Varsity Blues and the Value of Elite Education
American high school students are increasingly set on getting in to a handful of prestigious colleges. Are these schools really so much better than everywhere else?
Updated 3/29
I, like pretty much everyone else in America, recently watched Netflix’s “Operation Varsity Blues: The College Admissions Scandal.” Two major themes stuck out to me. The first is the fact that elite colleges are becoming an increasingly all-consuming goal for U.S. high schoolers. This was most obvious in the wealthy families willing to spend millions and risk prison to get their kids into Yale, Stanford, and USC, but the documentary also shows scenes of top college mania in the culture at large, including vignettes where students break down in emotion after receiving acceptance or rejection letters from elite schools.
If the film’s first goal is to convince us that there is an ever-increasing demand for spots at the country’s top universities, its second goal is to convince us that this is wrongheaded. The documentary’s experts hit the same message three or four times: there is nothing worth competing over at elite colleges. A representative quote is reproduced below:
It’s typically accepted that Ivy League institutions are the quote-unquote best in the country, but all of those differences have almost nothing to do with the academics of the institution … U.S. News started ranking colleges in the ‘80s, based on one criteria—prestige. That’s it.
Prestige is actually a French word. And in the original French, it means something people don’t realize. It means deceit. That’s what prestige is, in the college. It’s imaginary. It’s an illusion. Yet people believe in it.
This is a bold claim, but it doesn’t sound entirely crazy to me. Yet the documentary gives no evidence to back it up. For all we know, Harvard students learn three times as much as everyone else and all go on to become millionaires—in that case, the country’s fascination with elite schools would be well justified. Without supporting evidence, the documentary’s platitudes about the emptiness of academic prestige are just that—platitudes.
This question is somewhat personal for me. In my senior year of high school, I spent a tough three weeks trying to decide between Yale and the (much cheaper) honors program at my local state university. I like to think of myself as disagreeable enough to make an unpopular decision in these kinds of situations, but in the end I gave up trying to reason things through and went along with what cross-admit statistics imply pretty much everyone in my situation has ever done: I went to Yale.
I’ve been asked about this sort of decision by a number of college applicants in the time since, and I’ve never had a good answer. It’s time to take a look at evidence.
For the moment, I’m going to ignore the squishy, subjective stuff like how enjoyable the experience is and focus only on the measurable improvements in post-graduate success. It seems to me there are roughly two ways an elite university might increase your level of economic capacity: through teaching you important skills or through giving you the credentials and social network to land higher-paying jobs. Luckily, there has been quite a bit of research in both of these areas.
The Learning
Before we look at elite schools in particular, let’s take a step back and see what kinds of human capital increases students get from their post-secondary education in general. The answer, according to the well-publicized case made by the economist and blogger Bryan Caplan, is almost none. Caplan argues that college is almost entirely about signaling conformity, intelligence, and work ethic to employers. This belief is not as fringe as it might sound; a survey by Caplan found that 85% of econ bloggers believe half or more of the value of an education is signaling, not human capital formation.
This case sort of makes sense to me. I’m studying Computer Science, which is ostensibly very practical, but there are still plenty of subjects we study that don’t seem like strict prerequisites for starting work. The strong emphasis the school places on fairness in grading and sniffing out academic dishonesty doesn’t really make sense if classes are all about the learning. Even more tellingly, the increasing desperation around getting into top schools suggests that Yale classes should be under stricter guard than Afghan poppy fields; instead, it’s totally possible for anyone to just waltz in and watch whatever lecture they want. The fact that more students don’t choose to get an Ivy League education for free by just sitting in the back of classes (or even taking them online) suggests that most people intuitively understand this state of affairs.
