit of both?

      • Protagoras says:

        If you graduated from a top 3 law school, you wouldn’t be calling it a top 14 law school, you’d be calling it a top 3 law school (or maybe you’d say top 5 if it was number 3 and you didn’t want everybody guessing it was number 3). As a result, one guesses the specific rank of this top 14 school was in the double digits. From what I’ve heard, the job market for lawyers in recent years has been sufficiently awful due to the excess supply that it’s really only the top 3 law schools that still get you high odds of a good job.

      • broblawsky says:

        Did you do an internship? Most prospective lawyers intern at the firm they intend to eventually work during their summers at law school. Even if you don’t work at that firm, it gives you some references.

        • BBA says:

          Oh, most of those “job interviews” I failed were internship interviews. I ended up doing an unpaid externship at the Patent Office, which left me even more jaded about the state of the patent system.

      • Clutzy says:

        I graduated from a law school a tier below that in 2013. I think there are some reasons you were possibly passed over.

        1. Bad grades are a real thing. The firms that typically hire from these elite colleges also sometimes are run by grade snobs. This is why Yale stopped with grades, IIRC.
        1b. Did you interview with any niche firms? Those probably would have taken you regardless?

        2. Were you applying for patent prosecution positions? That is obviously what you should have been doing from my POV.

        3. 2007 = Entering a depression in the US, the legal market was one of the worst hit ones.

        4. There is a guy who I know who graduated top of our class (like top 5ish), huge GPA in his undergrad and Ph.D programs, impressive research, impressive law review. Couldn’t secure a single job because he gives off the worst vibe ever in person. He is an adjunct at a minor state school, not at the law school now. Its the only job that will accept him. He’s so creepy (and apparently this is a significant problem at our school) that the school created a program to try and teach our creepy students social skills, because Jones Day and some other elite firms stopped coming to on campus interviews because so many of our top students were creepy.

        • BBA says:

          I think it’s all of the above. I was applying to IP specialist firms, and had already passed the patent bar at that point – still no luck. And I’m lousy at interviews, and was getting depressed about my future towards the end of the interview process, and it showed. (I’m still depressed, but I’ve learned not to show it too much, except in whiny internet comments.)

      • aristides says:

        I was in roughly the same boat as you. I should have taken it as a warning sign when I finished a test with an hour and a half to spare. My guess is it was either the recession or internships, and probably both. The only places that would look at me post graduation were extremely similar to my internships. I heard this was a change during the recession years that is here to stay. Personally I consider it a blessing in disguise. I only work 40 hours a week and my salary is more than half of the law firm salary that I hear takes 80 to 100 hours to earn. It really might be a case that the law firm job market does a good job weeding out the people that would hate being a lawyer. I just wish I didn’t have the debt.

      • Lasagna says:

        2010 was a terrible year to graduate. During the previous three years the big firms – the ones that survived, several went under – cut their entering class in half, or by 90%, or didn’t hire anyone new at all, or didn’t hire anyone new AND postponed the start date of grads they had hired the year before. There has been a recovery since, obviously, but not as much as you might think. But things hadn’t improved much in 2010.

        If you went to one of the very very top schools, you might have still been OK even with bad grades. But not all of the schools on that Wikipedia list Edward Scizorhands posted would qualify. They’re all real good schools – I’m not saying that if you weren’t on law review at Cornell you were doomed – but grades would have counted for a lot that year.

        And the biggest problem, of course, is that if you don’t get that big firm job right out of law school, you don’t get another shot the year after. And smaller firms don’t hire baby lawyers. It’s a lousy industry.

    • Edward Scizorhands says:

      I’ve been eating out on that degree ever since,

      I have a degree from an elite institution, and it’s amazing how many doors it opens, and how many people think I can shit gold because of it. Multiple bosses have introduced me as “the guy from [college] .”

      (It’s kind of embarrassing. I know that sounds coy, given how many other people have Real Fucking Problems. But it is. Often I feel it’s just because I had my stuff together from age 14-22.)

    • Clutzy says:

      Just a bit of a tip, there is a ton of full ride scholarship opportunity out there at various schools based on academics, particularly from undergrad (and its actually quite easy to go from state school undergrad to Stanford/Ivy/etc grad school if you are a good student). Both me and my sister got full rides to undergrad at good state institutions that were trying to climb the rankings. Both our degrees have appreciated over time (hers much more because Nick Saban has revolutionized the bama degree) and she just easily got admitted to a Ph.D program after a few years of working at a top 20 uni for her field, and top 25 generally.

      Another idea is just to go in-state (if your state has mildly reputable institutions) with the same thing in mind. Most states have a good state school that a smart kid can jump off from even if his HS career wasn’t insane. Guy from HS I competed with in wrestling, went from our state school to Princeton medical school, for example.

      • mismath says:

        As a digression, shout out to Nick Saban. Dude almost certainly helped make applying to math grad school for me surprisingly fruitful. I had three friends the year under me who ended up at Princeton for ChemE PhDs after graduation. Talk about halo effects.

      • DragonMilk says:

        Princeton doesn’t have a medical school, or any professional school for that matter.

      • eccdogg says:


        Good grades and good test scores from a state flagship will get you into prestigious grad schools. And a prestigious grad school is way more important than undergrad.

  33. Douglas Knight says:

    You talk about how ease of application can create the illusion of competitiveness, but you fail to mention that colleges want that illusion. First, colleges became competitive. Then US News (among others) ranked colleges, drawing attention to their admissions rate. Then colleges gamed the rankings. The common application didn’t just happen. It isn’t just misleading. It’s engineered to be misleading. (as DV said above)

    So I think that you trust comeptitiveness as measured by admissions rate way too much.


    Also, they probably like having a high reject rate to allow opaque standards. This is related to what John Schilling says above, but he doesn’t address why schools would like it. One reason is that it masks legacies and outright purchase of admission. Steve Sailer suggests elsewhere that admission is random to make students grateful rather than entitled.

    • Steve Sailer says:

      “Steve Sailer suggests elsewhere that admission is random to make students grateful rather than entitled.”

      That’s more cogent than anything I can recall saying on the subject, but I like that! I’ll be happy to take credit for it, even if most of it owes to Douglas Knight’s impressive powers of insight.

      It’s kind of like the dog-training idea that if you want your puppy to grow up to be irrationally devoted to you, you treat him in a random fashion, sometimes loving, sometimes punitive.

      This traces back to a friend who was a professor at a Top 50 private college saying to me about 15 years ago that he was told while serving on the Admissions committee that the idea that legacies get an advantage in admissions is a hoax perpetrated to encourage alumni to donate.

      My impression, though, from reading accounts of gigantic 9 figure donations to colleges is that they tend to come from an:

      – Alumnus
      – A legacy (ideally with both a parent and a child who attended the college)
      – An athlete
      – A male
      – A white
      – A fraternity member
      – A Republican

      In other words, from Haven Monahan.

      This helps explain some of the tensions on campus these days: the boogie-men also tend to be the biggest donors.

      • Douglas Knight says:

        There is analogy between fraternities and colleges. People are grateful to both for introducing them to a network and in return propagate the network, which is probably more important than donations. One might call this asabiyyah.

        We have one image of fraternities at most schools, all male 3 year residences, but another image at elite schools, namely, Skull & Bones, which is non-residential, doesn’t throw parties, and only lasts one year. (How do they maintain continuity when the only contact between cohorts is rush?!)

        Harvard and Yale have been pressuring these clubs to go coed for decades. Skull & Bones went coed 20 years ago, while Harvard has upped the pressure recently. I suspect that they just want to crush a competing center of power. Harvard wants people in or out, not internal competition.

  34. Douglas Knight says:

    The framing in terms of the last ten years is pretty weird.
    If you’re rebutting some meme, OK, but who cares?
    But the article isn’t really about that, so I don’t have a substantive complaint.

    • Douglas Knight says:

      To combine my two comments, I’m skeptical that anything interesting happened in the past 10 years at all. Maybe California is special, though.

    • Steve Sailer says:

      Right. I’m 60, so I compare 2019 to 1976, not 2009. A lot has changed since 1976.

      On the other hand, I can imagine that one big recent change was that post-2008 Crash, American colleges have emphasized letting in more rich Chinese foreign students to pay full list price tuition, even if, say, they can’t speak or even read English.

  35. PersonOfInterest says:

    In many fields, a prestigious graduate school is almost an absolute requirement for becoming a professor. But there doesn’t seem to be an undergrad equivalent of this.

    I’m wondering the rapid growth in PhDs has affected the competitiveness of individual colleges. In other words, does the glut of PhDs on the market have a trickle-down effect on college reputation?

    A tenure-track position at any university will have very tough competition. This suggests that the average quality of the faculty in a given institution would have increased over time. Having better faculty would produce better research, acquire more institutional funding, etc. and eventually attract higher quality undergraduate students.

    Anecdotally, at my upper-middle-tier private university, my professors disproportionately had PhDs from top-tier universities. Also, I know people with Ivy League PhDs, prestigious post-docs, and multiple papers in Nature and Cell who are struggling to find teaching faculty jobs. Junior academics can’t afford to be picky and tend to take any job they can get.

    This could help explain the growing prestige of UC schools and why UC Davis, UCSB, UC Santa Cruz, etc. are all hypercompetitive. This is happening in other states, too. According to my father, the local state school where I grew up used to be considered “high school with ashtrays” but is now fairly respectable. I’m not sure if there’s any way to measure this, but it seems to make sense.

    • The Nybbler says:

      A tenure-track position at any university will have very tough competition. This suggests that the average quality of the faculty in a given institution would have increased over time.

      It suggests they’ve gotten better at getting tenure, not necessarily in any other way.

  36. rlms says:

    It’s apparently still technically possible in four states (including California) to become a lawyer by reading law, but it’s rare and not very encouraged.

    Kim Kardashian is currently doing this.

    • ana53294 says:

      Considering the cultural influence she has, I wonder whether that will mean more people will try this option.

      Also, how hard is it to find a lawyer who will agree to hire an apprentice when you are not a famous millionaire?

  37. caffeinezombie says:

    I’d look at the cold war and 20th century educated immigration waves as possible factors making “externally measured academic status” important. The cold war made “hard science” more important. This increased the status of research scientists (who need more schooling than other technical professions) and put science in contact with government bureaucracies, which could not rely on “local” data. And relatedly, waves of educated immigrants looking for academic jobs (first the Nazi refugees like Einstein, then late 20th century immigrants from Eastern Europe and China) made “cut of his jib”-style arguments less valid in attempts to hire objectively better researchers. This is just a STEM story, but it might have spilled over to other departments.

  38. eightieshair says:

    In 2002, I was a senior at a California high school in a good neighborhood. Most of the kids in my class wanted to go to famous Ivy League universities, and considered University of California colleges their “safety schools”.

    Hmm. When I was a senior at a California high school in a good neighborhood in the late 80s, a surprising number of my classmates actually turned down offers from Ivy League schools because the very idea of leaving California was unthinkable for them. Berkley and Stanford were the pinnacle as far as they were concerned.

  39. Peffern says:

    I think point 3.4 is the most important, based on anecdotal evidence.

    I’m in undergrad at a top school right now.

    In high school, I was a good student – perfect SAT, good APs, reasonable GPA, etc. Despite having good accomplishments here it didn’t feel particularly effortful – I’m good at math and can structure a coherent argument, so taking the SAT was mostly just getting a good night’s sleep and studying vocabulary for a week.

    I also did a lot of extracurriculars, and the work and stress load from those absolutely destroyed me. I’m not even talking about “starting homework at 11pm” kind of workload, I mean the social aspect, the cutthroat politics, the status games, and the showmanship. It’s not that I think those things are necessarily bad in and of themselves, but they’re infinitely more difficult than classwork. I’m incredibly busy these days with class and I’m not even on the same order of magnitude of stress as I was in HS.

    High school students are vicious bastards. When you take the AP calculus exam, or the SAT, or even just the final for some class you take, you are only really competing against the teacher, the test, and yourself. When you do extracurriculars, you are competing against horrible entitled jerks with rich parents who make your life miserable. I would take a hundred AP exams before doing another pointless extra curricular.

    I don’t know if it explains the college enrollment statistics but it certainly explains the outrage, pessimism, and anger of people my age over the process. I spent what was for past generations an exciting and important time of my life locked in a box sanding off all aspects of myself that didn’t perfectly resembke an ivy league student just to get beaten out by some kid whose father got him an internship at somewhere prestigious. That does things to people.

    • Randy M says:

      Interesting, thanks.

    • johan_larson says:

      Do you regret doing high school the way to you did?

      Suppose you had done high school in a less intensive way. You took the college-bound stream of classes (presumably doing well in them), and some extra-curricular stuff you were actually interested in, and then whatever placement tests you needed to take, without any extraordinary prep-work. Where would you be studying now? Do you think you’d be happier?

      • Peffern says:

        Interestingly, no. My life now is pretty great, as I’m taking lots of great classes and have a pretty good social group. And high school wasn’t mono misery – I did make a few friends and do a couple fun things. I feel a pretty strong continuity between myself at age 14 and age 18 – it feels to me more like I did my time for 4 years and then got to live my actual life.

        However, for one thing, that seems like an inefficient and harmful way to prepare students for college, and for another thing, I made an explicit promise to myself to keep reminding myself how much high school sucked, so as to not develop rose-tinted glasses with increasing hindsight.

        Given the circumstances, I don’t regret it, but I still feel entitled to bemoan the circumstances themselves.

      • reifel says:

        I sort of took the less intensive option. I also did very well on the SAT and APs, got decent grades, but I had very few extracurriculars and didn’t chase any of the leadership spots. I didn’t get into any highly selective schools and ended up going to the respected but not elite state flagship school near my hometown (admission rate ~50%). In college I basically repeated the process of doing well in academics and poorly in extracurriculars and managed to earn admission to a good (but again not elite) medical school.

        I may be a good exemplification of the “for a given level of ability, elite vs average college doesn’t make much difference” rule. But it did cause me a fair bit of distress as a high school senior to be rejected from almost every school I applied to.

        • eccdogg says:

          This is the route I am encouraging my daughters to take (and the one I took).

          Enjoy your life in HS, do well enough to get into big state U. While there focus on a STEM degree, have fun but get good enough grades to get you into a decent grad school if that is the path you want.

    • Reasoner says:

      Sounds like our future elites are being selected for ASPD :/

      • whereamigoing says:

        Whaddya mean, “future”?

        But seriously, it’s well known that higher income correlates with sociopathy and disagreeableness (the Big Five personality trait). Also, here’s a nice survey by gwern.

  40. Lambert says:

    If pedagogy has improved a lot in the past century (e.g. teaching in the vernacular seems like a decent idea), are we comparing malī to apples when we talk about what it means to go to Harvard?

    i.e. are you getting more ‘Learning’ done per course done now than you used to?

    • whereamigoing says:

      Even if there was no learning at all in the past, given the low current amount, it can’t have improved very much.

  41. honoredb says:

    Probably part of the problem is that not everyone agrees that it’s a problem–we’re incentivizing Academic Excellence and Charitable Activities and Well-Rounded Character Building in high school, leading to better adults and more habitats for humanity, and also we’re making our great meritocracy even more fine-grained. And that’s probably not completely wrong–there’s probably a marginal student who studies harder to get into a good college and ends up accidentally learning something useful or joyous as a side effect, and so on.

    Come to think of it, when I wanted to drop out of high school and teach myself, I used college admissions as an argument when persuading my parents to go along with it (not an easy task)–“I can either be a mediocre student because I hate school, which would tank my chances of getting into an elite college, or I can drop out and do something weird, which at least gives me a shot.” That’s I hope the kind of outcome the people running college admissions are shooting for–pushing students and their families to actually try to optimize years 14-18.

    Almost certainly still net negative for a variety of reasons.

    • Conrad Honcho says:

      when I wanted to drop out of high school and teach myself, I used college admissions as an argument when persuading my parents to go along with it (not an easy task)–“I can either be a mediocre student because I hate school, which would tank my chances of getting into an elite college, or I can drop out and do something weird, which at least gives me a shot.”

      If you don’t mind sharing, how’d that work out?

      • honoredb says:

        It worked out well. Hedonically (my actual motivation), I got to be mostly outside a power structure for a few years–I could make myself a sandwich whenever I wanted! College-wise, I pursued some passion projects, took a bunch of standardized tests, did well, applied to Harvard for the bragging rights and was waitlisted but got in to my actual top choice, University of Chicago, partly due to lucking into an admissions officer who was deep in one of the weird areas I was deep in and so could verify I’d actually done something.

        • kwr says:

          Bam. You nailed it honoredb. Admissions officers simply can’t find enough candidates like you. But that kid would get in every time. You went deep in a subject and someone knew you actually did something. You can’t imagine how many ISEF winners and science olympiad participants get into college and never pursue their topic again or don’t even pursue STEM majors.

  42. eqdw says:

    Hello. I work at a major company involved in job search, job ads, hiring, etc. And I would like to share with you something from a quarterly status update I saw the other day.

    6. All the hand-wringing about getting into good colleges is probably a waste of time, unless you are from a disadvantaged background. For most people, admission to a more selective college does not translate into a more lucrative career or a higher chance of admission to postgraduate education. There may be isolated exceptions at the very top, like for Supreme Court justices.

