Cracks in the Ivory Tower
The Moral Mess of Higher Education
I. Neither Gremlins Nor Poltergeists
Gremlins are mischievous little creatures with sharp teeth that like to break your stuff when you’re not looking. An even more nefarious creature is the poltergeist, an incorporeal spirit that terrorizes people by slamming doors and throwing objects. Throughout history and across different cultures, people have shown their propensity to believe in different unseen forces like spirits, demons, monsters and conspirators. If you’ve spent a few minutes reading the news, you know that people are also very willing to attribute bad outcomes to nefarious actors inside of institutions.
Cracks in the Ivory Tower takes a different approach to explaining the failures of higher education. Philosopher Jason Brennan of Georgetown University and Phillip Magness of the American Institute for Economic Research believe that the current failures of education are largely a consequence of bad incentives rather than gremlins and poltergeists. In this context, gremlins are corporeal individuals—such as the Koch brothers or George Soros—who are hurting higher education to achieve their own sinister ends. Poltergeists are intellectual movements such as neoliberalism or social justice. Despite being averse to attributing the state of higher education to intentional wickedness, the authors are extremely critical of the system:
From a business ethics standpoint, the average university makes Enron look pretty good. Universities’ problems are deep and fundamental: Most academic marketing is semi-fraudulent, grading is largely nonsense, students don’t study or learn much, students cheat frequently, liberal arts education fails because it presumes a false theory of learning, professors and administrators waste students’ money and time in order to line their own pockets, everyone engages in self-righteous moral grandstanding to disguise their selfish cronyism, professors pump out unemployable graduate students into oversaturated academic job markets for self-serving reasons, and so on.
To understand why things are so awful, we have to understand what the relevant actors—faculty, students and administration—want out of our education system and why they want it. The university is not a single entity that has a desire for creating knowledge or maximizing profit. It’s best thought of as a complicated network of actors with their own goals and motivations.
Above all else, the faculty would like to have a job, preferably the highly sought-after tenure track position. Many will end up settling for less desirable positions such as adjunct lecturer, a position with worse pay and more job instability. The people who want to secure a faculty position at a university are typically PhD graduates, who are hard-working and intelligent nerds who have a deep interest in their field of study and want to continue studying it for the rest of their life. These people love learning and knowledge, but they are also like the rest of us; they like money, fame and status.
Professors usually make good money. According to the American Association of University Professor’s 2017-2018 Annual Report, full professors averaged $104,280, associate professors averaged $81,274, and assistant professors averaged $70,791. There is a large amount of variation between disciplines, but becoming a professor is unlikely to be the most lucrative option for a person intelligent and conscientious enough to complete a PhD. Aspiring professors value the freedom to research topics that interest them without being told specifically what to research by a corporation or government. When professors devote a great deal of their life to a topic they find important, they like to receive some appreciation and recognition for their accomplishments.
Your average student is not so concerned with the love of knowledge or acquiring fame and status in an academic discipline. What students want is a credential that grants them access to more desirable and lucrative careers. They usually care about this credential even more than they care about learning. Students usually celebrate when the professor cancels class. Students want less academic work and more time to play while they are young, healthy, immature and in an environment that makes socializing extremely easy. We really shouldn’t be too surprised that students spend more time on leisure than they do on their academic work (Arum and Roksa, 2011, pp. 59, 120, 135).
What administrators want most of all is a promotion to get more money, prestige and power. To achieve these ends, they want to hire more people for their department. There is empirical support for the claim that—similarly to how firms can be seen as profit maximizers—university departments can be seen as budget maximizers (Kress, 1989; Ginsberg, 2011). To preserve their job, administrators want to appear not only useful but busy as well. They also prefer to have more non-tenured professors around to get fired before them in the case of a budget cut.
II. Negligent Advertising
A university’s wealth and status are a function of the students who enroll in the university. This gives administrators a strong incentive to make sure that they get students, especially talented ones, enrolled at their university. A good way to do this is with advertising. While companies like to explain why their product is the fastest, most effective and most beautiful, universities like to take a different approach. They go a lot further in claiming that you will be transformed into a newer, better and more enlightened version of yourself.
The Yale University website says that they are “committed to the idea of a liberal arts education through which students think and learn across disciplines, liberating or freeing the mind to its fullest potential.” The Princeton University website claims that “By exploring issues, ideas and methods across the humanities and the arts, and the natural and social sciences, you will learn to read critically, write cogently and think broadly.” At the time of writing, the James Madison University website claimed: “Engaged learning happens everywhere at JMU. Every discussion. Every lab. Every study group. Every team project. Every interaction with a professor.” Universities of various levels of prestige, all across the country make extremely bold claims about the powerful transformative effects of their courses.