Before we get too excited about the idea that college doesn’t teach you anything useful, I should note that a 2016 meta-analysis found that critical thinking scores improved considerably over a student's time at university. I recently wrote a post on how my general problem-solving strategy had been improved by the classes I’ve taken, so this result sort of matches with my experience. On the other hand, the study also found that results in the same ballpark could be observed after just one semester of philosophy, leading the authors to write that “there is probably some upper limit to the amount of critical thinking skills that will be learned and retained in college.” It seems unlikely that students are really getting high-levels of sustained critical thinking growth over all four years of their undergraduate degree.
As far as I can tell, there have been no studies that examine whether more prestigious universities are more effective at imparting knowledge and critical thinking skills than less prestigious ones. The inference we can draw from the general data, however, doesn’t look great for elite education. If Caplan is right and no school teaches anyone anything useful, then clearly Harvard has no advantage over UMass. Even if we conclude there are some human capital benefits to attending college, the fact that it is hard to detect the difference between UMass and no school at all means that the gap between UMass and slightly faster classes at Harvard is probably very small indeed. I draw a similar inference from the critical thinking data. If there is a hard upper limit on critical thinking gains, it seems likely to me that UMass students will approach this limit just as surely as their Ivy League peers over the long run.
The Credentials and Connections
After a disappointing performance in the first round, Ivy League boosters should be hoping for a home run in the credentials and connections department. This isn’t a far-fetched idea; a degree from an elite school seems like a valuable signal of quality, and it’s plausible that an introduction to upper-class social networks could turbocharge one’s social mobility.
The data on student outcomes looks really good for elite schools here. A glance at the Payscale college salary report confirms that there are big differences between schools in both early and mid-career pay, with a clear bias towards traditionally prestigious institutions. The average Harvard new grad, for instance, makes $22,000 more each year than the average UMass Boston graduate. This isn’t all about career choice, either. Harvard, Yale, and UPenn all perform quite well despite having only about 20% STEM majors.
Of course, correlation does not equal causation, and that should be especially obvious in this case. Almost by definition, elite schools select for students with high test scores and a track record of success. We would expect these students to outperform others post-graduation, even if they all went to the same university. To get to the bottom of this question, we need to make comparisons between students of similar caliber.
Dale and Krueger’s 2002 paper “Estimating the Payoff to Attending a More Selective College” is the first in a series of studies that performs this comparison through the “matched applicant” model—that is, the authors compare outcomes for students who attended a prestigious school with the outcomes for those who were admitted and chose not to attend. A key assumption of the model is that students who chose to attend the prestigious school are fundamentally similar to the students who chose not to attend. If this is not the case, Dale and Krueger write, “estimates from the matched-applicant model would be biased upward or downward, depending on whether more talented students chose to matriculate to more or less selective colleges conditional on their options.”
Using the College and Beyond data set and National Longitudinal Survey of the High School Class of 1972, the researchers found that students who attended a school with a 100 point higher average SAT score earned about 7.6 percent more in 1995. So far, so good. This matches about what we would expect from eyeballing the Payscale data. When comparing “matched applicants”, however, Dale and Kreuger find that “the effect of school-average SAT is slightly negative and statistically indistinguishable from zero.” That is, among qualified applicants, there is no positive effect of attending a selective school on income. Suddenly things are not looking so good for elite colleges.
This paper gets cited a lot (1266 times, per Google). The matched applicant model and similar studies are often taken as a slam-dunk indictment of elite schools. I think there are a number of good reasons to question this interpretation:
The fundamental assumption of the paper is suspect. As the authors acknowledge, the study relies on the assumption that students who are admitted to but choose not to attend elite schools are the same as those who do. I think it’s pretty easy to imagine a world where this is not the case. Consider the fact that only 18% of applicants turned down Harvard in 1973. If we make the reasonable assumption that most of these students went to another very prestigious school, we are left with only a very small sliver of the population who willingly took a big reduction in school prestige. In other words, these people are weird. It seems reasonable to assume that people who go boldly against the grain are statistically different from the rest of the Harvard admits; perhaps they are iconoclasts by personality, or so confident in their own abilities that they don’t feel the need for the Harvard stamp of approval. On top of this is the fact that elite schools are famously generous with financial aid, offering full-ride scholarships to all students below a family income threshold. Given that elite schools are free for very poor students and expensive for wealthy students, I would expect the students who turn down Harvard to be wealthier on average than those who don’t. It’s pretty easy to tell a story connecting any of these traits to increased post-graduate income. Anecdotally, the one person I know who got into an elite school but decided to go somewhere significantly less prestigious was a genius who felt he didn’t need credentials to succeed in the tech industry.