    Status update presentation had a slide outlining the results of a user study we did. We surveyed employers to find out what are the most vs least important details they look at when making a hiring decision.

    Out of something like 20 different options surveyed, “where the candidate went to college” was rated dead last in importance. “Formatting of resume” was rated as more important than “where they went to college” for making a hiring decision.

    This would seem to confirm conclusion #6 above

    • SamChevre says:

      I keep hearing this, and I flatly don’t believe it.

      It may be true at the “made it to an on-site interview as an experienced professional” stage. But I am willing to bet a reasonable amount that when applying to McKinsey, or Bain, or BCG, or Morgan Stanley, or Goldman Sachs right out of undergrad, a degree from Harvard gets you a lot more interviews than one from the University of Hartford (assuming the same classes and class rank).

      And in turn, top-tier law and MBA programs select heavily from people who worked at those top-tier firms.

      • Randy M says:

        Maybe it’s the case that middle tier employers are satisfied with middle tier college graduates. Elite schooling might only matter for getting elite employment.

        • Andrew Cady says:

          A very small proportion of advertised positions are hiring new grads at all. (Well under 10%.) Employers generally don’t care about college because they are hiring based on work experience.

        • Urstoff says:

          Entry-level positions at non-elite employers (which is most of them) will get tons of embarrassingly bad applications with “what the fuck were you thinking” cover letters or very obvious mistakes on their resume, so I can see why “formatting of resume” would be a higher priority than where the candidate went to college. However, after making that first cut, the specific college might matter more, but I would wager that skillset and cover letter still matters more.

      • Charles Kinbote says:

        The hiring personnel may not care what college you went to, but the hiring institution makes the decision for them by devoting unequal resources to hiring at each college. McKinsey might send 30 people to interview Harvard students and only 1 to UMass. But yes, this would apply only to hiring people right out of school. I’m analogizing from my own experience with a law firm. I don’t think the big consulting companies and banks work much differently–the hires are just younger.

        As for mid-career hires, the most useful heuristic for finding a very good attorney at my firm is where they graduated from law school. The less-selective the school, the better the attorney. For obvious reasons–they wouldn’t have made it this far if they weren’t good at their jobs.

        • SamChevre says:

          e less-selective the school, the better the attorney. For obvious reasons–they wouldn’t have made it this far if they weren’t good at their jobs.

          It amused me looking through the list of new Skadden Arps partners that the one new US partner who didn’t go to a high-ranked school is in tax-which is an incredibly nerdy legal field (I work with corporate M&A tax attorneys).

      • eqdw says:

        I mean, I am also surprised at how stark the data was. I don’t have access to the raw data, I don’t have access to the survey, I know nothing about how it was administered or whatever.

        I’m just reporting what I’ve seen. Take it as evidence, not as slam-dunk proof.

    • Reasoner says:

      What was most important?

      • eqdw says:

        I don’t remember, and I can’t find the slides in any of the drives I have access to so perhaps I’ll never know

    • OriginalSeeing says:

      Where can a person find a good resume format? haha

  43. VirgilKurkjian says:

    I dunno, “wandered into science, found a good position at a good lab, worked up the ranks to become a lab director, and ended up making great discoveries” describes me (ok, no great discoveries just yet) and pretty much everyone else I know in my field and related fields. We’ve developed a very good nose for climbers and resume-padders, and that approach is not competitive, at least in the fields I know.

    Take a look at what Paul Graham has to say as well.

  44. John Schilling says:

    One possible consequence of the increased number of applications per student, is the increased possibility of “statistical rejection”. Under the old rules, a reasonably competent student would be almost certain to pass the Harvard entrance exam and earn a spot at Harvard, and ditto Yale, Dartmouth, etc, but the exams were tedious so you picked the college you wanted to go to and presuming you were right about the “reasonably competent” part, you’re in. Now let’s assume that every reasonably competent applicant copies their application to all ten Ivies, and the equilibrium is that Ivy-league schools reject 80% of “reasonably competent” students – with 10x as many applicants, they’re sending twice as many letters of admission, but each accepted student has an average of two offers on the table and the number of attendees is fixed(*).

    The schools don’t coordinate, and each has no basis for preferring one “reasonably competent” student over any other, so the rejections are going to be essentially random. And a student competent enough to have been guaranteed a slot at Harvard under the old rules, now has a 0.8^10 = 11% chance of being rejected by every school they applied to, even Cornell won’t have them, now you have to go to some crappy state school die in a ditch.

    That 11% chance of complete rejection despite competence, is going to make college admissions look more competitive than it actually is. Er, was, because now everybody afraid of winding up in the unlucky 11% is going to pay for some test-prep classes and load up on extracurriculars that make them look a little bit more than just “reasonably competent”, and shift the odds in a direction even scarier for students who settle for demonstrating their reasonable competence without exaggerated signalling. Same with the part where, if you actually had a strong preference for Harvard over Yale, your odds of getting what you really want dropped from ~100% to 20%.

    * First-order approximation only, and this isn’t the only stable equilibrium

  45. rbwabd says:

    On the UC selectiveness issue, i remember reading that Berkeley or Caltech (which is not part of UC system) have race-blind, entirely test-driven admission processes, and hence disproportionately high Asian student populations. This would have increased selection pressure on all other ethnic groups. Not sure how admissions work across the UC system as a whole.

    • Jon S says:

      Caltech’s admission process isn’t strictly test-driven, there were still plenty of subjective elements on the application when I applied ~15 years ago. But to a first approximation, I think they’re running a model to predict applicants’ GPAs if they matriculate at Caltech. For the most part they admit the students with the highest predicted grades. Note also that Caltech has much less grade inflation than most US universities.

      • Steve Sailer says:

        Around 2002, I was walking across Caltech’s campus. I came upon a young lady undergrad conducting a tour for high school students and their parents. A dad asked her, “How’s the homework situation?”

        She replied, “Well, when I was a freshman …” and then she began sobbing. After about 15 seconds of tears, she rallied: “But now that I’m a sophomore, it’s much better,” she said with a heartbreakingly hopeful smile.

  46. Picador says:

    “This culminated in the creation of the College Board in 1899, which administered an admission test that acted as a sort of great-great-grandfather of the SAT.”

    I love the idea of 19th-century parents being like, “You should study Latin, it’ll help you on the SATs. No, I mean, it’ll REALLY help you, because 75% of the questions are about Latin declension and conjugation.”

  47. Deiseach says:

    Commentary on this: Harvard started out as a de facto seminary, Yale was explicitly so at its foundation. William and Mary university was set up by royal patent to provide an educated Anglican elite governing class; the original purpose of colleges was to educate and train clergy and the rulers. So while there may have been plenty of places, and ambition to fill those places, they weren’t for the ploughman’s son (unless he was very bright and had a call to the ministry) to attend. It was slowly during the 18th and more so during the 19th centuries that they became primarily secular institiutions (and in this, they were explicitly copying the great English universities which had started out as clerical institutions and during the Reformation become Anglicanised).

    So “college for everyone” was certainly not the aim or even during the 19th century expansion, the expectation: it was to provide the future governing class with a good education and turn out qualified candidates for the professions. Smart lower-middle and working class boys might make it there, but this was implicity acknowledged as lifting them out of their original class. College was open to “all”, where the “all” were “of the class and rank able to pay for a college education out of their private means”.

    As for fags, not in universities. Those were sizars or the equivalent term. Again, another way for smart but poor kids to finance their education. Fagging happend in public schools (which are private fee-paying schools) between the ages of twelve and sixteen as servants to the boys in the Sixth Form. For a good description of the system pre- and post the reforms of Thomas Arnold at Rugby, see Tom Brown’s Schooldays.

    As for the sexual element of single-sex public schools, there is Stephen Fry’s description in Moab Is My Washpot of his experiences at school, and C.S. Lewis’ description of his time at another in Surprised by Joy:

    One’s first hours at a public school are unforgettable. Our House was a tall, narrow stone building (and, by the way, the only house in the place which was not an architectural nightmare) rather like a ship. The deck on which we chiefly lived consisted of two very dark stone corridors at right angles to one another. The doors off them opened into the studies — little rooms about six feet square, each shared by two or three boys. The very sight of them was ravishing to a boy from a Prep. school who had never before had a pied-à-terre of his own. As we were still living (culturally) in the Edwardian period, each study imitated as closely as possible the cluttered appearance of an Edwardian drawing-room; the aim was to fill the tiny cell as full as it could hold with bookcases, corner cupboards, knick-knacks, and pictures. There were two larger rooms on the same floor; one the “Pres’ Room”, the synod of Olympus, and the other the New Boys’ Study. It was not like a study at all. It was larger, darker, and undecorated; an immovable bench ran round a clamped table. But we knew, we ten or twelve recruits, that not all of us would be left in the New Boys’ Study. Some of us would be given “real” studies; the residue would occupy the opprobrious place for a term or so. That was the great hazard of our first evening; one was to be taken and another left.

    As we sat round our clamped table, silent for the most part and speaking in whispers when we spoke, the door would be opened at intervals; a boy would look in, smile (not at us but to himself) and withdraw. Once, over the shoulder of the smiler there came another face, and a chuckling voice said, “Ho-ho! I know what you’re looking for.” Only I knew what it was all about, for my brother had played Chesterfield to my Stanhope and instructed me in the manners of the Coll. None of the boys who looked in and smiled was a Blood; they were all quite young and there was something common to the faces of them all. They were, in fact, the reigning or fading Tarts of the House, trying to guess which of us were their destined rivals or successors.

    It is possible that some readers will not know what a House Tart was. First, as to the adjective. All life at Wyvern was lived, so to speak, in the two concentric circles of Coll and House. You could be a Coll pre. or merely a House pre. You could be a Coll Blood or merely a House Blood, a Coll Punt (i.e. a pariah, an unpopular person) or merely a House Punt; and of course a Coll Tart or merely a House Tart. A Tart is a pretty and effeminate-looking small boy who acts as a catamite to one or more of his seniors, usually Bloods. Usually, not always. Though our oligarchy kept most of the amenities of life to themselves, they were, on this point, liberal; they did not impose chastity on the middle-class boy in addition to all his other disabilities. Pederasty among the lower classes was not “side”, or at least not serious side; not like putting one’s hands in one’s pockets or wearing one’s coat unbuttoned. The gods had a sense of proportion.

    The Tarts had an important function to play in making school (what it was advertised to be) a preparation for public life. They were not like slaves, for their favours were (nearly always) solicited, not compelled. Nor were they exactly like prostitutes, for the liaison often had some permanence and, far from being merely sensual, was highly sentimentalised. Nor were they paid (in hard cash, I mean) for their services; though of course they had all the flattery, unofficial influence, favour, and privileges which the mistresses of the great have always enjoyed in adult society. That was where the Preparation for Public Life came in. It would appear from Mr. Arnold Lunn’s Harrovians that the Tarts at his school acted as informers. None of ours did. I ought to know, for one of my friends shared a study with a minor Tart; and except that he was sometimes turned out of the study when one of the Tart’s lovers came in (and that, after all, was only natural) he had nothing to complain of. I was not shocked by these things. For me, at that age, the chief drawback to the whole system was that it bored me considerably. For you will have missed the atmosphere of our House unless you picture the whole place from week’s end to week’s end buzzing, tittering, hinting, whispering about this subject. After games, gallantry was the principal topic of polite conversation; who had “a case with” whom, whose star was in the ascendant, who had whose photo, who and when and how often and what night and where…. I suppose it might be called the Greek Tradition. But the vice in question is one to which I had never been tempted, and which, indeed, I still find opaque to the imagination. Possibly, if I had only stayed longer at the Coll, I might, in this respect as in others, have been turned into a Normal Boy, as the system promises. As things were, I was bored.

    • Deiseach says:

      Universal education only came in late in the day; in Britain it wasn’t until 1870 that non-denominational elementary school boards were created (if you remember the Sherlock Holmes reference to Board Schools, this is what he’s talking about) and not until an act of 1880 that compulsory education came in. Even after that, attending school after the age of twelve to fourteen was extraneous, and why such schools were called colleges (not to be confused with university colleges); you were educated enough and old enough to work at that age, unless there was a good reason for you to continue your education (such as being of middle class or higher, or a bright spark like Newton). My grandmother was unusually educated in that her father continued sending her to school past the age of fourteen, her daughter (my mother) by contrast left at fourteen to go into work (and you can see the reduction in family circumstances behind that).

      The relatively high literacy/education rate in America, as instanced in the post, came about as an effect of wanting a literate population who could read the Scriptures (hence the emphasis on Greek, Latin and Hebrew, the Scriptural languages) and why the early colleges were primarily or first set up to educate clergy. “Education for all” came late in the 19th century and downstream of that “college for all”. If (pulling figures completely out of thin air) 70% of your population is literate but you only realistically expect 5% to attend some form of college, that’s entirely different situation to 97% literacy and expecting 60%+ to go to some kind of college/further education.

      • bean says:

        It probably bears pointing out that there was an informal education system after 12-14, based on apprenticeships. At the time, the theory was that you had to start people young, because they couldn’t possibly learn the trade if they waited longer. This was a fairly serious problem for the RN, who wanted to get seamen from the merchant trades, but faced a dwindling supply in the later part of the 19th century. They started trading boys themselves, but this eventually became controversial as they required 12 years adult service (after 18, IIRC). Sorry if this is vague. I was rather surprised by it all, but I don’t have the book to hand.

        • Deiseach says:

          You’re entirely correct about apprenticeships and the like, bean. But today, where (a) parents want their kids to have better lives, so they encourage them to go for white collar jobs which necessitate some kind of degree and (b) increasing automation/efficiency/redundancy and the rest of it means, as Scott says, that knowledge-worker jobs are now not alone the best chance but the only chance to have a decent life.

          So the push for college only increases, and it doesn’t help that governments regard it as a panacea (“send the kids from deprived backgrounds to college to make sure they get good middle-class jobs and good middle-class lives! poverty solved!”). This just means that the aura surrounding top schools gets even more burnished; applications to all the colleges may be falling in absolute terms, but that certainly does not mean that Harvard is getting fewer applications – I’d imagine the reverse: the necessity to stand out from the crowd of ‘everyone has the same degree’ means that you have to wave ‘I have a degree from Harvard not from Smallville U’ so the pressure to apply is there (like Scott’s mother).

        • Garrett says:

          > They started trading boys themselves

          Rum, sodomy and the lash?

    • Nick says:

      One of Lewis’ points later in that chapter, or in one of the following chapters (you got me to read practically the whole book, damn you, Deiseach) is that he had hardly any free time during his school years. Between schoolwork, “games,” and fagging he could scarcely keep up, and hardly anyone around him had time for any fun either. It’s not exactly one to one with the America of today, but I wonder to what extent this was deliberate, given that a lot of parents and teachers don’t want louts like teenage Lewis to have free time.

      What ruins kids’ lives worse, on the whole? The rat race, or sex, drugs, and rock and roll?

      • Steve Sailer says:

        My recollection of C.S. Lewis’s autobiography is that homosexuality was a massive problem at his school.

        • S_J says:

          I think this is an instance where people can quibble over the meaning of homosexual. C.S. Lewis describes same-sex attraction between older teenage boys and younger teenage boys in an all-male environment, but also claims that the boys would have pursued teenage girls, if such girls were available.

          Is that a sign of situational same-sex-attraction, or an internal orientation towards same-sex behavior that could only be permitted in an all-male environment?

          The quotes from Deiseach are mostly from Chapter 6 of Surprised by Joy. At the beginning of Chapter 7, Lewis steps aside from his narrative to note–and defend–a decision to not to condemn the behavior of the Bloods and Tarts at the school.

          Here’s a fellow, you say, who used to come before us as a moral and religious writer, and now, if you please, he’s written a whole chapter describing his old school as a very furnace of impure loves without one word on the heinousness of the sin. But there are two reasons. One you shall hear before the chapter ends. The other is that, as I have said, the sin in question is one of the two (gambling is the other) which I have never beentempted to commit. I will not indulge in futile philippics against enemies I never met in battle.

          (“This means, then, that all the other vices you have so largely written about…” Well, yes, it does, and more’s the pity; but it’s nothing to our purpose at the moment).

          Lewis spends quite some time in Chapter 7 detailing how his short experience at that school turned him into, in his own words, a Prig. He was very low on the social scale, and would be punished if he acted too uppity for his position. But he found himself passing judgement on his social uppers.

          He counts the environment that produces Prigs as a worse moral problem than an environment that encourages same-sex interaction. Probably because, as he hints in the quote I gave, that the pride and haughtiness of being a Prig is a sin that he is very familiar with.

          • Nick says:

            He counts the environment that produces Prigs as a worse moral problem than an environment that encourages same-sex interaction. Probably because, as he hints in the quote I gave, that the pride and haughtiness of being a Prig is a sin that he is very familiar with.

            I think this is a good bit of why Lewis was willing to talk about such sins, but there’s longstanding reason for considering sins such as pride and haughtiness worse. Recall that pride is long considered the most deadly of the seven deadly sins. And insofar as vices are inversions of virtues, by the principle that corruption of the best is the worst, higher virtues make worse vices. Dante, for instance, placed Lust on only the second circle of Hell, the first one where punishments begin. Deeper lay gluttony, greed, wrath, heresy, violence, fraud, and treachery. Granted, sodomites are under the circle “violence,” their sin being violence against nature, but Lucifer, whose vice was pride and sin was treachery against God, lies deepest of all.

