Kevin Carey’s provocative The End of College asserts that our four-year universities and colleges are ineffective and much too expensive. And the jig will soon be up as new technological delivery systems—in particular, “massive open online courses,” or MOOCs—drive the vast majority of existing institutions out of business in the not-too-distant future. Even in the short run, Carey’s warning to parents today is that “universities have been ripping off parents and students for decades… Students are being left to the whims of professors who haven’t been trained to teach and aren’t accountable for helping students learn… Don’t let this happen to you.” After enumerating a few programs that he finds acceptable, he concludes that the true value of the bachelor’s degree will be exposed, and students “will be left with little but debt and lost time.” Therefore, graduating high school seniors should pass up traditional four-year programs and instead construct their own curriculums with a combination of MOOCs, apprenticeships, and work experiences that can lead to “badges,” an emerging form of credentialing.

Carey is the director of the education policy program at New America and has written widely on many aspects of higher education. In The End of College, he embraces the school of thought that technology changes everything—more specifically, it changes how college education will be delivered and how people need to be prepared to become productive workers. His blanket condemnation of the current higher education system is worthy of our consideration. He correctly points out the contradictions of the “hybrid” university that tries simultaneously to serve practical functions (preparing young people for work), to advance research (done by professors and university centers), and to expose students to humanistic views in a diverse liberal arts curriculum. Further, he notes that graduate schools do virtually nothing to prepare their students to teach and that colleges have become excessively expensive.

But there are also some questions we should ask of him, too. First, his finding that many people don’t show learning gains in college is based on a faulty understanding of one study. Second, much of what young people learn in college involves perseverance, task completion, and cooperation—matters that Carey gives short shrift. These are important qualities that allow philosophy and English majors to become mid-level managers. Third, the relatively high pay that college graduates receive compared to non-graduates seems to me a key validation of the current system.

For all the current system’s flaws, a comprehensive education program based on nonresidential MOOCs and other forms of skill validation can never be effective. Carey portrays the future as one in which individual students develop their own plan of course-taking and other forms of training. He believes that new institutions will develop standards that will accredit classes, and other institutions will facilitate the living arrangements and social interactions of groups of “students” who aren’t tied to a single education or training institution. But in this brave new world of free-flowing higher education, many young people, especially those from low-income families would be left behind. They would miss the structure that an institution can provide to “lead” them through the process of taking electives, declaring a major, and meeting all the requirements to graduate with degrees. Current college students also benefit greatly from the social interactions with other students and from the peer pressure to succeed. Finally, studies also show that the relatively few who complete MOOCs are those who are the most self-motivated and are often the ones with the most prior education.

Higher education essentially has two functions: First, for those who reside on or near campus, it provides a period of semi-independence and autonomy in a protected environment with many social interactions; and second, it develops the workplace skills of general cognition, ability to learn, task completion, group and organizational skills, and, for many students, a field-specific knowledge base. While Carey does show the limitations of the current system, he falls very short in showing how a MOOC-based system can be scaled up and produce better results in preparing young people to enter and succeed in the labor force.

Before World War II, higher education was reserved for the few elites and for those who would form the beginning of a professional class with the advent of medical and law schools. The expansion of four-year college degrees in the United States saw two big surges—the first in the wake of the 1944 GI Bill and, later, as a result of the Baby Boomers coming of college age in the 1960s. By the mid-1960s, 25 percent of Americans who turned 25 earned a bachelor’s degree, and more than one-third of this group earned a graduate degree as well. In the 1970s, many commentators thought that having one quarter of workers getting a B.A. was too high (see, for example, Richard Freeman’s The Overeducated American). But educators convinced political leaders that improvements in the cognitive and organizational skills of the citizenry would pay off in the form of higher productivity.

In time, this approach proved successful, and most other industrialized countries followed suit. Carey is misleading when he writes America has many more B.A.’s than other advanced countries. While this is true for older people, we are now in the middle of the pack among advanced industrial countries in the share of 25-to-34 year-olds with a B.A. The new standard here and abroad is having about one-third of each new age cohort earning a college degree. For many leaders, including President Obama, our movement from the top to the middle of the pack is disturbing, and there have been calls to expand higher education. By contrast, Carey and the people who advocate the widespread use of MOOCs don’t think in these terms because they don’t think that degrees reflect skills.

But in virtually all modern, industrialized societies, higher education has become the main path for preparing workers for the new service economy based in offices, health care, and education. The costs of such education in dollars and time are indeed immense—but there is a large payoff for the economy as a whole.