The paper did find an effect for poor students. Dale and Kreuger report that for students from families in the bottom decile of income, the matched-applicant model shows that attending a college with a 200 point higher average SAT score is associated with an 8% increase in 1995 income. This suggests to me that the connections and credentials gained at an elite school can be incredibly valuable to students who don’t already have them. My experience in coming to Yale from a public school in the Midwest leads me to wonder whether this effect generalizes to students from families that are financially comfortable but not plugged into the elite job market; as I wrote about in a previous post, Yale has connected me to career opportunities I’m not sure I would have found otherwise. The unconvincing results found in the general population may simply reflect the fact that many of the students admitted to Ivy League schools are from wealthy backgrounds and therefore already have the upper class social connections they need to succeed.
The returns to elite schooling depend heavily on sector. Undergraduate prestige is much more important in some sectors than others. A study of prestigious universities in Chile, for instance, found that elite business programs had large effects on the income of graduates, most of whom went into finance. As far as I know there have been no sector-specific matched-applicant studies in the U.S., but Ivy League students are overrepresented at top consulting and finance firms in a way that suggests a prestigious degree gives you a major leg up in these fields. This phenomenon is common knowledge among my peers at Yale—I seriously considered consulting, mostly because of how well established the pipeline to top firms is. An elite undergraduate degree is probably even more important for students pursuing careers in academia, where prestige and access to research with top professors can help secure spots in selective grad programs. This fact may give some insight into why the Dale and Kreuger study found no effect of college prestige on income. In 2020, almost 20% of Yale students went to work in academia, a field where we wouldn’t expect even profound per-university advantages to manifest in significantly increased mid-career pay.
The sum of the evidence here is not straightforward, but some general themes emerge. The credentials and connections gained from elite education clearly have a profound positive effect on students from disadvantaged backgrounds. For students who are already connected to a strong social network, it’s not clear that elite education is a panacea guaranteed to improve life outcomes for everyone. That being said, ambitious applicants with their heart set on high-status jobs in consulting, finance, or academia will benefit considerably from attending a prestigious undergraduate institution.
Conclusion
All told, the economic case for elite education is complicated. It seems unlikely that the education offered at prestigious schools leads to significantly higher levels of human capital than one could get from the same classes at their local state university. The degree and connections are more valuable, but exactly how valuable they are depends considerably on the student’s background and career ambitions.
This data seems to align pretty well with my own experience at Yale. My classes have been fast-paced and engaging, but it’s hard to imagine that what I’ve learned will make me significantly more productive on the job than the version of me who went to my local state university. The credential and exposure to opportunities at top companies have been very useful, although I think there’s some chance I would have figured out how to land prestigious internships as a junior at the state school as well. This is partly due to the fact that the tech industry tends to place a greater emphasis on test scores than credentials and partly due to the democratization of job information facilitated by forums like Reddit’s r/csMajors. If I was going into consulting or finance, I think I would be much worse off at the state school.
Overall, this seems like a conclusion college applicants can find somewhat comforting. Elite schools can give students a big leg up in the competition for research positions and high-prestige jobs, but they are not the only path to success. According to the data, smart, driven people seem to do well wherever they go.
Postscript
I was careful to avoid the question of personal enjoyment/fulfillment in this post, as I think it varies too much across individuals to be analyzed in any kind of helpful way. That being said, I think an intellectually curious person will find some exhilaration in the broad-based, fast-paced learning one gets at an elite school.