Is U Worth It? The Economic Value of College

A study of the economic benefits of a college degree does not provide the evidence that people think it does—and the flaws in measurement have profound implications for debt laden Americans.

College costs through four-year public and private colleges and universities have been rising faster than health care costs, yet demand for college has not diminished. For many, college is simply a no-brainer, with an economic benefit that outweighs just about any associated costs—and the rising debt associated with going to college reflects this sentiment.

A study of the economic benefits of a college degree does not provide the evidence that people think it does—and the flaws in measurement have profound implications for debt laden Americans.[/box] But with that total student debt hitting one trillion dollars and growing, and with the sticker price of College having grown more than 250 percent in inflation-adjusted dollars since 1982 it is hardly surprising that questions have been raised about whether that debt is worth it. The most recent answer—from the Pew Research Center—is that, yes, it is.

“In a modern, knowledge-based economy, the only thing more expensive than going to college is not going to college,” Paul Taylor, executive vice president of special projects at Pew, told NPR, after the Center’s report showed that for millennials aged 25 to 32, a college degree will improve earnings by $17,500 more per year, based on median incomes. The report is titled “The Rising Cost of Not Going to College.” The gut feeling that soaring prices may trump the benefit had been countered—and most in the media seemed pretty satisfied with the data.

But not everything looks as it seems in the land of trillion dollar dreams. Leaving aside the value of college intellectually or experientially, and focusing solely on financial justification, the Pew study doesn’t provide the evidence many people seem to think it does; indeed, the financial benefits of college are significantly exaggerated when it comes to the outcomes for individual people making decisions about their education. And this is because the financial benefits ascribed to a college degree derive from looking at median salaries for those with degrees and comparing them to those without. The assumption is that if you get a college degree, you will enjoy the median benefits, and if you don’t, you will earn the median salary of those without a degree. But this simple picture doesn’t work on the individual level; people who go to college have other differences from people who don’t aside from the degree. Put another way, a college degree can neither guarantee a good salary (even statistically speaking) nor is it a requirement.

It’s a classic case of causality gone awry. While undoubtedly a college degree is required for certain high-paying professions, it may not be appropriate to ascribe the full increase in salary benefits to the degree itself, nor to assume for any individual that the college degree will lead to a $17,000 increase in salary.


Highest average debts at graduation by state in 2013, according to The Project on Student Debt. From left to right, they are New Hampshire, Delaware, Pennsylvania, and Rhode Island.

The problem with reasoning to the median benefit is that it’s like saying that school teachers (with salaries at about $53,000 after 2-4 years of experience, according to had they not gone to college, would, en masse, have ended up working in McDonald’s rather than being paralegals ($56,000 per year) and office managers ($64,000). Such reasoning neglects that the people who choose to be teachers have skill sets and employability beyond their degree.

By lumping together all those with a “bachelor’s degree, ” and comparing them to o all those with a “high school diploma,” you exaggerate the differences in income for those people in the middle group, those who might consider going to college, or who are unsure of whether to go to college.

To illustrate the point further, suppose that whether you went to college was based purely on high school grades. Suppose that all students with a grade point average above 3.0 go to college, and 50 percent of students with a GPA 2.0-3.0 go to college, while no students with a GPA under 2.0 go to college. We also have to assume that there are no other differences among these students that might impact their eventual jobs. Now we track those folk—and find that those who went to college got better paying jobs than those who did not. How would we know to attribute the effect to college? Perhaps salary is tightly correlated with high school GPA rather than college. To see the impact of college, you would want to look only at those students with a GPA of 2.0-3.0. For this group, some went to college and some did not, but they have similar profiles otherwise. The different in their salaries may be attributable to college.

The way the Pew study analyzes the issue, the A students are mixed in with the B-/C+ students who went to college, and the A-student salaries push up the median salary of college graduates. Similarly, the C/D students are mixed in with those who got B-/C+ and did not go to college, bringing down the salaries. As a result, the salary difference between degreed and non-degreed workers, and we can’t make any conclusions about whether college is the source of benefit.

