Why Don’t More People Go To College?

At the Upshot today, David Leonhardt asks if college is “worth it” and answers with a resounding “clearly,” citing data he obtained from the Economic Policy Institute. Leonhardt’s answer, however, raises a bigger question, which he leaves unexamined: if college is such a good investment, why aren’t more people making it?

The data he presents show a big increase between 1979 and 2013 in the earnings of college graduates relative to high school graduates (the top line in the chart below). The gap, which has always been large, grew steadily for more than 20 years from the end of the 1970s into the early 2000s before slowing down in the 2000s. Leonhardt makes much of the uptick in the last couple of years, which puts the returns to college at an all-time high, but the growth in the college premium has clearly decelerated somewhat since about 2002, even with the finishing flourish in the chart.

Financial return to a college degree, 1973-2013
Source: New York Times.

The chart, however, also poses a serious puzzle. Leonhardt concludes that: “from almost any individual’s perspective, college is a no-brainer.” If that is true, then, in a well-functioning market, we would expect that this extraordinary increase in the payoff to college would have led to a large increase in the supply of college graduates. In fact, as economists have documented for several decades now, the supply of college graduates decelerated at about the same time that the return to college started to increased. The share of college graduates in the workforce continued to grow after the early 1980s, but at a slower, not a faster, pace than it had in the 1960s and 1970s.

Unless we think that the main reason young people aren’t going to college is because they don’t know that the average return to a college degree is very high (and always has been), then the much more interesting question is why it is that more young people aren’t going to college? Given the size of the returns to college –Leonhardt cites numbers from MIT economist David Autor that suggest that the present discounted value of college is roughly $500,000– the barriers must be substantial.

I’ll offer two quick reasons why college may be, in practice, less of a “no-brainer.”

First, as Heather Boushey and I (and, more recently, MIT economist Frank Levy and colleagues) have argued, not everyone who goes to college receives the average financial return. A non-trivial portion of college graduates earn less than the average otherwise comparable high school graduate. (In the current issue of Science, which the Upshot piece refers to, David Autor makes this same point: “Although the average college graduate earns substantially more than the average high school graduate, the least successful college graduates may earn substantially less than the median among high school graduates, and the most successful high school graduates may earn substantially more than the median among college graduates.”) At least some high school graduates on the fence about college might realistically expect –because of their grades, test scores, health situation, work responsibilities, or family obligations– that they will find themselves with a below-average pay-off to college.

Second, as Heather and I also argued (and as Ben Casselman notes in a post today at 538), the big financial returns to college depend heavily on finishing your degree, but more than 40 percent of those who start college for the first time don’t finish within six years. Those who don’t finish have little to weigh against their lost earnings, lost years of labor-market experience, direct outlays in tuition and fees, as well as any student debt they might have acquired. Again, from the point of view of a young high school graduate considering college, knowing that there is a 40 percent chance that you will start, but won’t finish is pretty discouraging –again, especially if you have reasons to believe that you might more be more likely to be in the 40 percent than in the 60 percent. (Autor also makes this point in Science: “for students who acquire substantial student debt but do not complete the college degree, it is far from certain that college will prove a good investment.”).

That I can identify, Leonhardt mentions only one reason that we don’t have more college graduates and does so only in passing. He quotes Autor as saying “we have too few people who are prepared for college.” But, if this is the main barrier to increasing the share of workers with a college degree, then we are in more trouble than Leonhardt seems to think. If we have not prepared enough young people to do college work, then we can hardly expect that these unprepared young people will finish college even if they start or that they will receive the high average returns earned by current graduates.

I may be wrong about why young people aren’t going to college in bigger numbers, and almost certainly there are other factors at play, but knowing why the young aren’t going to college in much greater numbers is a far more important question than knowing whether college is “worth it” “on average” and “all-else equal” for the people who have already gone.

(This post originally appeared on the CEPR blog. Updated to correct authorship of Benson, Esteva, and Levy paper.)