      • Quite a while back I was at an alum event for my high school (University of Chicago Laboratory School–high end private school). My conclusion from listening to the school people was was that they believed in “the Devil finds work for idle hands” theory of education. Keep the kids sufficiently busy with homework and after school activities and such and they won’t have time to do drugs or get pregnant.

        Not a theory I subscribe to. They also won’t have time to educate themselves.

        • Steve Sailer says:

          I always had the same GPA from high school through MBA school: 3.7 to 3.8. I maybe could have gotten a 4.0 if I’d redirected extra effort from courses I really liked to courses I didn’t find valuable, but I preferred to work extra hard on some courses at the expense of other courses.

          I enjoyed the lack of busy work in the 1970s, such as the virtual nonexistence of Test Prep. When I was 14 I spent the summer of 1973, for example, reading, very slowly, the first 200 or 300 pages of Adam Smith’s “The Wealth of Nations” (I skipped the “Digression on Silver”). It was a challenge to me to read more than, say, a page per hour, but I came away grasping that an Enlightenment genius like Smith really was vastly more enlightened and civilized than I was.

  48. Null42 says:

    Oh, come on, Scott, you can do this one. 😉

    Harvard tuition is $45,278 according to Google.
    Price of ground beef is $3.725 according to the Bureau of Labor Statistics.
    Harvard tuition is thus 12,155 pounds of beef, or about an 8500% increase.

    You’re welcome. 😉

    (Great job on the rest of the article!)

    • moonfirestorm says:

      Is ground beef the same product as a side of beef, though? My understanding is that ground beef is lower quality meat, and I’m skeptical that Harvard will accept that.

      You’re sitting there with your U-Hauls full of ground beef like a sucker, and you lose out to the guy who’s leading a couple dozen grass-fed naturally-raised cattle into the admissions office. It’s just another arms race.

      • Nornagest says:

        Ground beef is lower quality but a side of beef has less labor going into it, and includes lots of inedible bones. I expect it’d work out to about the same price, or slightly lower for a side.

    • J Mann says:

      Objection – a 141 lb side of beef =/= 141 pounds of ground beef. 🙂

      I got bored before I found a market price, but I found a bunch of farm web pages that will sell a half beef for $4-5 per pound.

      • Picador says:

        Also, the graph has an error: it says 161, not 141. That’s going to show up all over reddit and everybody is going to miscalculate how much beef they have to pay Harvard every year.

  49. JPNunez says:

    Surprised I did not see the Flynn effect mentioned. If SATs are basically IQ tests, then yeah, college should be harder to get into…for previous generations.

    I know I would have had a hard time getting into my college a few years after I got into, but then they thankfully changed the admission tests giving me plausible deniability.

  50. benquo says:

    What you’re describing is a major cultural and economic shift from a system where even elite colleges functioned as a sort of utility, fee-for-service with some standards, to qualify people in some objective sense for certain sorts of activity, to a system in which colleges function as a kind of sorting mechanism for an increasingly positional political economy.

    It’s not plausibly an uncaused coincidence that this coincided with the time period in which the US Government rapidly centralized the economy under central and increasingly unprincipled management in the context of the World Wars, and the US population shifted from majority-rural to majority-urban.

    Managerial and knowledge-worker jobs are no longer a critical expense for the sake of which we reluctantly tax free households participating in direct production – they’re the main way people can live OK lives, so we try to make there be as many of them as we can.

    More thoughts on the problem here.

  51. DV says:

    I think you’re missing the effect of the USNews rankings for admissions and pricing decisions. Colleges and Universities scrambled to raise their rankings and they realized very quickly that “selectivity” mattered greatly, so they worked to make application easier and encouraged as many students to apply as possible (with no intention of actually admitting them). They also realized that % of students with financial aid and average amount of financial aid mattered greatly. So instead of charging $20,000 and giving out limited financial aid, they charged $30,000 and gave out tons of financial aid. Those who could afford to pay the higher price did; everyone else paid about the same, but felt special because now they were getting lots of financial aid, and it boosted the scores in the rankings. As a prof for the last 25 years I saw first hand both of these decisions being made.

    • Freddie deBoer says:

      Colleges and Universities scrambled to raise their rankings and they realized very quickly that “selectivity” mattered greatly, so they worked to make application easier and encouraged as many students to apply as possible (with no intention of actually admitting them)


    • koreindian says:

      Goodhart’s Law strikes again!

  52. Freddie deBoer says:

    I ran the numbers myself several years ago. Out of 3000+ accredited two- and four-year colleges, something like ~150 reject more students than they accept. The large majority accept almost every student who applies. And of course only half of the population will ever enroll in college and only a third will ever finish. The people who say this is a niche problem are correct.

    • Freddie deBoer says:

      Oh, and – you’re correct that generally, declines in public funding compel schools to admit more students, not fewer. That’s because most schools are tuition-dependent and require high (and typically growing) enrollments to survive financially. It’s precisely what’s happening in my university, CUNY; Brooklyn College has seen record enrollments exactly because state funding has dried up so substantially.

    • Deiseach says:

      The large majority accept almost every student who applies.

      I’ve long had a suspicion that as regards Irish universities, they work on the principle of “accept everyone you can cram in on the points race, get their fees, then expect 50% of them to drop out after failing First Year. You get to keep the fees*, but don’t have to build extra capacity since the numbers going on to actually graduate will remain the same. Profit!”

      Granted, things may have improved in recent years and I have no idea how things are shaking out with all the ‘new’ universities and colleges of technology springing up, but back when I worked in adminstering student grant applications the difference between first year applicants and second years continuing on was striking.

      *Rules at the time were that fees were only refunded if the student dropped out within a particular time frame, go over that and the college kept the fee.

      • whereamigoing says:

        “the difference between first year applicants and second years continuing on was striking”

        Same thing in Estonia, though I don’t know if for the same reason.

      • BBA says:

        This is basically how the scammiest of the American for-profit colleges worked – except the courses were online, so costs to the college were even less (while they still charged the same astronomical tuition as everywhere else – puuuure profit!), and nearly all the students had federal loans that paid up front and left the students on the hook.

        These schools were known for advertising on daytime TV, targeting the unemployed and unemployable. It’s one thing to sell degrees that nobody will respect, but when it’s in the college’s interest to enroll as many students as possible who have no chance of ever graduating… I just don’t know.

  53. Forlorn Hopes says:

    Does section 3.2 adjust for test scores or other merits of measurement?

    If the percentage of Asians in Harvard’s incoming class was higher than the percentage of Americans who are Asian; but lower than the percentage of Americans who got over X% on the SAT, I would say that’s Asians being under-represented.

    • whereamigoing says:

      Judging by this link, the graph isn’t adjusted for test scores (indeed, it would be surprising for Harvard to point out such differences), but what’s important for Scott’s point is that Asians are less under-represented than before, so under-representation can’t be a cause of perceptions of increased competitiveness.

  54. RalMirrorAd says:

    The public school spending chart is shown on a per student basis. Has the absolute spending on state universities gone down or has it simply not grown fast enough to accommodate the shift from private to public schools? Is there a chart that shows the combined per pupil per year spending for public universities?

    Also, great post.

  55. AlphaGamma says:

    (These sorts of servants were colloqually called “fags“; some etymologists speculate that a tendency for rich boys to sexually abuse their servants led to the current use as a slur for gays).

    AFAIK fagging was the system at public schools (in the British sense) rather than universities, and was one where younger boys acted as servants to older boys- the hierarchy was based on age rather than wealth. It was abolished in the 1960s and 70s- I had a teacher who claimed that he had had to serve as a fag, but it had been abolished before he could benefit from it.

    Poor students who had to work as servants in exchange for tuition, like Newton, were called sizars in Cambridge and Dublin or servitors in Oxford. There was a whole hierarchy of social ranks above them (with noblemen at the top), differing in how much they paid, where and what they ate*, and in some times and places their academic dress.

    *At Cambridge, the rank below nobleman was fellow-commoner, so called because they ate with the Fellows of their college.

  56. rlms says:

    Suppose in the old days, each top student would apply to either Harvard or Yale.

    The system in the UK enforces this; you can only apply to one of Oxford and Cambridge (unless you play organ), and if you’re applying for medicine you have to choose four universities rather than five for other subjects.

    • Tarpitz says:

      Is that still true? I know you couldn’t apply to both Oxford and Cambridge when I did (2001) but I thought I’d heard that restriction had since been done away with.

      • rlms says:

        Pretty sure it is (for non-organ-scholar undergraduates).

      • whereamigoing says:

        At least it was true a couple years ago.

      • The original Mr. X says:

        It was still the case in 2009, when I was applying, unless you were applying for an organ scholar, in which case you could apply for both.

        • Garrett says:

          > organ scholar, in which case you could apply for both

          As in … pipe organ? Of all things, why would that be the exception?

          • Lambert says:

            The Church.

          • rlms says:

            Yes. The universities of Oxford and Cambridge are divided into around 30 colleges each, which are basically glorified, old and often extremely rich halls of residence. For historical reasons, most colleges have a chapel and associated choir that sings at services. College choirs are accompanied/partially directed by student organ scholars (usually but not always music students, I think always undergraduates). Presumably there are few enough candidates that it’s necessary to let them apply to both universities to avoid a situation where one university doesn’t get enough.

          • ec429 says:

            For historical reasons, most colleges have a chapel

            And of course now I have to fly the flag for Churchill by pointing out that we technically don’t. (And instead of an organ scholar, we have a Music Sizar.)

            colleges […] are basically glorified, old and often extremely rich halls of residence

            I don’t know about Oxford, but in Cambridge your college is responsible for your teaching apart from lectures. Mostly that means that your supervisions (1:1 or 1:2 teaching sessions) will be with fellows and postgrads from the same college, although not always (especially if your subject is an obscure one or one that your college doesn’t have many places for).

          • rlms says:

            That’s a good story:

            Crick had agreed to become a fellow on the basis that no chapel be placed at Churchill. A donation was later made by Lord Beaumont of Whitley to Churchill College for the establishment of one, and the majority of fellows voted in favour of it. Sir Winston Churchill wrote to him saying that no-one need enter the chapel unless they wished to do so, and therefore it did not need to be a problem. Crick, in short order, replied with a letter dated 12 October 1961 accompanied by a cheque for 10 guineas saying that, if that were the case, the enclosed money should be used for the establishment of a brothel.

            I was exaggerating a little, although other than my DoS most of my supervisors weren’t at my college.

  57. rlms says:

    I’m not totally convinced by the Dale and Krueger paper. From the abstract:

    We find that the return to college selectivity is sizeable for both cohorts in regression models that control for variables commonly observed by researchers, such as student high school GPA and SAT scores. However, when we adjust for unobserved student ability by controlling for the average SAT score of the colleges that students applied to, our estimates of the return to college selectivity fall substantially and are generally indistinguishable from zero.

    So college selectivity *is* significant even after controlling for student quality as measured by SAT scores. It only ceases to be significant when you also control for some vague measure of ambition as signalled by the average SAT score of all the colleges they applied to.

    • NoRandomWalk says:

      I think that’s a reasonable approach to use as a control though.
      Assume students are good at knowing if they are capable/well-credentialed or not. Everyone applies to universities they think they can get into, the more confident you are in yourself the higher the average scores of colleges you apply to.
      Then, admins introduce a bunch of randomness into this, and admit some students and not others who had the same profile.
      If you have two students who only applied to Harvard/Yale/Princeton and one got in and the other didn’t, they’re the two students you want to compare to figure out if college is important, not ‘performance of Harvard admins’ vs. ‘performance of non-Harvard admins including those who didn’t apply to Harvard or applied to Harvard and 50 state schools’

    • Douglas Knight says:

      Are you concerned that this is a bad thing to control for? Are you concerned about p-hacking?
      These are very different concerns and you should spell it out.

      • rlms says:

        I don’t think they’re that different. Controlling for a lot of irrelevant things is one way of doing p-hacking. But I’ve not looked into it in enough detail to want to make a confident or specific claim. I mainly want to point out that the claim

        Although Harvard students earn much more than University of Podunk students, this is entirely explained by Harvard only accepting the highest-ability people. Conditional on a given level of ability, people do not earn more money by going to more selective colleges.

        is somewhat misleading; it’s not true (according to this paper) if you take the obvious interpretation that “ability” means SAT score. I’m also skeptical about the idea that average SAT score of colleges applied to is correlated with some kind of go-getterness that it has more of an impact on future earnings than which college you go to, but I don’t have an opinion on whether the effect of controlling for that is dubious or whether there is another mechanism (maybe average SAT score of colleges applied to is a proxy for socio-economic status or something).

        • Douglas Knight says:

          Sure, controlling for a lot of irrelevant things is bad, but so is controlling for a single irrelevant thing, or controlling for too many relevant things. And those are different.

        • NoRandomWalk says:

          If you want to claim that college admissions are uncorrelated to aptitude/grades/work ethic, fine, but the main claim people care about is ‘does Harvard give you more prestige or skills that result in higher earnings’. This paper seems to suggest ‘no’. Which is surprising to me, but sometimes research turns up surprising results and my emotion is not a reason to ignore the research (maybe a weak prior to be skeptical, but I haven’t seen any critiques of the paper to validate that skepticism).

          Can you unpack a bit more what model you’re positing?
          It’s possible that ‘average sat score of applied colleges’ is a proxy for prestige/class, not intelligence (I personally expect it to be a big mix of both).

          Would you agree that the paper shows:
          Although Harvard students earn much more than University of Podunk students, this is entirely explained by Harvard only accepting the highest-[whatever mix of self-perceived ability, socioeconomic class, etc. caused them to apply to elite colleges] people. Conditional on that, people do not earn more money by going to more selective colleges.

          • rlms says:

            If the paper found that after controlling for SAT score, or SAT score and intelligence/class measured in some obvious way (household income, parental education), college SAT score was not found to affect future earnings I would be moderately surprised (why do people try so hard to get into Harvard if it doesn’t matter? what happened to college equals signalling?) but not hugely. But as far as I understand this study, they found that college choice does affect selectivity even after you control for SAT score. When you control for average SAT score of colleges applied to (henceforth ASSOCAT) there’s no effect, but it seems fishy to me that this would be majorly relevant after you control for SAT score achieved (I expect ASSOCAT is a good proxy for intelligence by itself, but not after you take SAT into account, in the same way you would expect e.g. number of books read per year to be a reasonable predictor for chess ability in the general population, but probably not among chess players). You could make an argument for why ASSOCAT would carry information about intelligence beyond that given by SAT (maybe some people have a bad day on SAT and their perceived intelligence tells you more about their actual intelligence than SAT) but I think that’s highly implausible; as pointed out in this post there’s the huge opposing factor of people overestimating their ability (and maybe in some cases underestimating too).

            More plausible is that ASSOCAT carries significant information about class beyond SAT. You can imagine rich but thick kids who do badly on SAT but apply to competitive colleges because they’ve got a good chance of getting in due to other factors (being a legacy etc.) and that those other factors have a big effect on future earnings. But it still seems unlikely to me that this information is a bigger deal than SAT itself. Maybe I’ve got an unreasonably rosy view, but I would hope that the US is meritocratic enough that (within the population of college graduates) actual competence has more of an effect on earnings than daddy’s money. The other mechanism (the one they seem to suggest in the paper) is that ASSOCAT carries information about ambition or something; so the driven applicant who applies to better colleges than their SAT score suggests is wise have better life outcomes than the applicants with the same SAT who accurately estimate their ability. Again, it doesn’t seem that likely to me that this would have more of an effect than intelligence-as-measured-by-SAT (and if I’m wrong, someone needs to set up programmes to instill this mysterious beneficial quality in kids right away).

            So because I can’t see a reasonable mechanism for ASSOCAT to have the effect they describe, I’m dubious about the results. There are various ways they could have good the stated results without ASSOCAT being an important factor; I haven’t spent enough time reading the paper to suggest which are most plausible.

          • Douglas Knight says:

            RLMS, you seem to be complaining about a very narrow part of the paper. Your alternate hypothesis does not at all rebut the paper. It just rebuts Scott’s choice of the word “ability.”* The question both Scott and DK ask is the causal question of whether selective college has an effect, which must control for the prior state of the student, whether that is ability or class. Is ASSOCAT a good control? If it is class and you think class is important for future earnings, then yes. Lots of people claim that selective colleges are a way of buying class; this would be a direct rebuttal.

            (Which is not to say that ASSOCAT is actually a good measure of class, p-hacking, etc.)

            * but ability is a poorly specified word. Why not include class as the ability to extract money from certain power structures?

          • rlms says:

            @Douglas Knight
            The class mechanism is more plausible than the other two I mentioned, but it’s still not convincing to me. Aren’t we biodeterminists here? A purely environmental (and, even worse, shared environmental) effect having a dramatic effect on life outcomes should raise eyebrows, especially since intuitively I’d have expected “upper-class people (conditional on SAT score) can get into better colleges and hence better signal competence to employers” to be one of the main mechanisms class would affect earnings by. And that’s disregarding that ASSOCAT seems like a pretty weak proxy for class: “upper-class people are more optimistic with their college choice” is reasonable-sounding story that explains something rather than an obvious fact.

