What the data actually shows
The landmark work here is by economists Stacy Dale and Alan Krueger. Their clever design compared students who were admitted to similarly selective colleges but made different choices — some attending the more elite option, some choosing a less selective one. By matching on the schools students applied to and got into, they could isolate the effect of attendance from the effect of the underlying student. Their 2002 study found that, for most students, attending a more selective college did not produce a meaningful earnings premium once this matching was done.
A follow-up analysis (Dale and Krueger, 2014) with a larger, longer-run dataset reached broadly the same conclusion: the apparent payoff to selectivity largely reflected which students attended those schools, not the schools themselves. The raw correlation between elite colleges and high earnings is real, but much of it appears to be selection — talented, driven people both get into and graduate from selective schools — rather than a causal boost from the institution.
The notable exception in the same research: for Black and Hispanic students, and for students from less-educated families, attending a more selective college was associated with higher earnings. For these groups, the elite school appeared to provide access — to networks, information, and opportunities — that they were less likely to have otherwise. So the average finding masks a real difference between who benefits and who does not.
Why this feels different from how it actually is
It feels like the school must matter enormously because the visible examples are skewed. The most prominent people in finance, tech, law, and politics disproportionately attended a handful of elite institutions, so the brand looks decisive. But this is survivorship bias: we see the elite-school graduates who succeeded, not the equally capable people who succeeded from ordinary schools, nor the elite-school graduates who did not.
Admissions culture also amplifies the stakes. The intensity around getting into a 'good' school sends a message that the decision is permanent and life-defining, which makes the institution feel far more determinative of your future than the long-run data supports.
And because prestige is a clear, visible signal and ability is not, people lean on the signal. A famous college name is easy to point to; the harder-to-observe traits that actually drove the outcome — drive, talent, the ambition to apply high in the first place — quietly do most of the work.
The student matters more than the school — which cuts both ways: an elite school doesn't guarantee the outcome, and an ordinary one doesn't foreclose it.
What the research says to do about it
Treat fit, cost, and completion as the variables that the evidence supports caring about. Graduating with manageable debt, finishing the degree, and being in an environment where you can do well tend to matter more for most students than incremental gains in prestige. Avoiding heavy debt for a marginal bump in selectivity is, for the average student, a reasonable read of the research.
If you come from a disadvantaged or first-generation background and have access to a more selective school, the research suggests that access can be genuinely valuable — partly through the networks and information it opens up. The benefit the average finding washes out is concentrated in exactly these cases, so it is worth weighing seriously.
For everyone, the durable predictors of career outcomes — skills, effort, the field you choose, and the relationships you build — are largely within reach regardless of where you study. The data points toward investing in what you do during and after college, rather than treating the admissions decision as the thing that settles your trajectory.
What the research says does not help
Taking on large debt for a marginally more prestigious school does not, for most students, pay for itself in the way the brand premium implies. Because the earnings gap between similarly capable students at more and less selective schools is small on average, the borrowing can outweigh the benefit.
Treating the college name as a verdict on your potential does not help and is not supported by the evidence. The research consistently finds that the student matters more than the school, which cuts both ways: an elite school does not guarantee the outcome, and an ordinary one does not foreclose it.
Assuming the school never matters is the opposite error. For some groups the access an elite institution provides is real and measurable, and in a few specific high-prestige fields, brand and network effects do exist. Flatly dismissing the school's role overcorrects past what the data shows.
The benefit the average finding washes out is concentrated in exactly the students least likely to have other routes to the same networks.
What this looks like in real life
Matching students who got the same offers
Dale and Krueger compared students admitted to similarly selective colleges who then made different choices — some taking the more elite option, some a less selective one. By matching on where students applied and got in, they isolated the effect of the school from the effect of the student. For most students, attending the more selective college produced little to no earnings premium once that matching was done.
Debt for a marginal name
Because the earnings gap between similarly capable graduates of more and less selective schools is small on average, taking on large debt for a marginally more prestigious name often doesn't pay for itself. For the average student, avoiding heavy borrowing for a small bump in selectivity is a reasonable read of the research — though for a disadvantaged or first-generation student, the access an elite school opens up can change that calculation.
Real numbers in context
The headline figure from this body of work is essentially a non-finding for the average student: once researchers matched students by the selectivity of schools they applied to and were admitted to, the earnings difference attributable to attending the more selective college was roughly negligible (Dale and Krueger, 2002; 2014). The strong raw correlation between elite schools and high pay appears to be mostly selection — who attends — rather than a causal effect of the institution.
The exception is specific and worth stating plainly: for students from disadvantaged backgrounds, including Black and Hispanic students and those whose parents did not attend college, attending a more selective school was associated with higher earnings. The benefit of an elite college, in this research, is concentrated among the students least likely to have other routes to the same networks and opportunities — not spread evenly across everyone.