3 Comments

  1. Eric L says:

    “Although the average college graduate earns substantially more than the average high school graduate, the least successful college graduates may earn substantially less than the median among high school graduates”

    I saw this, but it strikes me as very apples-to-oranges. The bottom of one group should not be compared to the median of another; there is undoubtedly variation within both groups. A better study would be, how do people who got an 800 on the SAT and did go to college compare to those who got an 800 and didn’t? (Actually the higher ranges would be very interesting here too; I would suspect that the college premium is overestimated due to selection bias.)

    “In fact, as economists have documented for several decades now, the supply of college graduates decelerated at about the same time that the return to college started to increased.”

    This is easily explainable if you assume causality goes the other way — a reduced supply of college graduates would tend to raise the premium employers have to pay to get one. Also graduates may now be a more select group.

    “I may be wrong about why young people aren’t going to college in bigger numbers, and almost certainly there are other factors at play[…]”

    To make the obvious guess — rising costs, meet stagnant wages and public support. The truth is people have never been willing to risk as much debt as they have to now to go to college. The net present value of an education is beside the point; the future is a luxury good.

  2. John Schmitt says:

    Thanks, Eric.

    On the supply issue, that is certainly the standard story. The deceleration in supply led to the increase in the college premium. But, the whole question is why the strong, sustained rise in the college premium hasn’t induced a supply response? (And a closely related question, what led to the deceleration in the first place?) If we think it is important to increase the supply of college graduates, it seems pretty important to understand why market forces alone have, so far, not been enough to create additional supply. (In his piece in Science, David Autor argues –and as far as I can tell, he is the first to make this argument– that the flattening in the college premium since about 2002 reflects a long-delayed supply response to the rising premium.)

    I agree that rising costs are likely to be important in all of this, and that is a point Heather and I also made in our 2010 paper. What is interesting to me is that this argument is often met with a lot of skepticism. The usual response is that net of financial aid, the cost of college has increased far less than the sticker price suggests. This could well be true, but student debt levels have also been rising, which does suggest that aid is not keeping pace.

    As to exactly who and how to compare, I think it is reasonable for this kind of discussion to compare, say, the bottom fourth of college earners, with the top fourth of high school earners in the same age range. More rigorous approaches like the one you suggest have their advantages for some purposes, but the sharper the comparison (perfect SATs, with and without college, for example), the less likely those results are likely to tell us about a large-scale expansion of college to a heteregeneous group of students (by age, high school grades, SAT scores, health status, care responsibilities, need to work while in college, etc.). But, I agree that the “real” return to college is likely to be less than the return calculated in ways that don’t separate out any selection effect.

  3. Eric L says:

    “I think it is reasonable for this kind of discussion to compare, say, the bottom fourth of college earners, with the top fourth of high school earners in the same age range.”

    Sorry, is this what you meant to say? I don’t see why it would be reasonable to compare the bottom fourth of one group to the top fourth of another, could you clarify why that is appropriate here? I guess if all you’re trying to establish is that the range of incomes made by college grads and the range of incomes made by high school grads overlap, this is sufficient, but that’s hardly surprising or interesting and doesn’t tell us much about whether those college grads earning the least are earning more than they would if they hadn’t gone to college.

    Do you find it plausible that 1/5 of college grads have, due to their education, found it impossible to get better paying jobs they would have been able to get had they not gone to college?

    “But, the whole question is why the strong, sustained rise in the college premium hasn’t induced a supply response?”

    Is the bigger question why the rise in the premium hasn’t caused a rise in graduates, or why with the premium having been consistently high the number of graduates hasn’t been higher all along? Both interesting questions, possibly with different answers. But the first question seems easy to me once you note that the cost of education has also been rising. The second question though has many potential answers, ranging from expectations of dropping out to hyperbolic discounting to uncertainty about which particular majors will command a premium.

    “The usual response is that net of financial aid, the cost of college has increased far less than the sticker price suggests.”

    I wonder how rising sticker prices have affected the rate of college applications? Applying to college is a time consuming task and comes with fees, and at the time people are deciding to apply they only know the sticker price, not the net price after financial aid.

Leave a Reply