    • jblum says:

      Fwiw, I read the Dale Krueger paper a while back and, despite it fitting my priors, found it pretty unconvincing. A few issues with it from memory: 1) the sample of colleges it looked at was very narrow – maybe 20 – and all of them were name brands. There was no Podunk University so you were more talking about the difference between going to Vanderbilt and Yale or something like that. 2) the earnings measure was top coded at a pretty low level, maybe $250k or something, so if the belief is that you need to go to a top school to be an earnings star, this study didn’t even try to measure that. 3) i don’t believe it looked at industry so if you had a lot of “less competitive” school grads like Vanderbilt going to med school but a lot of “high competitive” school grads going into teaching, this would get you a result where school quality doesn’t lead to higher earnings. 4) related, the main result was that the quality of schools you applied to was more predictive than the quality of schools you attended. An alternative hypothesis would be that students looking to go into competitive/high earning fields would be applying to higher competitive schools. 5) the SAT scores assigned to the individual schools in the original 1976 paper were weird looking and didn’t correspond to what you would think of intuitively as their rank ordering of selectively. The tuition numbers were important in their regressions but they struck me as also reflecting selectivity and not only resources which was the author’s supposition.
      There was more, but bottom line for a very commonly cited paper i’d urge caution before considering it’s conclusions reliable.

  58. ana53294 says:

    From what other generations say, in the USSR, it was much harder for the previous generations to get to university. They had all kinds of formal and informal quotas (the Jewish exam questions were a thing), and it was in general harder for a child from intelligentsia to go to university than for a proletariat. One of the ways of getting admitted was to go work in a factory for six years or so, and go through the prole quota.

    A Slovakian co-worker told me it was the same in Slovakia. So it seems like at least for former communist countries, admission has become easier, with less bullshit (although now you may have to pay money).

    In Spain, it doesn’t seem like admissions have become harder or easier, and they have stayed more or less the same, at least from what other people have told me.

    • Kindly says:

      The impression I get from my relatives living in Russia is that college admission has been (in the USSR period) and still is based in large part on bribery and on having the right connections.

      • ana53294 says:

        Well, bribery is a thing, of course.

        But the bribery was to go to some types of universities, like the MGIMO or the RUDN (previously known as the Patrice Lumumba University). Those were universities for the party, and only party members would go there. In the case of the RUDN (which had the objective of educating the next generation of leaders in Africa and South America), there were mainly KGB agents, party members and the foreigners.

        But there were also plenty of other, normal universities, and most people during the Soviet era just got accepted there.

        • Douglas Knight says:

          I have been told that a habilitation (доктор наук) was a way of signalling KGB connections.

    • Viliam says:

      In Slovakia, students usually don’t pay for university education. University education is paid by government. The government evaluates universities on criteria such as “do they employ at least N professors?”. Universities that pass these criteria get paid by the number of students.

      This leads to obvious strategy for the universities: take as many students as possible. Doesn’t matter if they are literally retarded; each one of them still brings you extra money.

      Even the universities who do not primarily want to play this game, still have to sacrifice some of their value to Moloch. Because if they make education nontrivial for students, many students will choose a university that doesn’t make problems instead. And if you get few students, you get little money from the government, so you have to fire some of your teachers; and that would be an irreversible step the universities wants to avoid.

      The situation is made worse by the fact that the best students often choose to study in Czech Republic instead; specifically in Brno, which is geographically close to Slovakia; 90 minutes by train from our capital city. (Czechia provides free education for students from Slovakia, and the languages are mutually intelligible, so it’s just a question of traveling a few extra miles.) This leads to vicious circle — formerly good Slovakian universities have to admit many bad students, because not enough good students apply; good students decide not to study in Slovakia, because they know they would have a lot of bad classmates and therefore slower lessons.

      I don’t see any incentive to make university admissions nontrivial. The universities are literally paid by the number of students. Fewer students would mean less money. You can’t take the extra money from students, because they are used to receive free education (and the best ones can always get it in Czechia). From the government’s perspective, more university students, regardless of their quality, means less unemployment, which is politically desirable.

      (For reasons I don’t really understand, employers in Slovakia care a lot whether you have a university diploma or not, but don’t care at all whether you have the diploma from a prestigious university or from a diploma mill. Therefore, being a prestigious university is something that can make you proud as a person, but economically it does not make a difference.)

  59. vV_Vv says:

    My understanding is that college used to be an institution that catered to the intellectually curious and was vaguely high-status but wasn’t really seen as preparation for work, unless you actually wanted to work as a college professor.

    Sort of like taking violin lessons now: you do it if you are interested in music, and it is vaguely high-status, but it is not really useful for getting a job, unless you want to be a professional musician and even then there are many successful professional musicians with no formal musical education.

    College however became a mainly credentialing institution for the general labor market. There might be legitimate (in the sense of market-efficient) reasons for it: the labor market might have just became much more competitive.

    In your review of On the Road you noted how easy it was, as late as the 1940s, to get a job: if you were an able-bodied man and you just showed up at a business chances were that they had a job for you. And until couple few decades before, countries such as the US, Canada, Brazil and Argentina were all encouraging mass immigration from Europe. And before that there was the trans-Atlantic slave trade. This is evidence that the economy of the Americas, and possibly to a certain extent Europe, was labor-limited.

    But the economy of modern developed countries is limited by high-skilled labor: e.g. Goldman Sachs, JP Morgan, etc. would be bigger if they could hire more competent quants, and there are lots of people who want these jobs but most of them are not as smart and conscientious to perform well, and the cost of hiring an underperformer can be high (between training costs, management costs and disruptions to the existing teams, they might well have negative productivity for years until you figure out and resolve to fire them), therefore Goldman Sachs, JP Morgan and so on rely on college degrees as a form of filtering.
    Google, Facebook, Apple, etc. also have the same problem, but at least Github commits might work as a partially viable substitute of college degrees. As more and more middle class jobs become more complex, they also start to rely on college degrees for filtering.

  60. Robert Jones says:

    My parents’ generation tells stories of slacking off in high school, not worrying about it too much, and knowing they’d get into a good college anyway. Millennials tell stories of an awful dog-eat-dog world where you can have perfect grades and SAT scores and hundreds of hours of extracurriculars and still get rejected from everywhere you dreamed of.

    Like another commenter, I doubt the accuracy of the memories, but if true, I’m not sure that this is bad. If young people today are working harder and slacking off less, then they’re better actualising their potential, whether or not they gain admission to a good university.

    • Robert Jones says:

      PS People live longer now, so it is rational to invest more resources in personal development at a young age, because you can expect a longer pay-off phase.

    • Tarpitz says:

      If what they’re working hard on actually has some long term benefit, rather than being Molochian busywork. I am not optimistic on this score.

  61. Rachael says:

    If you’re not Harvard material, who is?? I am demonstrably Cambridge material, and I think you’re definitely smarter and more hard-working than me.

    • vV_Vv says:

      If you’re not Harvard material, who is??

      Rich people.

    • Robert Jones says:

      It does seem surprising: I have the impression that Scott is cleverer than me and I was considered an obvious admit by Peterhouse. Perhaps Scott was a lazy child? Or perhaps his mother was right but his self-doubt caused him to sabotage his admission? Or perhaps Harvard is much harder to get into than Cambridge?

      • sohois says:

        As evidenced in this post, admission to US universities is no longer really about mere cleverness any more – you’ve got to do the extra curriculars, the AP classes, the entrance essays – many of which will pick for a certain type of driven student. Even if Scott was merely an average teenager in his effort levels that would probably be sufficient to exclude him from consideration for a Harvard position.

        The UK, meanwhile, is much more about academics. While I know that Oxbridge do conduct interviews with prospective candidates, by far the biggest driver in university admission is A-Level results, meaning any sufficiently clever student can swan into an end of year exam, get all As, and gain entrance to top tier institutions.

        • Robert Jones says:

          That doesn’t sound right: Cambridge rejects lots of straight A candidates.

          • AlphaGamma says:

            Oxbridge (and medicine anywhere) are the exception in many discussions of British university admissions. But the point stands that performance in a single high-stakes test is much more important than things like extracurriculars, essays/personal statements, references, etc. It’s just that that test is your interview (plus arguably STEP if you’re a mathematician) rather than your A-levels.

            Explanatory note: Oxbridge admission interviews are pretty much purely testing academic ability and ability to benefit from the institution’s style of teaching. They have nothing to do with checking that you are a well-rounded person. Medicine may be an exception here too…

          • rlms says:

            Fewer straight A* candidates though. From the most recent statistics, candidates with A*A*A* have a ~50% of getting in, compared with ~20% for all applicants. Plus (especially for sciences) many applicants will have four A-levels, and offers are generally but not necessarily made before A-levels are completed; I expect a candidate with A*A*A*A* already achieved would be in a very strong position.

          • whereamigoing says:

            My impression is that Cambridge cares a bit more about extracurricular stuff, whereas Oxford purely tests for academic ability. At least when applying to computer science specifically, you need to take the MAT, which is basically an IQ test (little background knowledge needed), then do an interview (which is purely academic, like a CS job interview without any bulls**t puzzles or career questions) and then get decent points on a country-dependent exam (to the extent that’s unmeritocratic, it’s not Oxford’s fault).

            The UCAS application does require a cover letter, but they don’t care about it too much (except perhaps as a tiebreaker). I got to talk to a few undergraduates and the admissions officer when I was there for my interview, and she said that she doesn’t remember the cover letter ever making a big difference, except one guy who they might have accepted if not for his really terrible cover letter.

            Other UK universities seemed somewhat less meritocratic though, though not as bad as US universities.

            Edit: Just remembered another anecdote — one interview started with the tutors asking about something on the applicant’s personal statement, the applicant replied with “Do you mind if we just skip to the technical questions?”, and then the tutors laughed a bit and said “sure”.

        • Tarpitz says:

          Also, don’t mistake the purpose of the Oxbridge interviews: they are there to help tutors distinguish between the academic abilities of exceptional applicants and merely bright ones, because A levels don’t provide enough definition at the very top end. If anything, extra-curricular interests are seen as a downside: they might be a distraction from academic work. In one awful-but-hilarious incident I know of, a tutor slept with an applicant, got her pregnant… and didn’t offer her a place.

          When this came out he was suspended for a year. Then promoted to chairman of the faculty.

          • johan_larson says:

            What are the interviews like? Are there skill-testing questions? Or is it more about showing yourself to be a proper young gentleman with just the right amount of ambition?

          • Deiseach says:

            In one awful-but-hilarious incident I know of, a tutor slept with an applicant, got her pregnant… and didn’t offer her a place.

            Oh goodness. The very cynical part of me replies that this was a test and the young person lost her slot because she showed lack of forward planning and inability to properly ration her time and effort; plainly she’d rely on last-minute cramming to pass exams and that wouldn’t be good enough, she’d never be able to keep up with the work.

          • rlms says:

            There will always be at least one largely academic interview, there are a few examples on youtube (e.g. this one).

          • The original Mr. X says:

            What are the interviews like? Are there skill-testing questions? Or is it more about showing yourself to be a proper young gentleman with just the right amount of ambition?

            When I was interviewing (for Classics) I had four interviews. Two were mostly about the books I’d mentioned reading on my personal statement (one at the college I applied to, one at another college, I suppose so that the admissions tutors could get a second opinion), and for the third I was given a passage about a book on education policy and had to discuss it with my interviewer. I also had to do something called the Language Aptitude Test, which basically consisted of being given sentences from made-up or obscure languages and trying to work out the grammatical rules. My last interview was going through this with teachers from the faculty, basically running over the questions I’d got wrong to see if I could spot my mistakes.

            So in answer to your question, whilst coming across as a “proper young gentleman with just the right amount of ambition” would doubtless help your chances of getting accepted, the interviews themselves are geared towards skill-testing.

          • Peter says:

            A similar story with Cambridge interviews here – for Natural Sciences[1]. Three interviews. One was a “tutorial” interview which was fairly general and they said it didn’t count for much. The other two were very much skills oriented. One of the things they’re after is your ability to think around your subject a bit and go beyond what you’ve been taught (I vaguely recall the word “flair” being used), there’s a bit of intellectual self-confidence being tested there.

   should give you a flavour of it. Like I say, it takes a certain amount of intellectual self-confidence to say, “oooh, good question” and yammer on for a few minutes while explaining the educated guess you’re making on the fly. This came very naturally to me, but can be intimidating to some.

            At least in my day, another thing that often happened was that your teachers would write a statement about you, and a good recommendation helped.

            [1] Cambridge has this odd thing where you register for big broad subjects, do fairly broad stuff in your first year, and specialise later on. You don’t sign up for e.g. Chemistry or Japanese or Sociology, you sign up for Natural Sciences or Modern And Medieval Languages or Social And Political Sciences. If someone had told me this was a plot by the Geology department to keep numbers up on their courses, I’d half believe them.

          • Nick says:

            If someone had told me this was a plot by the Geology department to keep numbers up on their courses, I’d half believe them.

            I don’t know about Cambridge, but this happens in American universities where there’s a core curriculum. For language courses, obviously, but also math, theology, philosophy, etc.

          • Robert Jones says:

            At least in my day, another thing that often happened was that your teachers would write a statement about you, and a good recommendation helped.

            A friend of mine who conducts interviews for his college says that he entirely discounts the schools’ recommendations because he has no way of knowing whether the person writing the recommendation has any understanding of the standard required.

          • Michael Watts says:

            Cambridge has this odd thing where you register for big broad subjects, do fairly broad stuff in your first year, and specialise later on. You don’t sign up for e.g. Chemistry or Japanese or Sociology, you sign up for Natural Sciences or Modern And Medieval Languages or Social And Political Sciences. If someone had told me this was a plot by the Geology department to keep numbers up on their courses, I’d half believe them.

            You might not like hearing about the US. That system is the norm for all US colleges; you aren’t expected to take courses in your specialization (“major”), or even declare what your specialization is, until year 3. For the first two years, you are usually required to take intro courses in a variety of different fields.

          • rlms says:

            The Cambridge system (for those subjects where this is the case, which isn’t all of them) is still a lot more specific than the US one. I think the most interdisciplinary you could get would be formally studying Human, Social, and Political Sciences and “borrow papers” from history, non-social psychology and Japanese studies. Most people will still be fairly concentrated in e.g. biology, or a couple of languages. The main incidence of people being forced to diversify is that first-year physicists end up also studying chemistry and materials science (or biology or computer science if they’re adventurous), whereas at most British universities they would just do physics and maths.

          • Tarpitz says:

            Much like others here, I had two interviews and two written tests at interview. The tests were on French grammar and logic (no prior knowledge of formal logic was assumed – these were challenging but informal puzzles).

            The French-orientated interview was actually based on discussion of a short passage from an English literary text I had been provided with an hour or so beforehand (whether this was to allow them to assess language and literary criticism skills separately or simply so that the same text could be provided to applicants to study every language I don’t know), combined with a brief section of French conversation.

            The philosophy interview involved perhaps ten to fifteen minutes each discussing the nature of philosophy as distinct from other subjects (on which I was shockingly bad), the nature of knowledge (on which I was tolerable but a little slow), and the nature of personal identity (on which I did well); I also explained a problem from the logic test that I had subsequently realised I got wrong and asked about the intrinsically circular nature of logic. The two dons present played clever cop/stupid cop, and the clever cop, David Bostock, is the one person I have ever met who could routinely, in one simple sentence, make you feel an utter fool for not having previously realised the obvious truth of what he was saying. The highest compliment I will ever receive was a moment two years later in which he briefly paused for thought before saying, “I don’t agree with you at all, but I can’t actually prove you’re wrong.” I went in pretty confident. I came out and my legs gave way about two steps down the stairs. Between both interviews combined, I spent perhaps ninety seconds on non-subject-matter-related pleasantries.

      • Wency says:

        My sense is that, as a white person with non-famous, non-legacy parents, you need a positively bizarre background to get into Harvard. Zuckerberg sold a company to Microsoft for a million dollars as a teenager — that sort of thing. And I don’t even know if he’s legacy or not (i.e., whether his parents went there).

        All near-perfect grades + SATs + lots of extracurriculars do is put you in consideration, so your application doesn’t instantly go to the trash. I basically had those things (just short of perfect on both SATs and grades), and I was wait-listed at Harvard back in the early 2000s, which means “avoided instant-trash status”, but I had no real chance. As I recall they wait-listed like 60% of applications.

        • Reasoner says:

          Zuckerberg sold a company to Microsoft for a million dollars as a teenager

          Huh? It’s not on his Wikipedia page.

          • Wency says:


            The software was Synapse, but I guess I remembered that wrong. They say offers were as high as $2 million, but they turned them down. Elsewhere Zuck explained that those companies wanted them to go to work for them, and he wanted to go to Harvard.

            Anyway, even without that payday, and even if those offers were fiction, he was doing some pretty extraordinary software dev work for a high schooler, and something tells me that Harvard admissions was aware of this. Plus he was maintaining the rest of that Harvard resume.

            Suddenly the fact that you’re President of your school’s Computer Club doesn’t count for so much.

        • j1000000 says:

          Zuckerberg is also one of the most successful people ever to attend Harvard, so I’m not sure you could generalize from that anecdote. I know a fair amount of white people who got into Harvard around Zuckerberg’s time, and they were your run-of-the-mill super geniuses who were debate club presidents, not people who turned down millions from Microsoft.