People who go to college have many advantages compared to those who don’t, a number of which have nothing to do with college. As Michelle Singletory noted in discussing the Pew findings in the Washington Post, “poverty rates are higher for people with only a high school diploma or the equivalent—22 percent compared with 6 percent of today’s college-educated young adults.” Should we attribute this difference to college? Without a doubt, poverty rates are higher among parents of kids who don’t go to college than those whose kids do go to college—but is the lasting impact of a poor childhood, which also impacts college readiness, success and attendance—going to be broken by actually getting a college degree? The data simply do not tell that story.

The conversation about college gets to the heart of the causal pathway to a well-paying job. Does college create a skill set for someone who would otherwise be in a low-wage job? Will college give a bright lower-middle class graduate an interview for a job she wouldn’t otherwise get, even if the job itself doesn’t require a degree? Is college worth it for an average student in a family with an average income?

The point is not that the data are incorrect: people who have gone to college do make more money than people who don’t. The data, however, do not indicate a clear causal relationship for people who are in the “middle group,” trying to carry out a cost analysis about the relative value of college. Will college have a financial payback that exceeds its cost for young adults coming from families earning $25,000-$75,000 per year?

The Pew study neglects another population, for whom college is decidedly detrimental financially: those who do not graduate. Nearly half of those who start a four-year education drop out (or don’t graduate within 6 years), yet their accumulated debt is simply not taken into account by the Pew report, which considers debt among students who graduate. Evaluating the financial benefit college while only considering the success stories is a bit like asking whether a diet works to lower cholesterol, but we only consider people who have lost weight: if the diet causes half the population to gain weight, and half to lose weight, but we only check the cholesterol of the weight-losers, we will have a ridiculously skewed perspective on the diet’s benefits.

The reasons that students don’t finish are myriad; and for this population, college is neither giving them the skill set nor the earning potential. While there are a few jobs for which “college experience” may be useful, and some students will eventually earn their degree, generally college is not worth significant debt for a student who isn’t successful in obtaining the degree. That part of the “cost of college” is simply not factored into the Pew report,

The misreporting of the impact of college on individuals’ salaries may be understandable: most journalists have attended college, and, according to the Pew survey, graduates overwhelmingly say it’s “worth it”. But then again, people who attended college are also much more likely to have money. Not surprisingly 91 percent of graduates without loans told Pew that they felt college was “worth it,” while only 79 percent of those who did take out loans agreed.

But opinions and aggregate data aside, what is really needed is an in-depth and data-driven analysis of what college means for middle to low-income families, for whom college is neither assumed nor dismissed. This is the slice of the American population for which an honest discussion of what college means to these individuals matters. It’s a matter of whether these students can manage the academic responsibilities to be sure they obtain their degree, whether their job interests require a degree, and what else they may otherwise do if they do not obtain a degree. The benefit may still be present, but it’s likely to be worth less than $17,500 a year.


Rebecca Goldin is Professor of Mathematical Sciences at George Mason University and Director of She received her undergraduate degree from Harvard University and her Ph.D. from the Massachusetts Institute of Technology. She taught at the University of Maryland as a National Science Foundation postdoctoral fellow before joining George Mason in 2001. Her academic research is in symplectic geometry, group actions and related combinatorics. In 2007, she received the Ruth I. Michler Memorial Prize, presented by the Association for Women in Mathematics.

Please note that this is a forum for statisticians and mathematicians to critically evaluate the design and statistical methods used in studies. The subjects (products, procedures, treatments, etc.) of the studies being evaluated are neither endorsed nor rejected by Sense About Science USA. We encourage readers to use these articles as a starting point to discuss better study design and statistical analysis. While we strive for factual accuracy in these posts, they should not be considered journalistic works, but rather pieces of academic writing. 

Submit a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Share This