Economists have devoted a lot of research to tracing the earnings of workers with different levels of education. A standard way to present the data is to show how much greater the earnings of workers with just a B.A. are than the earnings of those with just a high-school diploma. This premium has changed dramatically over time and has grown considerably since 1980 (as Claudia Goldin and Lawrence Katz detail in The Race between Education and Technology).

Carey has very little to say about the earnings premium: There is one brief mention early on, and another short discussion starting on page 194. Since he doesn’t believe most students develop skills in college, he views college as a signal that they were smart and from good families when they left high school—getting a diploma, he writes, is “a kind of acculturation [and] rite of passage for entrance into the upper middle class.” Carey quotes the late university president Robert Hutchins to argue that attending classes, passing exams, and managing the complex processes needed to graduate are described require little more than “faithfulness, docility, and memory.” He modifies this statement a couple of pages later when he says that “for a large number, the college degree is overwhelmingly a signal of general cognitive ability, acculturation, aptitude for becoming part of a large organization—and nothing else.”

But these are skills that are very important to employers. They lead to higher earnings, and because of these high earnings, the demand for a four-year degree in the United States has never been higher: Parents and high school students want it, and large majorities of people who graduate say that it was worth the cost and time. In an earlier work with Georgetown’s Anthony Carnevale, we found that over a 40-year career those with a four-year degree earned $1 million more than those with a high-school diploma and no postsecondary education (this calculation excludes the 35 percent of B.A.’s who went on to earn a graduate degree and who have much higher earnings.).

Some argue that it is not the college but the students—i.e., smart and talented people know that they should go to college, and these people would make more money even if they didn’t attend college. Empirically it is very difficult to differentiate between the students’ raw abilities and college going. Regression analyses on the few data sets that have measures of pre-college ability and post-college earnings mainly show that it is the added value of attending college that is responsible for the extra earnings. While it is hard to disentangle the effect of growing maturity with age, it certainly is clear that college seniors are much more capable than first-year students.

Finally, no industrialized country has chosen the path of limiting the share of workers with a four-year degree to the under 10 percent that existed in all countries prior to 1950. If we are overeducated and wasting money, so is the rest of the world. When countries are predominantly agricultural or in the early stages of industrialization, there is no need for them to have many people with college degrees. However, this investment becomes worthwhile once the economy has shifted to a high-end service economy in which the majority of employment is in office work, public administration, health care, communication, and education.

For these tasks, strength and manual dexterity have been replaced by analytic, organizational, and communication skills. On the one hand, at least half of all graduates have majors that are tied, at least moderately, to a specific set of occupations—e.g., teaching, business, nursing, engineering, and so on. On the other hand, the four years of classes (often taken over more than four calendar years) challenge students in many ways: They learn how to juggle courses, to understand what is required to satisfy each instructor, and to work in concert with a wide range of other students. While most of the specifics of what students learn in college are not retained for much longer than the end of the semester, the very process of going through these four years creates adaptability and other skills that will be useful in future careers.

Carey cites Richard Arum and Josipa Roksa’s Academically Adrift showing that “45 percent of students made no gains on a widely used test of critical thinking, analytic reasoning, and communication skills during their first two years in college.” But Alexander Astin, in an early review of Academically Adrift in the Chronicle of Higher Education, argues persuasively that Arum and Roksa did not prove that 45 percent had no learning gains. What they did report was that 55 percent of students had a “statistically significant” gain. These last two words are very important because all aptitude assessments contain a large degree of variation—i.e., the same person taking different versions of the same assessment will get different scores. Therefore, the 55 percent who registered gains should be called “large gainers” because the second score must be considerably higher (at least 1.96 standard deviations) than the first score, while most of the other 45 percent had small gains that were not statistically significant. Furthermore, two other studies (one by the creator of the assessment instrument that Arum and Roksa use) found larger learning gains.

Consequently, this central pillar of Carey’s argument is weak—and he fails to note that college graduates do very well as adults in terms of many metrics in addition to earnings: low unemployment rates, higher marriage rates, better physical and mental health, and higher self-reported indicators of happiness.

A large part of Carey’s book is devoted to his experiences taking the online version of the MIT freshman biology course “The Secrets of Life,” taught by a wonderful professor who covers considerably more territory than is usual. One indication of the extra content is the use of complex problem sets that often require great effort until the concepts are learned. Further, there is an online community of students available for more interactions about the course.