          (Edited to reflect that Zuckerberg did not graduate…)

    • eric23 says:

      I wasn’t “Harvard material” either. I didn’t bother applying there (didn’t expect to get in, but also didn’t like the culture). I hoped to go to Princeton but was rejected, and ended up going to Cornell.

      Why do I say I wasn’t Harvard material? My high school GPA was high but not perfect (too many hard AP classes and too little maximizing of extra credit/honors grade opportunities). My SAT score was near perfect. My 10 AP scores were perfect. But my extracurriculars were pretty thin. I was told that MIT (where I also applied) ranked me 5/5 in academic potential and 2/5 in intangibles/extracurriculars (they waitlisted then rejected me). I’m sure the other colleges I applied to rated me about the same way.

      I often see people who were “Harvard material” and went to Harvard, but who are less talented than me and have no greater career achievements. But based on the peculiar mix of academics and intangibles which we all know colleges are looking for, I’m sure these people are better “Harvard material” than I was. The same is true regarding Scott, who is almost certainly more talented than me, yet it’s easy to imagine that he failed the Harvard admissions criteria in about the same way that I did.

      • Robert Jones says:

        If it’s not too personal a question, why did not you not do the extracurriculars?

        • eric23 says:

          Partly because I was spending so much time on study that little time was left. But more because I was an awkward and introverted nerd who did not fit in most of the activities…

          • A Definite Beta Guy says:

            Does it even help to just be a member of the chess team or the math club? I’m guessing you not only need to participate, you need to actually be accomplished within your chosen extracurricular to be worth a damn at the top schools.
            Like, I was a top 50 competitive speaker in IL, but that just means there were 49 people ahead of me. If you’re Harvard, do you want the state champion, or #50? There are going to be a lot of champion mathletes, a lot of kids who won the robotics competition, a lot of All-State basketball players…does just participating even matter?

          • Robert Jones says:

            I’m not in the US, but I think you should do 5 or 6 things, and be genuinely good at one of them. Nobody expects you to be state champion in all 6.

            My school had a master for Oxbridge admissions, who identified the likely candidates around the start of their penultimate year, and advised them if their applications could be strengthened in certain areas. I’m pretty sure the school would fudge the selection criteria for the activities if necessary. I suspect this sort of thing goes some way to explaining the overrepresentation of public schools at Oxbridge.

          • Tarpitz says:

            My impression is that the widespread professed belief of British schools that Oxbridge is interested in the non-academic accomplishments of applicants is, with a few narrow exceptions (exceptional skill at playing the organ or rowing, mainly) simply wrong. And even for the rowing, you’d better be applying to read colouring-in geography at Gryffindor Oriel or Pembroke or some such. A few specific tutors might have their own particular (and usually contemptible) quirks (girls are ill-advised to apply for English at Hertford, for example) but by and large tutors are trying to select people they actively want to teach the subject to in very small groups. Hell, even Tom Paulin probably thinks he’s just rejecting all those girls because they’re thick.

            Now, whether the schools stress the admissions value of worthy extracurriculars because they really mistakenly believe Oxbridge care or because they think it’s a useful carrot to get teenagers to do things they’d like them to do for other reasons I am much less sure.

          • rlms says:

            Yes, Cambridge certainly stresses that non-academic extracurricular activities are not relevant and for the most part I think they’re being genuine. Some people get in on the basis of sporting talent but they will do graduate degrees (if you look at the boat race crews there are a lot of people doing MPhils in Management or similar). Possibly extracurriculars are considered by non-Oxbridge universities though.

    • fluorocarbon says:

      According to Wikipedia, the acceptance rate in 2015 for Cambridge was 21% and 17.5% for Oxford. Harvard’s was 6.2%.

      The admission rate for normal people is even lower than that since 14% of students are legacy (and legacy students have the higher admission rate of 34%). Harvard is also known for accepting famous people, people they think will become famous, and people from all across the US. This is by design, since it means that no matter where you go in America, the local elite will contain Harvard graduates. The news will also always be full of Harvard graduates. But that makes it very difficult for people to get into Harvard if they’re from a region that sends a lot of people to Harvard already and their only standout quality is being academically gifted.

      Or as Steven Pinker says:

      At the admissions end, it’s common knowledge that Harvard selects at most 10 percent (some say 5 percent) of its students on the basis of academic merit. At an orientation session for new faculty, we were told that Harvard “wants to train the future leaders of the world, not the future academics of the world,” and that “We want to read about our student in Newsweek 20 years hence” (prompting the woman next to me to mutter, “Like the Unabomer”). The rest are selected “holistically,” based also on participation in athletics, the arts, charity, activism, travel, and, we inferred (Not in front of the children!), race, donations, and legacy status (since anything can be hidden behind the holistic fig leaf).

      But the other side of that is (as Scott points out), it doesn’t really matter where you get your undergraduate degree in the US. It maybe matters if you want to become a supreme court justice, but that’s mostly for law school, which is a different beast entirely.

      • AlphaGamma says:

        According to Wikipedia, the acceptance rate in 2015 for Cambridge was 21% and 17.5% for Oxford. Harvard’s was 6.2%.

        When comparing these, it should be noted that British universities use a centralised application system, and applicants are only allowed to apply to 5 universities in total and only one of Oxford or Cambridge (in the vast majority of cases). This presumably cuts the number of applicants down somewhat.

        • fluorocarbon says:

          That’s a good point. I got my original statistics by Googling “admission rate” for each of the schools and looking at the sidebar in Google. I thought the data came from Wikipedia, but it’s not and I’m not sure where Google is pulling the information from. Regardless, the numbers in my original post are slightly wrong.

          Below is the actual data (still from 2015) taken from Cambridge, Oxford, and Harvard:

          Percent given offer: 25.8%
          Percent accepted offer: 81.1%
          (Note: these numbers do not match the Wikipedia article. They’re using some other document and I’m not sure why.)

          Percent given offer: 19.9%
          Percent accepted offer: 87.8%

          Percent given offer: 5.4%
          Percent accepted offer: 79.2%
          (Note: this is for the class of 2020. I’m not familiar with the application cycle in the UK, so I might be comparing slightly different years.)

          The individual numbers are slightly different, but the general theme is the same. Harvard gives offers to a lot fewer students even accounting for yield/acceptance rate. This is probably not true for most US institutions, but Harvard is Harvard, and I imagine most people who get in will want to go.

      • Robert Jones says:

        I’m not sure how comparable those percentages are because (with a tiny number of exceptions), applicants are only permitted to apply to one of Oxford or Cambridge, whereas per the OP it appears that applicants can and do apply to multiple Ivy League schools.

        There was a famous case of a student who was rejected by Oxford and then obtained a place at Harvard (Laura Spence): Gordon Brown (who was Chancellor at the time) accused Oxford (Magdalen) of having rejected her for elitist reasons.

    • Plumber says:


      The only person I knew who went to Harvard from Berkeley High School when I was there in the ’80’s had been elected “Northern California President of the Junior Statesmen of America” (basically a high school debate society), and my wife went to the Harvard law school in the late 1980’s (she dropped out before her last year), and we used to get the alumni magazine which was a pretty good read until they sent a letter saying “Pay up or we stop sending the magazine”.

      She went to a Catholic high school in Seattle where she say she was “an indifferent student”, and then went to the University of Washington – where she became a much better student “Because the classes were interesting”.

      I’m doubtful that would be possible now.

      • j1000000 says:

        Why don’t you think that’d be possible? Do Harvard grad schools take your high school GPA into account?

        • Plumber says:


          My guess is that it would be harder to get into the University of Washington today than then, so the opportunity to go to Harvard Law from there would be much less.

      • sandoratthezoo says:

        Two people in my high school class of 47 went to Harvard. I went to a nice private school and graduated (high school) in 1995.

        One of the two people who went there was my ex-girlfriend and I knew a lot about her. She was asian, her family was quite wealthy, and she was smart and extremely hard-working, following a sort of classic-for-the-90’s tiger-mom curriculum. She wasn’t brilliant, but was obviously able to excel in academics, and put in the work.

        The other I knew much less well. She was academically very solid, and I didn’t interact with her enough to know how much was raw intelligence and how much was hard work. She was also both black and hispanic, so that might’ve helped.

    • aristides says:

      Let’s not forget that Scott has admitted on many occasions that he is relatively bad at math. I was going to say that Scott strikes me as the kind of person who got a perfect score on his SAT verbal and in the 600s in math, but when I looked for the link, he essentially confirms it.

      SAT and GPA both favor well rounded students over specialized. The holistic look at the candidate might have helped Scott, if he won a national writing competition, the topics of which likely wouldn’t have suited Scott. But it would have also penalized him if he wasn’t a natural leader.

      None of this is to minimize Scott, I personally think he is one of the greatest writers of our generation. But it would be hard to show that in paper to Harvard.

    • drunkfish says:

      Evaluation criteria are a insane, as others say.

      I was admitted to Harvard for a PhD, which presumably indicates that I’m Harvard material, and my girlfriend at the end of high school was admitted to Yale and Princeton but rejected from Harvard, so I think I have a good sense of what their standards were. There was no chance in hell I would’ve been admitted to Harvard for undergrad, full stop. You don’t just need to be smart, to get into those schools for undergrad you need: perfect grades, to cure a reasonably common form of cancer, and a ton of luck. Having good grades, piddling extracurriculars, and a normal amount of luck, I knew not even to apply. I can easily picture Scott having a similar resume to me if he was even remotely relaxed in high school, despite him definitely being one of the most brilliant people I’ve ever encountered.

  62. Rachael says:

    That Common Application thing with no upper limit seems broken. What’s to stop every student applying to every college? Here in the UK we have a similar system called UCAS, but it only lets you apply to a maximum of five universities (down from six when I did it in the 90s). This cuts down on admin for everyone, and also means the universities have a more realistic idea of how many of their applicants are “serious”and this don’t have to be artificially over-selective.

    • vV_Vv says:

      What’s to stop every student applying to every college?

      What’s wrong with every student applying to every college? Let the colleges compete!

      This cuts down on admin for everyone, and also means the universities have a more realistic idea of how many of their applicants are “serious”and this don’t have to be artificially over-selective.

      Or they could just use a standardized criterion like SAT.

      • rlms says:

        What’s wrong with every student applying to every college? Let the colleges compete!

        Colleges would waste energy considering students for whom they are the bottom choice (and also students who are nowhere near good enough, although that’s less of a problem).

        Or they could just use a standardized criterion like SAT.

        UK universities do do that. Achieving the A-level requirements for a university basically guarantees you a place except for Oxbridge and medicine (and even there if you get the maximum score your chances of getting in are pretty good).

    • sohois says:

      Worth bearing in mind that the UK only has ~120 universities whereas the US numbers in the thousands, it would be practically impossible for a student to apply to even 10% of the universities in the US

    • FormerRanger says:

      What’s to stop every student applying to every college?

      Applying is not free. The Common App makes it less work but it still costs money.

      • aristides says:

        This is the main reason. Applications fees are large enough to cover the cost of screening the next marginal candidate. Average application fee is $77, and most applications can be screened out in minutes not hours. More than enough to cover the hourly salary of a screener.

    • Anthony says:

      But colleges are *incentivized* to be artifically over-selective.

      And since the Common Application thing is online, it’s easy to have the computer screen out any arbitrary percentage using any set of arbitrary criteria, so that your actual admissions workload isn’t much worse than before.

      • onyomi says:

        Back when I was applying for college almost twenty years ago I had the impression that if you used the common application for a selective university your chances were hurt relative to jumping through the hoops required to fill out a more bespoke application. Not sure if that was actually true (didn’t get in anywhere I sent a common application to, I seem to remember), but if it was/is it would make sense that the common application serves primarily as a kind of “applicant fishing” for colleges to artificially boos their numbers under the guise of lessening the burden of applying.

  63. BlindKungFuMaster says:

    My friend’s daughter got 3.85 GPA, had 5 AP classes in high school, was on competitive swimming team, volunteered 100+ hours, was active in school activities, yet she got rejected by all 4 UCs that she applied to. And these were not even the highest tier of UCs, not Berkeley. She did not apply for more schools and thought that UC San Diego and UC Santa Cruz were her safe choices. The whole family is devastated.

    It could be the case that college admissions became more random as kids apply more often and the metrics become more vague. I.e. there used to be x acceptable applicants of which 50% were somewhat randomly selected. Now there are 2x acceptable applicants of which 25% are somewhat randomly selected. Then there are just much more kids being unlucky and missing out on all their choices, though it hasn’t really been getting harder to get in. Of course individually the remedy is applying to even more colleges.

    • Clutzy says:

      Also those metrics are pretty useless these days. A 3.85 is basically available to anyone who shows up for all 180 days of school, 5 AP classes doesn’t mean she got a 4 or 5 on all 5 tests she took, and it tells us nothing of her SAT/ACT scores.

      Such a story is entirely compatible with the idea of there being little increase in the objective increase in difficulty for people to be admitted, just a shift in the number of applications per student, and perhaps a slight shift in who applies.

      UCSD’s average SAT is 1257 and UCSantaCruz’s is 1263. Neither is all that high. Indeed, the anecdata a lot of people have about parents/grandparents coasting through high school and getting in anyways is probably mostly driven by people being better/more objective at evaluating the IQ of their children and directing them properly. A 1257 SAT person should not expect to get into any elite college, and in the past they didn’t whether they slacked or tried hard in High School because people were more honest with them. Indeed, with that score you are a slightly above average college student (hell you’d have been below average in my high school’s honors math courses which was the worst school in the district).

      Much that has changed is parents’ perceptions of their children have become less logical. Apparently they now think a kid in the 80th percentile is special (and note that is the average at those schools, not the score to be in the bottom quintile). Lets say you got a 1250 on your SAT. You probably can go to school, but you are not really a gem to be coveted by the universities, so your application pool should be strongly favoring the cal state system over the UC system.

      • Conrad Honcho says:

        I agree with this. I think it’s telling that the family used GPA and number of AP courses as their metrics. What were the SAT and AP test scores?

        It’s like the kid who complains “How did I get a C?! I studied for 8 hours!” The test doesn’t measure how long you studied.

      • j1000000 says:

        Agreed. What does taking an AP Class even mean as a credential? It’s like bragging about taking an SAT prep course. Someone’s gotta get the 1’s and 2’s on the tests. (Well, I think someone does — I don’t know how scoring works)

        • Eric Rall says:

          AP classes are typically more rigorous than the non-AP equivalent class, covering more material, assigning more work, and grading to a higher standard. So a B in an AP History class might indicate at least as much diligence and mastery of the material as an A in a regular History class.

          When I was in high school (late 90s), there was also usually some selectivity applied to which students were permitted to sign up for AP classes, usually requiring you to get sorted into the “Honors” track (most courses had regular and honors versions, with honors being a moderately accelerated for above-average students in that subject) for a subject area (math, history, english, science, etc) and do well in it, with the end of the track splitting into an honors course and an AP course. This may have changed over the past twenty years or so, though.

          As a side note: many high schools (especially in California) have “weighted” GPAs, where courses intended to be particularly challenging get a bonus in the GPA calculation to adjust for the difficulty level. A common implementations is to add 0.5 points for an honors course and 1.0 points for an AP course, so an A in an AP course counts as 5.0, a B counts as a 4.0 (same as an A in a regular course), and so on. So if the 3.85 GPA is a weighted GPA for a student who took mostly honors and AP classes, that means the student took relatively difficult classes but got a B to B+ average in them. It it’s an unweighted GPA, it’s a bit more impressive, indicating an A- average in the same classes.

          • aristides says:

            Agreed. My weighted GPA was over 5.0. That doesn’t tell you anything beyond the fact that my high school inflated GPA more than this students. Class rank is a little better, but favors kids at schools with disadvantaged schools. I know plenty of students that took 5 AP classes and failed all of the tests . I even know one that took 9 AP classes and couldn’t pass a single test. Her Teachers didn’t want to fail anyone and she did all the homework, so she got an A or B In every class. My school just let you take whatever classes your parents asked, and she ended up in the top ten students GPA. It wasn’t a real surprise that that didn’t get her into any of my states universities.

    • sandoratthezoo says:

      Anecdotes are also just unreliable. She might’ve had a 95% chance of getting admitted to UCSD and UCSC. One in 400 such students will fail at both. There are definitely enough students out there to make there be quite a few comparable stories each year.

      • Michael Watts says:

        She might’ve had a 95% chance of getting admitted to UCSD and UCSC. One in 400 such students will fail at both.

        This isn’t the case unless UC admissions have radically changed recently. You are not admitted to different campuses independently; you apply to the UC system once, marking the campuses for which you wish to be considered in particular. If you aren’t admitted to any of the ones you asked for, you may be offered a place at a different one.

        So I see absolutely no reason to believe that, given a 5% chance of rejection by UCSD and a 5% chance of rejection by UCSC, your overall chance of being rejected by both is 0.25%. The admissions process explicitly takes account of the full set of schools to which you are applying; they might adjust that joint probability to anywhere in the range (0%,5%].

        (Side anecdote: I matriculated to UCSC, my local campus, in 2004. I had a classmate who applied there and was rejected as not meeting standards, likely related to a 7th-percentile SAT II score. She filed an appeal, presented the argument “but I really want to go to college”, and was admitted. This did not make me think of UCSC as being especially selective.)

    • moridinamael says:

      Yeah. 3.85 is now effectively “below average” because of grade inflation, at least in AP classes. Taking 5 AP classes is meaningless, colleges don’t care about this. In fact, it means you’re probably paying them less by testing out of courses. Being “active in school activities” is literally nothing. This example student is a significantly below-par sample. I say this as someone who was themselves a below-par sample.

  64. BlindKungFuMaster says:

    Part of what inspired me to write this post was listening to a famous scientist (can’t remember who) opine that back when he was a student in the 1940s, he kind of wandered into science, found a good position at a good lab, worked up the ranks to become a lab director, and ended up making great discoveries. He noted that this was unthinkable today – you have to be super-passionate to get into science grad school, and once you’re in you have to churn out grant proposals and be the best of the best to have any shot at one day having a lab of your own.

    I get that vibe from a lot of famous scientists talking about their career. No wonder you can’t remember who it was specifically in this case.

    • Robert Jones says:

      I doubt the “unthinkable today” part of the story. My anecdata suggest that exceptions often get made for sufficiently brilliant people.

      • Conrad Honcho says:

        I’m not sure the scientist in Scott’s anecdote was especially brilliant. Sounds like he just put in some good ol’ elbow grease sciencin’ and worked his way up the ladder.

        Let that be a lesson for you kids. Don’t waste your time with all that fancy collegin’, just walk into the lab, shoulders back, head held high, look the professor square in the eye, give him a firm handshake and say “I’m the man to unify special relativity and quantum mechanics.”

        • Worley says:

          Hahaha! It also helps, when someone tells the anecdote fifty years later, if you actually did then go on to do the unification … I mean, if Linus Pauling talks about doing fine despite slacking, I suspect it means “If I’d actually worked hard, I probably would have gotten another Nobel.” Doesn’t quite apply to us normals.

        • nadbor says:

          You mean *general* relativity and quantum mechanics. Special relativity and quantum mechanics were successfully unified in the 1920s ad 30s. Quantum Field Theory is the result.

      • The Nybbler says:

        Sufficiently brilliant and socially ept, anyway.

  65. spork says:

    I think the scramble for competitive college spots is more intense partly because of the growing recognition that college admissions sorting may be the only truly objective overall assessment of your potential as a human being that Americans will undergo. America is the land of grade inflation and trophies for everyone, but when it comes to college admissions, it’s a merciless war of all against all. Where you land after this sorting is actually pretty informative, and unlike your SAT scores or IQ tests, it’s something that employers are free to use to evaluate you.

    For now there is no sign that a one-rung promotion actually helps you later in life, but that might be misleading. For one thing, it’s based on data from old people. (Only they have measurable “outcomes.”) When HR is is full of millennials who spent their youth checking boxes in order to impress admissions boards, I have a feeling that they might get more judgmental about finer differences in how high a college bar the job applicant cleared. If landing in a fancy college is taken to be only job of today’s high school student, it sets up a self-reinforcing cycle. College prestige may not make a difference now, but if we think it does, it soon will. I’m stunned by how much detail high school juniors know by heart about the selectivity of the various colleges. They won’t forget this when they move into HR jobs. If they judge their self-worth by their admissions letters, how could they not judge others by the same?

    A second reason why the outcome studies might mislead is because the only methodologically sound ones study differences between the narrowly accepted and the narrowly rejected. But if sorting is as efficient as the article suggests, maybe the narrowly rejected land only one rung lower, and the prestige difference between adjacent rungs is small enough to be statistically negligible without huge sample sizes. Slacking off in high school or blowing off the mandatory virtue signaling you need on your application doesn’t just cost you one rung, though. It might cost you enough rungs to make a real difference to your life. So these are two arguments for why you actually might want to freak out about landing in a good college spot, despite the research.

    • AG says:

      Has the Asian system of entrance examming even earlier in life been provably better, though? Theoretically, it allows people to not spend so much of their life in education, because the sorting occurs sooner, but practically, it means that people lose their childhoods to test prep.

      • spork says:

        We live in a country with kindergartens that reject children based on an admissions test. Then there are three different levels of the SSAT, the last of which is given to 8th graders applying to competitive high schools. These applications are basically age-adjusted versions of college apps. So if you are an American kid playing the game “right,” you might very well spend your entire childhood preparing for tests and applying for things, and college is your first break from curating your life to look impressive to some admissions committee… unless you’re preparing for medical school.

    • Michael Watts says:

      A second reason why the outcome studies might mislead is because the only methodologically sound ones study differences between the narrowly accepted and the narrowly rejected. But if sorting is as efficient as the article suggests, maybe the narrowly rejected land only one rung lower, and the prestige difference between adjacent rungs is small enough to be statistically negligible without huge sample sizes.

      This isn’t correct; there are also studies that compare those who were accepted and who subsequently matriculated with those were also accepted, but chose to go somewhere else. Your criticism still applies to that scenario, but your description of “the only methodologically sound” approach is wildly off base.

      At a slightly less rigorous level of analysis, though, those who are accepted and choose not to attend do not tend to matriculate to schools only slightly lower in the prestige hierarchy. Why would they? If they were aiming for prestige, they’d go to the top school that admitted them. They tend to matriculate to schools that are cheap or close to family.

    • aristides says:

      I’m a millennial that graduated from a high tier university and work in HR, and I have noticed the opposite trend. It’s the Baby boomers and gen Xers that really believe that where you go to college matters. The director specifically ordered that my position goes to the person with the best college degree. Millennials don’t care where I went to college and just treat me normally.

      My hypothesis is that baby boomers and gen Xers were the driving force to telling millennials that if they went to the best college possible their life would be set, and still believe it. Millennials on the other hand are disillusioned, either because they couldn’t get in and are resentful or because they did get in and realized they learned nothing but have a mountain of debt. When I read someone’s resume and it has a top tier school I just assume that their parents helped them get in it or pay for it, and when it has a public university on it, I just assume they were cost and debt conscience and made the wise financial decision. And when I’m hiring an hr specialist I would rather take the HR assistant that has 4 years experience than the Harvard alumn with an English degree Ceteris Peribus. But I don’t make final hiring decisions yet.

      • Controls Freak says:

        I echo a lot of this. I went to different schools for undergrad/grad, and since I liked teaching, I worked with the undergrad program in both of them. The “better” school had a much worse program. Adding in the variety of stories I’ve heard over the years about how so-and-so chose such-and-such a program (and meeting both dumb/smart people from both good/bad schools), I pretty much completely discount school-specific credentials. You can roughly weigh bachelors/masters/phd-level and sometimes adjust for “came out of so-and-so’s lab”, but I don’t really have much of an opinion on whether you’re good or not until I just have a conversation with you and can figure out whether you’re actually smart.

      • spork says:

        That’s really interesting and encouraging. I really didn’t want my “you should freak out about where you go to college” argument to be right, because there is much better stuff to worry about for high school students. If your attitude is typical of millennials, then all the freaking out is really serving no point.

  66. prunesquallor says:

    I believe UCs are more competitive because of cost. Personally, I got into a number of strong private schools, and chose a UC because it is significantly cheaper. In the past, this would not have been as much of a factor, because college was affordable.

    I’m not sure if this applies to public schools in other states. The UCs are the best public schools in the country and are able to compete with high level private schools

    • sandoratthezoo says:

      This is a really good idea for an explanation, that I had not thought of. Are there any comparably low-cost colleges besides the UCs that can compete with at least tier-2 private schools that we could use to test the hypothesis?

      • gbdub says:

        Michigan and Georgia Tech are two perennially top-ten public universities comparable in price and prestige to the best UCs, and both are on that list of top 20 biggest decrease in admission rate.

        Other top ten publics that would be good comparisons are Virginia and UNC.

      • Steve Sailer says:

        Rice U. traditionally set a lower list price than other private colleges because its endowment was rich due to Rice being privy to inside information on the great East Texas oil discovery of c. 1930.

        I believe when I started there in 1976, tuition was $2,300 versus maybe $3,700 at most comparable private colleges. (Note: my parents took care of the money for me, so these numbers are very hazy in my memory.)

        I don’t know if that did Rice much good, though. In contrast, George Washington U. in D.C. followed a policy of trying to have the highest tuition in the country by a few dollars, which may have helped attract positive attention.

      • The Nybbler says:

        There are a fair number of low-cost colleges which have programs which can compete with tier-2 private schools, even if the school as a whole emphatically cannot. I want to school at the University of Maryland, College Park, which has remained fairly consistently near #20 for computer science, for instance (looks like it’s currently tied for #16 with UCSD and Harvard)

        The particular program rankings are for the graduate school but they seem to be used as proxies for undergraduate as well. For undergraduate engineering as a whole, there’s quite a few state schools in the top 20. Some are pretty expensive for out-of-state students.

  67. kaakitwitaasota says:

    Dale and Krueger(2011) examine this question, using lifetime earnings as a dependent variable.

    In general, they find no advantage from attending more selective colleges. Although Harvard students earn much more than University of Podunk students, this is entirely explained by Harvard only accepting the highest-ability people. Conditional on a given level of ability, people do not earn more money by going to more selective colleges.

    This may well be true, but it isn’t perceived as being true. Caitlin Flanagan’s excellent article in The Atlantic about the college-admissions scandal recently reveals that among the very, very rich what’s really being purchased is respectability:

    We are in the home of Gordon Caplan and his wife, Amy. Gordon was—until placed on “leave” post-indictment—the co-chairman of a New York–based global law firm, where he was a partner in the private-equity group. Amy is the heiress daughter of the late telecommunications magnate Richard Treibick. He also lived in Greenwich, summering in the Hamptons in a 32-acre spread in Sagaponack that included a seven-bedroom house on the dunes with a pool overlooking the ocean, which his family sold shortly after his death in 2014 for a reported $35 million.

    Now if I were worth a middling eight figures, I wouldn’t care where my kid goes to college. They can get an excellent education at a good Midwestern flagship state school–I certainly did–and when they get out they will have an eight-figure trust fund carefully set up to yield them a six-figure-a-year income from the interest from the rest of their lives. It’s one thing if you’re only middling upper-middle-class and your dream is to work on Wall Street (for which you’d better have an Ivy League degree), but if you’re worth $35 million-plus you’d think you wouldn’t care as much. But in reality it is precisely the rentier class, who are not anywhere near having to worry about putting food on the table or even having to work, who are most paranoid about getting their kids into Harvard at all costs.

    Anecdotally, I can bring two stories of my own to reinforce this. I went to a very good prep school in the Lexus-infested suburbs of a middlingly large American city that usually sent a couple of kids to Harvard or Yale every year. I knew from early on that I didn’t want to end up at an Ivy–my SATs were in the mid-2300s (back before the scoring was changed), but I was pathologically lazy into senior year, and decided early on that going to an Ivy amounted to taking out six figures’ worth of debt to develop crippling imposter syndrome and probably a drinking problem. I ended up at a large Midwestern state school and received an excellent education with nearly no debt.

    My college counselor stood by me through all of this, and was a consistently great guy to work with. A couple years after I graduated, however, the graduating class sent nobody to Harvard or Yale. Georgetown, Chicago, Berkeley, Columbia, UVA, Cornell, yes–but not Harvard or Yale. So he was kicked downstairs back to the English department.

    Also, a lot of top firms these days won’t even look at you if you didn’t go to the “right” college. My mother did her MBA at Northeastern, and recently had lunch with an old classmate who ended up at a top consulting firm. My mother’s classmate’s résumé would end up in the trash unread these days–Northeastern isn’t considered good enough.

    So while it’s probably true on the macro level that smart kids will do just fine anywhere they end up, there is a subset of extremely prestigious, extremely well-paid jobs which will not even look at you if you didn’t get into the right institution at the age of 18–which, in practice, means that the élite are chosen on the basis of who they were at the age of 14-17. When viewed in those terms, it’s completely nuts.

    If there’s a fix for this, it’s probably to make Harvard and Yale massively increase admissions rates (perhaps by threatening them with a high endowment tax if they don’t). Someone has proposed having Harvard/Yale/etc. auction off ten percent of all seats, and then choose the rest by lottery from all applicants who meet a particular baseline (perhaps mostly using a college entrance exam to squash the extracurricular arms race); this would make explicit what has already been going on (people bribing their kids’ way into top colleges) and, over time, decrease the value of a Harvard degree relative to those of good state or private schools (how can you tell if your résumé with a Harvard degree on it is from a prodigy, or just from a member of the idle rich whose father dropped $2 million on an admissions spot?)

    • Deiseach says:

      If there’s a fix for this, it’s probably to make Harvard and Yale massively increase admissions rates

      That’s not a fix. They can admit six thousand first years to every course, then have 99% of them drop out after failing the end of year first year exams, the re-sits, and the re-sits of the re-sits. The end product will still be the tiny number who made it all the way to graduation, which will be the normal size graduating class they had all along.

      If you want to bump up the numbers graduating, you’ve got a tough choice. Take every suitable student who could realistically graduate, and that’s got its own costs. Or dumb down the courses so what is considered “sufficient” numbers – of the right demographic mixture – graduate.

      Then the employers will want the honours degrees versus the bare pass, or higher levels. Now you need a PhD where an MA used to do, and an MA where the BSc or BA was enough.

      It’s one thing if you’re only middling upper-middle-class and your dream is to work on Wall Street (for which you’d better have an Ivy League degree), but if you’re worth $35 million-plus you’d think you wouldn’t care as much.

      For people at that level, it’s not the “education to get the degree that will get your foot in the door of a high-flying firm”, it’s the connections. Sure, you’ll get a nice degree at the Midnowhere college, but you’ll miss out on all the networking with the kids of the important people in those big firms and political circles and the likes of it. When young George goes to that big firm for his sinecure job, he’ll be as out of place with his non-Ivy degree as Jose the janitor’s son because neither of them will have mixed with, partied with and been in the same fraternity as the other gilded elite kids.

      At that level, forming connections and “who you know” who can pull strings and do favours for you in later working life is vitally important.

      • johan_larson says:

        I don’t know. It seems to me the top schools like Harvard turn down hefty numbers of students who would do just fine academically if admitted and who would do good work after college too. But clearly that can’t go on indefinitely. With ever larger classes, at some point the character of the students would change. The question is how much they could scale out before it would become evident.

        Anyone want to take a guess at a multiple?

        • Mizike says:

          I have a lot of experience with MIT admission.s The party line for admissions officers most of these schools is usually a number in the 2-3 range before the general awesomeness level starts to noticeably decrease, but like 5-10 before you get to people who can’t handle basic coursework.

        • It seems to me the top schools like Harvard turn down hefty numbers of students who would do just fine academically if admitted and who would do good work after college too.

          What does “do just fine academically” mean? Harvard could pitch their classes at a level that would be fine for many more students than they admit. They could pitch them at a level that was suited for fewer applicants than they admit. The more academically selective they are, the higher the level at which they can pitch them.

          When I was there, and I think still, there were two math sequences. Math 11 followed by Math 55 was for students particularly good at math. A different sequence, I think three years starting with math 1, was for most students. I believe a similar pattern existed in physics and chemistry. I don’t know about other fields.

          They could have offered only the 11/55 sequence–and a majority of the students would have failed to do just fine academically. They could have admitted more students at the low end of the range and fewer at the high end, and the additional students would probably not have flunked out, but the average quality of the education would have been lower.

      • Lasagna says:

        I think that the networking effects of elite schools are overrated. I went to one, but my important networking comes from my career connections, not my school connections (the degree helps in other, equally important ways).

        I suspect that important people/political circles networking you’re talking about is dependent more on who the parents are than where the kids go to school. I rubbed elbows with the children of the rich and powerful at school, I guess, but they hung out and “networked” with each other, not me and mine.

        • I went to Harvard, a very long time ago. I have only one current friend whom I met at Harvard, and I don’t think his friendship has ever been professionally useful to me.

          Of course, that is long ago and a sample size of one.

          On the issue of multiple applications … . I applied to five schools–two top universities, two top liberal arts colleges, and one somewhat lower liberal arts college as a safety school. That was in 1960. I don’t remember any impression that it was an unusually large (or small) number.

      • Alex Zavoluk says:

        Those schools get way more applicants who could graduate than they can admit. Getting in is much harder than getting out.

        edit: The reputation of being exclusive is probably part of the value of going to these schools. That effect weakens as you admit more people.

    • Worley says:

      As showed up even in the Boston Globe, there’s a lot of snobbery in well-off suburbs (not just among the actually rich). It’s worth $100,000 to the parents to get their kid into Harvard/Stanford/etc., even if it doesn’t matter that much to the kids.

      What I’ve not seen explained is paying large bribes to get into schools that aren’t in that top 20 or so. I mean, UCSD? I’m sure you can get a fine education there, especially if you work hard, but you’re not going to buffalo future employers/investors/whatever by simply having the degree.

      • Ketil says:

        What I’ve not seen explained is paying large bribes to get into schools that aren’t in that top 20 or so. I mean, UCSD?

        I guess it is a trade-off between the value of $100K to you, and your options among respectable (UCSD is still respectable, right?) colleges with corrupt officials, and the limitations of your little prodigy’s mental faculties — that is, his or her ability to get into any honest education at all.