In successive chapters, Carey argues that MOOCs will lead to much better learning outcomes for three reasons. First, the courses can be refined over time toward the most effective approach (some shudder at the thought that there is one and only one most effective approach). Second, students can learn at their own pace by replaying the videos of the lectures and by answering the problem sets online until they get the right answer. Third, computers can bring together the experiences of thousands of students to develop algorithms to tailor learning pathways based on how each student has done on the problem sets.

These factors are very important and there is no doubt that MOOCs will play an important role in the future of higher education. But Carey views MOOCs as superior to normal college learning. He even proclaims that his B.A. conveyed no knowledge about what he had learned but that his certificate of passing MIT’s course “was the most credible, discoverable evidence of learning I had ever produced—other than, perhaps, this book.”

This is an odd statement for a very successful mid-career professional to make. On the one hand, Carey passed a freshman biology course that wouldn’t get him a job as a lab assistant. He would need many more years of classes in biology for this course to have any effect on his career options. On the other hand, his B.A. led to a master’s of public administration, which was followed by several years working in various positions in the Indiana state government. He transitioned to a series of successes in the Washington, D.C. policy arena before establishing his own nonprofit educational research organization that is able to sustain itself through grants from many foundations. He is a national leader in his field.

In the end, Carey relies on his good single experience with a MOOC to generalize that this can be the best way for all students. He never fleshes out what the alternative to the current college system would look like. For example, would there be a minimum credit requirement for a degree? Would there be a required curriculum that covers a wide variety of subject areas? Would there be something short of baccalaureate that would mainly consist of a limited concentration of technical study? He brushes aside these specific issues and assumes that they will be worked out over time.

What he fails to grasp is that colleges provide the institutional heft to impose certain standards that students know they must meet. While I was a good student, my strategy was always to do the minimum that I needed to meet the expectations of my parents (which was to get A’s in science and math and high B’s in the humanities and social sciences). Ultimately, I just followed the path that was set out before me, and only really became a self-starter in my second graduate school.

The vast majority of young people lack the individual motivation and understanding to put together a variety of MOOCs that would lead to anything that would resemble what is required of college degree today. Further, there will be the temptation to limit MOOCs to only courses in one’s intended area of study. These “efficiencies” minimize the chance that an elective will open up a whole new life path.

Carey says that the instigators of change will be for-profit companies that will see an untapped worldwide market of more than $4 trillion. Yet he doesn’t discuss the current sorry state of for-profit education. Many of these institutions have focused on certificate programs (the kind of specialized skills that Carey seems to think are important) and have a mixed-to-poor record. While they do help hard-to-serve populations, their students are often left with lots of debt, and many of their certificates and four-year degrees have been of dubious value.

Carey’s book came out last spring. His optimism about MOOCs frankly seems a bit dated now as many of the problems that I cite above have become more prominent. So far, most online programs have not been viewed as equivalent to campus-based four-year degrees. For example, many graduate programs do not accept students with online bachelor’s degrees. Similarly, the University of Phoenix, a for-profit college that offers online courses, has expanded its model to include campus-based learning even as the number of students that it serves has dropped from 600,000 in 2010 to 227,000 in 2015.

By contrast, the current higher education model is working here and in other industrialized countries. Cost is of course an issue, and Carey’s one mention of debt involves a George Washington University student expecting to graduate owing more than $100,000. Left unsaid is that less than one-half of 1 percent of B.A. graduates leave with this amount of debt. The listed tuition and room and board fees at the most expensive private universities are sky high, but more than 30 percent of B.A. graduates leave with no debt. Finally, those who pay the full rate provide lots of money for institutional grants that go to the 60 percent or more of students that receive some sort of aid at these schools. As economists like to say, “there are no free lunches”; as long as students want many extracurricular options, someone has to bear the cost and it is not unreasonable that beneficiaries (especially if they are from above-average income families) bear a large share.

The real higher education problem today concerns the children from low-income and minority families, who aren’t getting a bachelor’s degree in large numbers. Much of the difference has to do with their poorer performance in high school. The good news is that record numbers of these students do enroll in post-secondary education in community colleges and four-year open-admission state schools near their home. But of those who did do well in high school, an alarming number don’t pursue four-year degrees at one of the better colleges because of costs (even when they get full scholarships, there are still travel and living costs that aren’t covered), poor counseling, and lack of awareness of financial aid.

While the current system needs improving, it is unlikely that these students could navigate successfully through programs based on MOOCs and distance learning. Further, MOOCs alone will not create the well-rounded students that populate the offices that are the main employers of college graduates. The success stories that Carey cites are almost invariably based on talented kids with lots of initiative and willingness to take chances. These programs are not scalable, and MOOCs most likely will be incorporated into existing programs rather than bring about the end of college.