  68. fnord says:

    The Ivy League probably has some kind of unofficial Asian quota which has been pretty stable over the past decade. Although the Asian population continues to grow, and their academic achievement continues to increase, this probably just increases intra-Asian competition rather than affecting people of other races.

    I mean, this would still show up as some amount of increased selectivity in the overall admissions numbers, right?

  69. DeWitt says:

    My parents’ generation tells stories of slacking off in high school, not worrying about it too much, and knowing they’d get into a good college anyway.

    Your parents’ generation almost certainly has a terrible memory about what life was like back when they were teenagers. Human memory in general is unreliable when looking so far a distance back, and anyone talking about events several decades ago had best show some real humility.

    • Tarpitz says:

      I handed in precisely four pieces of written work in my entire upper sixth (senior-equivalent) year. I routinely didn’t even turn up to school until after morning break to allow my hangovers to abate. I got into Merton, which at the time (2001) was the most academically competitive college in Oxford. This was possible because no ongoing measure of my performance mattered: I just had to perform well in exams and interviews. Has there been any procedural change in admissions which might favour conscientiousness over raw ability? Greater weighting of GPA as compared to SATs?

      • DeWitt says:

        To clarify, I do believe that things have gotten more competitive. People’s memories are just not the way to figure this out, and should always be taken eith the tiniest grain of salt when they are brought up as evidence.

    • Anthony says:

      I graduated from an East Bay Catholic school in 1984. We did have a significant case of grade inflation, but I really was slacking in a lot of classes, because I didn’t need to do much work to get As in most of my classes. (Though I didn’t slack nearly as much as Tarpitz.)

      A GPA a little above 3.5 and a 1540 SAT (and 2230 on the three “achievement” tests from ETS) got me in to Berkeley, where I suffered from not having good study and work habits.

      I doubt my daughters could do the same and still get into Berkeley.

      • Tarpitz says:

        Yes, to be clear I imagine the college regrets not having a better way to detect the slacking. I did well in my first year exams, but then spent four and a half years doing essentially no work at all before finally dropping out. Selecting for conscientiousness or general functionality is not necessarily a bad thing…

  70. HALtheWise says:

    One very powerful contributor that Scott did not mention is that in many cases schools are directly or indically intentivized to have a low admission rate. US news & world report released the first national college ranking in 1983, and donors and board members at various schools have increasingly been using national rankings performance, which directly includes low admissions rates, as a measure of how well a school is doing.
    These rankings and metrics also heavily incentivize having high yield (a large fraction of students that are admitted end up attending) which for a fixed size applicant pool also encourages accepting as few people as possible. This has led to the death of safety schools, because they would rather reject a high performing student than admit them and have them not attend.
    These factors might also be a driving force behind the rise of common app, since schools are trying to get as many applicants as possible, even if it hurts the quality of their pool.

    • Jon S says:

      +1 Possibly apocryphal, but this definitely leads to a perception that safety schools may reject overqualified candidates.

      • edmundgennings says:

        One college, that I was interested in because it was a good social fit, would clearly reject anyone from my highschool over certain a GPA/SATs. There were a number of data points for this on Naviance.

    • Erusian says:

      Anecdote time: A friend of mine who’s a reasonably successful entrepreneur with a strong educational background and who has won several research grants and awards related to feminism was told she wouldn’t be accepted to Harvard because it was worth more to Harvard to have rejected her.

      Specifically, entrepreneurship is risky so why should Harvard bear that risk? If she was successful they would point out they rejected as a sign of how high their standards were. If she wasn’t, no one would care about her anyway. (The fact she was able to get access to admissions officers who told her this shows something about how well connected she was too.)

    • Wency says:

      I lived in WA for HS, and UW was the only public school I applied to. At the time, UW had a helpful table that flat out told you your odds of being accepted based on your SATs and GPA. If you’re X smart you have a 99% chance of getting in (I imagine the 1% is to allow for those whose essay tells the admissions staff to go kill themselves). If you’re less smart, the rest of your application needs to make it up to some degree. And below some threshold, you have basically no chance. There was no safety school punishment factor.

      I am curious if this table still exists. But I strongly approved of it then and now. Public schools are supposed to serve the people rather than their own power as institutions, so why not provide some transparency in service to the people?

      That said, I did have a sense that you wanted to communicate to your safety schools that you took them seriously and might attend. It might not hurt to visit your favorite safety school in person, have some recorded correspondence with them, and make sure your essay indicates that you see some unique charm with their school, or you want to stay close to home for college, or whatever.

    • aristides says:

      Related, the key to getting into a safety school is to write a very personalized essay that stresses why that school specifically is a good fit for you. In the days of a common app, personalized essays are a good costly signal that you will attend. I spoke to an admissions official at a low rank law school, and whether you mention the name of their school was the first thing they look at in an essay. Several applicants listed the wrong school and got auto rejected, and one applicants essay was two words, “180 LSAT” and got rejected by them and admitted to Harvard.

    • Worley says:

      This incentivization is quite strong. I’ve kept in fairly good touch with my alma mater, and its officers note that the USN&WR rankings are considered to be garbage by everybody in higher education — except prospective students and their parents. When my a.m. slipped from one category down to the next, the professors reported a noticeably less good freshman class the next year. (Confirmation bias?) In any case, schools have worked hard to raise their rankings, including by methods that could be considered gaming the rankings. It’s commonly admitted that trying to get more applicants so you can reduce your admission rate is one of the popular methods.

      • The Nybbler says:

        the USN&WR rankings are considered to be garbage by everybody in higher education

        They aren’t the people who matter. Nor are the students and parents. It’s the prospective employers who matter. I’d expect a system designed by people in higher education to lead to a really terrible feedback spiral (see “US Education, primary and secondary”)

        • faoiseam says:

          USN&WR rankings are considered to be garbage by everybody in higher education

          Trustees at colleges know that the rankings are garbage, and do what they must in order to stay high in the rankings. Just because you know they are garbage does not mean you don’t feel obliged to submit.

        • John Schilling says:

          How can employers matter, when the arrow of time denies them any useful input? Sure, maybe Alice never gets her dream job because she went to State rather than Harvard and employers in her field really care about that, or maybe Bob winds up $100K in debt because he thought employers cared about Harvard when they would have been fine with state. But that’s history. Alice went to State and Bob went to Harvard and the fact that their respective employers would have preferred the opposite, can’t change that.

          The opinions of students, parents, and guidance counselors matter, because those drive decisions about what schools students will apply to and attend. Those opinions may be driven by their belief regarding the preferences of employers, but there’s not much in the way of feedback to keep that belief grounded in reality. If they believe employers care about “good schools”, they’re almost certainly going to use the USN&WR definition of “good schools” rather than call prospective employers and ask what they really think about the schools their precious children are planning to attend, and if they get it wrong they’ll probably never know why.

          • The Nybbler says:

            I think we only disagree in whether there’s effective feedback to keep that belief grounded in reality. I think if the USN&WR and employer preferences were to significantly diverge, parents would start finding out about it fairly quickly.

            (though it’s very possible that employer’s preferences aren’t primary either and that employers, parents, students, and schools are all simply following USN≀ the ratings themselves are primary. Which would be depressing if true)

          • John Schilling says:

            I don’t see how that feedback is going to work, without the sort of broad statistical information that no parent is going to have and that I have not seen anyone else collecting and analyzing on behalf of parents. If Junior goes to the USN&WR #10 school in his field and doesn’t get a good job, is that because employers are recruiting only from the Top Five, or is it because Junior’s school is really ranked #20 on the (informal) Secret Ranking of Really Good Schools used by employers in the field? Unknowable without a Scott Alexander level deep dive into employment in that particular field, and absent that the presumption will be that Junior should have studied harder to get into a school with an even higher USN&WR ranking.

            Law schools are probably an exception, due to the extensive focus on BigLaw’s hiring practices, and perhaps a few others, but for generic undergraduate education I’m skeptical.

          • The Nybbler says:

            Many of the parents looking to get their kids into the prestigious schools are going to be working at the employers that hire undergraduates based on school prestige. That’s going to be one place the feedback comes from. And in some fields, it’s going to be fairly well-known even outside the industry, because the industry is high-visibility enough.

  71. sharper13 says:

    Two additional data points to consider for potential explanatory value in your analysis:
    1. Accreditation wasn’t really a thing until after the 1880s. It wasn’t national and standardized until 1912. Related changes may play into the transformation from accepting everyone who can qualify to competing for admission.
    2. Affirmative Action admissions in the UC System were substantially limited as of 1998 via Prop 209. Moving lesser qualified applicants to lower tier schools and no longer restricting the number of higher qualified applicants might contribute to an increase in the overall competitiveness level. i.e. more depends on your grades and test scores and less on biased admissions policies disguised under the “roundness” or personal background of the applicant. This change also increased graduation rates with better accepted applicant to school matching.

    • BBA says:

      Accreditation wasn’t that important until the ’40s and the GI Bill. The first version of the GI Bill only specified that aid would go to veterans at degree-granting institutions authorized under state law. Problem was, at the time many states would give degree-granting authority to just about anyone who asked for it, and ongoing monitoring was minimal. So after the spate of waste, fraud, and abuse that’s needed to change any government program, the GI Bill was amended to specify that the college had to be accredited by a recognized accreditation board, as well as authorized under state law. Before then the accreditation boards were (as I understand it) mainly about cross-recognition of course credits, but with federal student aid becoming so essential to the system, now the accreditor is the primary regulator of universities, and losing accreditation is practically the death penalty. Yet they’re still nominally private, voluntary organizations.

      What’s weird is, there are six regional boards that are the primary accreditors in their parts of the country, plus a few national boards that specialize in particular types of institutions, but no overall national standard. The one you linked has the impressive name of “Accrediting Council for Independent Colleges and Schools” but it was known mainly for accrediting online and correspondence schools and other low-prestige institutions. A couple years ago two for-profit ACICS schools (Corinthian and ITT Tech) went bankrupt after years of unethical practices that ACICS ignored entirely, and the Obama administration took the extraordinary step of revoking its recognition of ACICS as an accrediting board. This has since been reversed by the Trump administration, but most ACICS schools have left for other accreditors in the meantime.

      • sharper13 says:

        Yeah, I wasn’t holding up ACICS as an example of accrediting bodies, their page just had the first actual history timeline of the first few results in my search results. No endorsement was meant. 🙂

        In addition to your points, I think over time accreditation has turned more and more into standardization and control of how Universities operate. You can’t have a law school without a big law library, even if your students no longer need it because of electronic books, you can’t have too inexpensive of education if you must maintain certain student/teacher ratios, etc… This also limits innovation and change, even as new accrediting bodies (like ACIS) either work out or don’t (again, like ACIS).

  72. Erusian says:

    Only about 8% of Congresspeople went to Ivy League colleges, which feels shockingly low considering how elite they are in other ways. The trend is going up among Democrats but not Republicans. There is obviously a 40-50 year delay here and it will be a long time before we know how likely today’s college students are to get elected to Congress. But overall this looks encouraging.

    Congressional elections can be more meritocratic than people imagine. If you can raise money and win votes, you’re in and there’s only a limited number of options (mostly very extreme/hostile) the party can do to you. There are a lot of people in the House of Representatives that are basically just local elites from a district. Off the top of my head, I can think of a man who wins his seat basically off of his connections to powerful garbage unions (D), another who owns a chain of very successful car dealerships and is well respected in his community (D), another who built a series of hotels and real estate investments in his community (R), and one who entrepreneurially organized a certain ethnic and religious group to dominate his district (R).

    The government, when it gets to appoint people, goes in deep for credentialism. But elections are basically meritocratic. If you can get enough votes, you win. Full stop. And once you’re in power, it comes down to how good you are at gaining, holding, and using power. And that can be intensely personal: Paul Ryan’s district is not important. Neither is Kentucky an electoral powerhouse.

    6. All the hand-wringing about getting into good colleges is probably a waste of time, unless you are from a disadvantaged background. For most people, admission to a more selective college does not translate into a more lucrative career or a higher chance of admission to postgraduate education. There may be isolated exceptions at the very top, like for Supreme Court justices.

    I’d like to point out another angle. Americas elites are still heavily regional. New York society vs Boston society vs the Great Lakes industrialists are still distinct groups with distinct patterns of norms and behaviors. Even in a city like DC, the people who run large companies are a distinct set from the governmen types. If you are elite, attending an elite college can teach you the norms of the national, governing elite and give you the social currency to be universally identified as elite.

    How many people know what Danaher is? Or Andreesen Horowitz? Well if you’re in DC or the Bay (respectively), you probably do. Otherwise, probably not. As an American elite, your power base is almost certainly sectional (unless you’re a hyperelite like Bezos or Zuckerburg). Having a Harvard degree wins you acknowledgement everywhere which greatly expands your options. J.D. Vance could go from New England to California to Ohio and, in all places, be recognized as elite because of his Yale degree. I doubt the rich people his parents despised, though undoubtedly local elites, could do the same.

    • Tarpitz says:

      How important is it to be generically recognized as elite, versus having specific elite contacts? And is it possible that even the Iviese (and Oxbridge, over here) are too large to really dramatically increase your chances of making friends who can seriously help your elite career? I’ve long suspected that for this reason the very top private schools are spectacular value for money and all the others a massive waste: send your kids to the Dragon and Eton, or send them to the local comp.

      • Erusian says:

        One of the big differences between Britain/France/Poland/Germany/whatever and the United States is the sheer size of the US. Specific contacts are very powerful on both sides of the pond but in Britain you can have almost universal bonds. The population and the elite is just smaller and more homogenous. The geography is smaller too, making it easier to cross-pollinate through social events. In Britain, it’s just possible that the elite network of an Etonian into Oxbridge type touches almost everyone.

        In the US, in conrast, you will meet wealthy, powerful, influential people who you have no links to. The US has roughly the equivalent to six Londons and five Oxbridges. And that’s not including areas of the country that are relatively unintegrated into those elite networks and have their own. The landowners of southern Indiana have had their habits scrutinized lately because of Mike Pence. But even before that, they were a local elite that had relatively little relationship to the national.

        So when you’re meeting someone who’s wealthy, powerful, and locally influential, and you don’t have a friend of a friend who knows them, how do you get them to acknowledge you as a fellow elite? An elite education is one of the few things that’s recognized basically everywhere as an elite status marker.

        How important is that? Well, it gets your foot in the door. Maybe not with the warm recommendation of someone they know (which is better). But that’s a lot better than most people get.

        • Tarpitz says:

          That speaks to why an American would be more likely to have fellow elites to whom they had no connection, but not to the question of whether that matters very much compared to the positive existence of elites to whom they have strong connections. It seems to me that most of the value of elite contacts might well lie in the absolute number and strength of those contacts, far more than the proportion of the elite to which one is somewhat connected.

          • Erusian says:

            Because elites can’t entirely isolate themselves in their home environments in that situation. Imagine you’re a Texas oil baron and you travel to (say) Chicago. Chicago has an entirely different elite structure where ‘being in oil’ is irrelevant. So you need something that commonly marks you out as elite.

    • Michael Watts says:

      If you are elite, attending an elite college can teach you the norms of the national, governing elite and give you the social currency to be universally identified as elite.

      The excellent book Pre-Industrial Societies: Anatomy of the Pre-Modern World describes this as the entire purpose of formal education in every pre-modern society. The point of education is that (1) it makes you recognizable as a member of the elite; and (2) elite society is geographically widespread (e.g. the peasants in Sicily and the peasants in Sparta had radically different cultures, but the elites had minorly different cultures — they all knew the Iliad). Point (2) actually makes education necessary in order to realize point (1).

      For Europe, the education that made you elite consisted of knowledge of Latin (this wasn’t credentialed — you knew it or you didn’t — and this meant that non-elite church officials could occasionally embarrass elite lords, since church education also taught Latin), so I think Harvard’s original entrance requirement there is less generous than Scott wants it to be. The accessibility of Latin to peasants in Massachusetts is an aspirational effort on the part of Massachusetts, an effect of the idea that people who deserve respect know Latin, not a cause of the entrance requirement.

      (Also, I was surprised how lopsided the entrance requirements were between Latin and Greek. Knowing all the noun and verb paradigms is material that will fit on a few pages. Composing fluent poetry in the native style is, let us say, a higher bar than that.)

      Megan McArdle has written speculating that the modern pressure to attend an elite college comes from the fact that, in the past, when the pressure was lower, many Americans of middle-to-high status were that way through owning a local business such as a car dealership, and could easily provide a career/livelihood to their own children. If your children are going to move somewhere remote, like New York, and obtain jobs unrelated to you, then they need to be accepted into that remote culture, and college is how that is done. So the failure of the “successful families operate local-level family businesses through multiple generations” model is what leads to the pressure.

      I perceive a good amount of support for McArdle’s theory in your comment and in Scott’s post.

      • Erusian says:

        The excellent book Pre-Industrial Societies: Anatomy of the Pre-Modern World describes this as the entire purpose of formal education in every pre-modern society.

        Definitely going to read that book. I disagree it’s the entire purpose. But it’s certainly one of them. I’d say education generally has the purpose of creating and reinforcing social structure and bonds. And that includes marking people out as elite or non-elite. It also includes inculcating ideas of what a citizen is, what knowledge society deems socially valuable, etc.

        Additionally, at its best, education inculcates real and useful skills to the population. But this is by no means necessary. One thing that has been written on is how the Chinese exam system eventually shifted from examining people on practical skills to plain knowledge of classics and other such pure humanities (in our terms). And how this left the Chinese elite increasingly unable to interpret the law, understand science, or have any practical skills beyond occupying a privileged position in government. Yet I think it has to be admitted the Chinese elite did a significantly better job in preserving Chinese culture, a sense of Chinese ethnicity/unity, and Chinese society than many other elites.

        Megan McArdle has written speculating that the modern pressure to attend an elite college comes from the fact that, in the past, when the pressure was lower, many Americans of middle-to-high status were that way through owning a local business such as a car dealership, and could easily provide a career/livelihood to their own children.

        As for McArdle, that may be true to some extent. But I think the primary motivator is a little more cultural: elites have started to jockey with each other for supremacy of late. My impression is it just used to be accepted that people like Lincoln (who was from a frontier farming region and loved to tell stories and had a local accent) could be local elites and then make it to the national stage from that platform. But now there are elites who consider themselves to reside as national elites, even if really they’re just the local elite of Hollywood or DC or New York.

        One thing I think gets lost in the shuffle of other, more legitimate reasons Cooper dislikes Trump: Anderson Cooper is a Vanderbilt, one of the old moneyed families of New York and part of New York’s high society. Trump… is not, to say the least. Likewise, I can say I’ve heard in certain cities (being very vague here for identity reasons) people sneering with contempt at people who might be very wealthy and successful but are from the wrong schools, the wrong part of the country, the wrong industry.

        The issue is that this requires them to play ‘eliter than thou’. And this in turn leads to competition. Would a person who inherits thirty million dollars and has a giant social media following and gets handed acting roles be elite regardless of whether she went to Harvard? Yes, absolutely. But if she can’t get something like a Harvard admission, something people even in (say) Chicago’s steel industry want, she has no claim to being more elite than the wealthy son of a steel mill owner. Which makes lecturing him about how Hollywood values are superior awkward.

        I think Harvard’s original entrance requirement there is less generous than Scott wants it to be.

        Also, one reason Harvard was so open originally: it trained clergy. If you wanted to be a wealthy/important non-clergyman, attending Harvard was no great advantage in the 17th century. It didn’t stop primarily being a seminary until basically the American Revolution (which is likewise the point when it became more advantageous to be educated in America than abroad). Likewise, when exiting Harvard, your connections to the clergy were important for what job you got. You’d definitely get one, but it was an intensely political process (though less so than back in England).

        Harvard as a status marker is really a 19th century and later thing.

        So I don’t doubt anyone who could read Latin could attend Harvard. But it was effectively declaring they wanted to be a priest rather than a generalized access to elite professions.

        • Steve Sailer says:

          One interesting question is why Continental European colleges have declined so much in global prestige. In 1913, a list of the most famous colleges in the world would include Heidelberg, Gottingen, Sorbonne, Padua, etc., as well as Oxford, Cambridge, Harvard, and Yale. For example, one of the biggest Broadway musicals of the 1920s was “The Student Prince” about how awesome it was to go to a German college and drink beer and get a dueling scar:

          My best guess is because the graduates of Oxbridge and Haryale won the Big Ones, while the graduates of the Continental schools were humiliated in 1914-45 by defeat.

          After 1945, Continental universities mostly switched to an anti-elitist open admissions, low tuition model with huge class sizes. In contrast, the victors’ elite universities stayed elitist, and in the case of the American colleges, became much more intellectually elitist.

          Today, few remember “The Student Prince,” while Evelyn Waugh’s 1945 novel “Brideshead Revisited” about life at Oxford in 1919 is hugely famous. Waugh was convinced in 1945 that the future was dismally proletarian due to the sacrifices made to win the war. But, what he didn’t realize was that winning the Big One meant that global elites in 2019 would find Oxford hugely glamorous, with Waugh’s vision of Oxford resounding among the nouveau riche of Bangalore.

          • johan_larson says:

            It wasn’t just the humiliation of losing. It was also the burden of reconstruction. The fighting in WWI and WWII didn’t take place in New England.

          • A1987dM says:

            Not sure about Sorbonne, but Heidelberg, Gottingen and Padua had the Jews (and I think other non-Nazis/non-fascists too) kicked out of them, many of whom fled to the US (e.g. the Manhattan Project) and started working at American universities.

          • Protagoras says:

            @A1987dM, I read some research on the productivity of German scientists before, during, and after the Nazi era, and it wasn’t just loss of the direct contribution of the Jews, though some of the other factors seem to have been indirectly connected to that. Non-Jewish German scientists were more productive before the Nazi era, and saw a major drop under the Nazis (despite the Nazis providing considerable increases in funding in many cases) and showed relatively little recovery after. It wouldn’t be surprising if similar factors operated on other academic fields apart from the sciences.

          • Steve Sailer says:

            BMW and Mercedes-Benz have been fully reconstructed, but not Heidelberg and Gottingen.

            The prestige of elite colleges has a surprising amount to do with who produces the elites who win World Wars. The graduates of Oxbridge and Harvyale won the Big One, and in the world of 2019 that still matters.

            It’s almost as if who built the A-bomb and who got crushed into submission still matters.

          • Nick says:

            For example, one of the biggest Broadway musicals of the 1920s was “The Student Prince” about how awesome it was to go to a German college and drink beer and get a dueling scar:

            Hilariously, from my perspective, in The Education of Henry Adams, Adams hates his educational experience in Germany, though he faults the system:

            He [i.e., Adams; it’s all written in the third person] had revolted at the American school and university; he had instantly rejected the German university; and as his last experience of education he tried the German high school. The experiment was hazardous. In 1858 Berlin was a poor, keen-witted, provincial town, simple, dirty, uncivilized, and in most respects disgusting. Life was primitive beyond what an American boy could have imagined. Overridden by military methods and bureaucratic pettiness, Prussia was only beginning to free her hands from internal bonds. Apart from discipline, activity scarcely existed. The future Kaiser Wilhelm I, regent for his insane brother King Friedrich Wilhelm IV, seemed to pass his time looking at the passers-by from the window of his modest palace on the Linden. German manners, even at Court, were sometimes brutal, and German thoroughness at school was apt to be routine. Bismarck himself was then struggling to begin a career against the inertia of the German system. The condition of Germany was a scandal and nuisance to every earnest German, all whose energies were turned to reforming it from top to bottom; and Adams walked into a great public school to get educated, at precisely the time when the Germans wanted most to get rid of the education they were forced to follow. As an episode in the search for education, this adventure smacked of Heine.

            The school system has doubtless changed, and at all events the schoolmasters are probably long ago dead; the story has no longer a practical value, and had very little even at the time; one could at least say in defence of the German school that it was neither very brutal nor very immoral. The head-master was excellent in his Prussian way, and the other instructors were not worse than in other schools; it was their system that struck the systemless American with horror. The arbitrary training given to the memory was stupefying; the strain that the memory endured was a form of torture; and the feats that the boys performed, without complaint, were pitiable. No other faculty than the memory seemed to be recognized. Least of all was any use made of reason, either analytic, synthetic, or dogmatic. The German government did not encourage reasoning.

          • A1987dM says:

            @Steve Sailer:

            1) What share of BMW’s and Mercedes-Benz’s productivity was from Jews/commies/other enemies of the Party compared to those of Heidelberg or Gottingen? 2) Like Protagoras, I’m under the impression that the decline began long before the Bomb was dropped.

          • Douglas Knight says:

            The German system is designed to avoid elite schools. Students go to the local school. I think professors are discouraged from congregating, too. (They now have elite research institutes, but those don’t lend prestige to undergrads, because there are no undergrads.) But I’m not sure that the system was that different before the war.


            Were people in 1913 enthusiastic about the Sorbonne or the University of Paris? Their status was up in the air since Napoleon. The schools he made elite are still elite, they just don’t attract foreigners.

            The University of Paris was modified in 1968 at the behest of protestors. But the same thing happened to CCNY, so that’s not such a difference. It’s just that the American system was private and not so centralized.

          • Steve Sailer says:

            CCNY was one of the few American colleges that took 1968 seriously. It switched over to open admissions, and instantly plummeted in prestige.

            CCNY post-1968 was pretty similar to most Continental colleges post-1945: fairly open admissions, huge class sizes, very low tuition, low prestige.

            There’s something to be said for the anti-elitist post-1945 Continental model, but it definitely doesn’t appeal to rising elites in Shanghai and Mumbai.

        • Garrett says:

          > reasons Cooper dislikes Trump: Anderson Cooper is a Vanderbilt
          It’s not that he’s a Republican. It’s worse: he’s New Money!

      • Nick says:

        (Also, I was surprised how lopsided the entrance requirements were between Latin and Greek. Knowing all the noun and verb paradigms is material that will fit on a few pages. Composing fluent poetry in the native style is, let us say, a higher bar than that.)

        Latin noun and verb paradigms will fit on a few pages. My introductory Greek textbook’s grammatical appendix was nearly 70 pages.

        • Michael Watts says:

          Greek noun and verb paradigms will also fit on a few pages, though more pages than Latin will take. They won’t fill 70 pages; I’m going to have to speculate that your grammatical appendix included other grammatical information. I’d be shocked if inflectional paradigms took more than 15 pages.

          • Nick says:

            The appendix is 379-438 inclusive, which works out to 60 pages. Nouns are 4 pages, pronouns are 5, adjectives are 5, numerals are 2, adverbs are 1. Ω-verbs are 22 pages and μι-verbs are 22 too. That works out to 61 pages because I counted 1 page for both pronouns and adjectives.

            I picked an intro Latin textbook off my shelf and it has a 15 page grammatical appendix, 446-460. The type is smaller and the lines more compacted, but it doesn’t use all the space; I’d say it’s fitting maybe 25% more per page. It covers all the same topics as I listed from the Greek appendix.

            Even discounting the greater number of irregular μι-verb conjugations, I think there’s just more to cover. Noun and adjective declension in Greek is more complicated and having 2 verb conjugations—not to mention the optative mood, contract verbs, and greater number of participial forms—mean there’s a lot more to cover. Even considering verb and noun paradigms alone, the Greek appendix has 48 pages and the Latin has only 10; since those pages are especially compact, we could say the Latin has 15 equivalent, but still.

    • Steve Sailer says:

      There’s a certain amount of populism in running for Congress, as George W. Bush discovered when he lost a bid for the House from Midland, TX. His opponent busted his chops repeatedly for having both Yale and Harvard degrees.

      Here in the San Fernando Valley, my congressman Brad Sherman (D-CA) constantly reminds voters he is a certified public accountant, but seldom mentions that he is also a Harvard law school graduate.

      There are also advantages to local college connections. Before Sherman, my congressman was Howard Berman, who went to UCLA for both his BA and JD. There he met Henry Waxman and they started a Young Democrats club that was a massive force in Los Angeles politics. Berman and Waxman (D-Beverly Hills) served 70 years in the House of Representatives between them.

      Berman, who made it up to Chairman of the House Foreign Affairs Committee, was definitely Ivy League material. (I met Berman in 1975 when I was 16 and he was my state legislator. I prepared a snotty question to outwit him. He instantly saw where I was going and swatted it away.) But his loyalty to the local public flagship university paid off for him in politics.

      • Erusian says:

        I think the populist/globalist divide has less explanatory power than local/national elite networks. These do somewhat overlap: local elites tend to be more populist because they can muster more genuine support from their local bases. But not always. But if you want to be prominent among a local elite, ties to the national elite are less valuable. They still have some value: but not infinite value. In particular, certain colleges already specialize in catering to other elites. Notre Dame and U of Chicago are both somewhat national but also have a huge impact on the Great Lakes region. Miami U isn’t nationally competitive but is very important to Florida and Latin America. And so on.

        I actually know of politicians in Central America who turned down Harvard and the like to go to more Latino universities in the US. They figured they’d meet more of their countrymen there and form closer bonds (which seems to be working for them).

        I personally think we’re entering a period where technology is allowing increasing centralization and this is recreating the 16th-17th century Court and Country fights. We’re seeing powerful local elites skirmishing with powerful national ones. Back then Oxbridge split in England: Cambridge was Country, Oxford was Court (with some individual colleges as exceptions). But the United States is instead creating colleges enmeshed with local networks instead of directly challenging national ones for national dominance. Miami U isn’t going to get anyone on the Supreme Court but it has a lot of US Latino politicians and powerful people in Latin America, for example.

        • Steve Sailer says:

          South Florida is pretty cut off from the rest of the country in terms of colleges. Tom Wolfe’s last novel took place in part at a college he called Everglades Global University, which is based on a real private university in suburban Miami that has something like 40,000 undergrads, but nobody in the rest of the country has heard of it (I’ve forgotten its real name).

          It makes sense for South Florida to concentrate on providing college to rich Latins. My impression is that Latin America has peculiar traditions that make it hard to compete in colleges. E.g., the Latin idea that the police aren’t allowed on campus makes it appealing to turn student political organizations into criminal gangs. Fidel Castro got his start robbing banks and then roaring back to sanctuary at the U. of Havana just ahead of the police cars, which, to be frank, sounds like a much more awesome way to spend your college years than studying. But you can imagine why wealthy dads in Latin America might prefer to send their daughters to the U. of Miami.

  73. Lambert says:

    I’m assuming the UK jump is when all the polys changed their names from $CITY Polytechnic to $CITY $HISTORICAL FIGURE FROM THAT CITY University.
    See: Leics. De Montfort, Sheffield Hallam etc.

    • atreic says:

      Yes, you’re right. Polytechnics were vocational tertiary institutions, that couldn’t award their own degrees. In 1992, parliament made them all Officially Universities, thus causing a huge step change in the number of people at university.

      (Here is an interesting contemporaneous news article about this. I didn’t know newspaper articles from 1992 were all on line now!)

      But just because that is marked in the graph as a sudden step change due to rebranding doesn’t mean there wasn’t a huge rise between the 50’s and 90’s in the number of people in tertiary education – they built most of the polys over this time period to meet a need, and also built a big pile of new universities in the 60s. (This article feels about right.) And they can’t have been dodging the draft in the same way as americans were.

      • Tarpitz says:

        My perception is that York and Warwick are vastly more prestigious than the other institutions founded in the 60s. Am I wrong? If I’m not wrong, whence the difference?

        • rlms says:

          York isn’t vastly above Lancaster and UEA, which I think have increased significantly in prestige recently. If you include converted ones, Surrey, Cardiff and Bath are also comparable. Wildly speculating about the differences between those and the others, it looks like being in a significant city or county town that didn’t have an existing prestigious university was necessary for success.

      • Aapje says:

        Interesting. My Dutch university changed from a polytechnic to a university in 1905.

        I guess that Dutch people respected engineering more.

      • nameless1 says:

        … and one of the results was that e.g. Wolverhampton, having a decent polytechnic (well, that general area was the heartland of the industrial revolution) turning into the third lowest ranking university in the UK. I am not British but I lived there and near it a while. And what makes it particularly bad is that while they were probably not good at teaching art history, they probably didn’t suck at the polytechnic type of stuff, like mechanical engineering. So Wolverhampton Uni’s mechnical engineering graduates were now burdened by a statistic that they graduated from the third worst university in the UK, while clearly it is NOT the third worst in mechanical engineering.

    • Salem says:

      Yes, 1992 is when they rebranded the polytechnics, but this isn’t all the FHEA 1992 did, despite what Scott’s source seems to think. The Act gave a lot more money and independence to the universities, and one thing they did with this was expand their student intake. If you look more closely at the graph you’ll notice that after the step change, the rate continues to go up faster than previously, until about 1999 (i.e. when tuition fees were introduced).

      Atreic is also right, and it’s not just the polys. Scott’s own graph shows that even before 1992, there’s a huge increase in enrollments. It just doesn’t match the timings of the US increase.

    • ana53294 says:

      Were Polytechnics not universities? Why not?

      In Spain, we have plenty of Polytechnics that are Universities, some of them are in the top of Spanish rankings (for what that’s worth).

      There is the Polytechnic University of Valencia (teaches arts, business, science, engineering and architecture), the Polytechnic University of Catalunya (architecture, sciences, design, engineering, business). None of them teach law, or medicine, or the humanities.

      I don’t see why a university should teach every subject to be a university. We don’t have the American model where you can go undecided, after all; you go to university with a chosen degree, so if you want humanities, you go to a non-Polytechnic. And there are humanities courses in every major (mostly language and ethics are requirements).

      • Protagoras says:

        There is no consistency in the systems or the terminology over time or in different places. and in some cases not much even in the same region at the same time. Evidently, Spain does things differently than England, which fits the overall pattern of no consistency.

        • dark orchid says:

          Just for fun, ETH Zurich (Switzerland) which is one of the top 10 universities in the world (source: statistics from ETH Zurich) is technically a polytechnic, unlike the next-door University of Zurich which is nowhere close in the rankings.

  74. Clutzy says:

    I think a chart/graph that is missing from this writeup that would give critical information would be an equivalent graph of your MCAT scores graph except for SAT scores & college admissions. Its probably easiest to do that as an addendum to your UC aside because that data is probably on hand.

    Other than increased international students, if UC SAT scores havent gotten much higher for (in particular the 2nd and third tier schools) would likely be related to people applying to more schools.

    I know even in 2004-06 when I applied vs. 2012 with my younger brother my process was much more difficult.