Universities claim that liberal arts education has a myriad of extremely positive effects. Amherst College claims that “A liberal-arts education develops an individual’s potential for understanding possibilities, perceiving consequences, creating novel connections and making life-altering choices. It fosters innovative and critical thinking as well as strong writing and speaking skills.” At the time of writing, the Montclair State University website claimed that studying liberal arts “prepares you for a career by instilling … the intellectual skills of critical thinking, analysis of information, and effective expression of ideas.” At the time of writing, Park University claimed that “Liberal studies can, in fact, free a person from ignorance, prejudice and apathy. A liberal arts education equips students with an appreciation for critical inquiry and independent thought and reasoning.”
If you picked up a self-development book that said it’ll free you from “ignorance, prejudice and apathy,” you’d probably roll your eyes. These are seriously impressive claims. When it’s a $7.99 self-help book, it’s not such a big deal if some people are duped—but universities are tens of thousands and sometimes hundreds of thousands of dollars. The authors ask you to imagine a pharmaceutical company that made the same claims as universities all across the nation.
Imagine if the pharmaceutical giant Pfizer placed the following advertisement: Introducing Collegra! Collegra is a drug unlike any other. If you take Collegra 256 times a year for four years, Collegra will improve your critical reasoning, moral reasoning, analytic, and quantitative skills. It will transform you into a better person with a global mindset. It will make you able to face any challenge. It will prepare you for any job. It will dramatically improve your cognitive skills. It will make you score higher on standardized tests, such as the LSAT or GMAT. Furthermore, it will help you make more money! Indeed, Collegra users, on average, earn an extra million dollars of lifetime income compared to nonusers.
The cost of Collegra varies from person to person. Collegra is not covered by your insurance. Pfizer charges rich kids and foreign students on government grants $60,000 a year. But if you can’t afford Collegra, Pfizer may be able to help. Some satisfied previous Collegra users have generously provided us with funds to help new users.
Warning: Taking Collegra is more like undergoing chemotherapy than taking a pill. Users need to spend at least thirty hours a week for thirty weeks a year over four years for it to be effective. Most users will be unable to work at a job while taking Collegra.
Side effects include increased tendency to engage in binge drinking and to acquire tens of thousands of dollars in debt.
Now imagine that Pfizer has conducted no clinical trials, no randomized experiments, and no elaborate statistical analyses to tease out the effects of Collegra. And that the only evidence that they have is simple statistics comparing users of Collegra with non-users and that Pfizer is careful in ensuring that the only people who are allowed to take Collegra are those that are above average in the skills it is designed to boost. Would this sort of advertising be ethical? The US government wouldn’t deem it ethical because Pfizer would be engaging in negligent advertising:
Negligent advertising: Selling a product based on the claim that the product delivers certain benefits, despite lacking evidence that the product, in fact, delivers those benefits.
It would be much more understandable, although still not ethical, to give Pfizer a pass on Collegra if it only cost a few dollars. But in our hypothetical, Collegra is tens of thousands of dollars. If a university wants to make claims like Pfizer is making, they need to do their due diligence in ensuring that what they are saying or implying in their advertisements is true. Universities have failed to do this. Differences in outcomes may actually be a result of what is called a selection effect.
If you want to claim that your intervention is effective at creating a change, you need a random sample. Selection effects occur when your sample is not actually random because of the process of obtaining your data. In this case, the more intelligent and conscientious high school students choose to attend college. An even more intelligent and conscientious subset finishes college. The fact that college graduates are successful after college doesn’t necessarily mean that college made them more successful.
The model of education which believes that education is mostly about giving you skills that make you more productive in the workplace, allowing you to earn a higher wage, is called the human capital model. The idea that businesses hire college graduates more mostly because they have the traits that are desirable regardless of college but just use the credential of education as evidence of the fact is called the signaling theory of education. The best defense of the signaling theory of education is arguably George Mason University economist Bryan Caplan’s The Case Against Education. Caplan argues that education is roughly 80 percent signaling of intelligence, conscientiousness and conformity.
An important argument for the signaling model is that students don’t see substantial improvements in their academic skills. In Academically Adrift: Limited Learning on College Campuses, Richard Arum and Joseph Roksa sought to figure out how much students were learning on campus. Brennan and Magness describe this book as a “treasure trove of depressing statistics.” Using a test called the Collegiate Learning Assessment (CLA), the authors evaluated the degree to which students improved in the “core outcomes espoused by all of higher education—critical thinking, analytical reasoning, problem solving, and writing,” which are the “broad competences that are mentioned in college”(Arum and Roksa, 2011, p. 21).
According to Arum and Ruska, “At least 45 percent of students in [their] sample did not demonstrate any statistically significant improvement in CLA performance during the first two years of college” (Arum and Roksa, 2011, p. 121). Of the students who did see an improvement, the gains were modest. Even if students are able to successfully increase their knowledge and skills, there is little reason to think this will last very long. Memory and skills decay quickly without reinforcement. The authors described their findings:
[M]any students are not only failing to complete educational credentials, they are also not learning much, even when they persist through higher education. In general, as we have shown, undergraduates are barely improving their CLA- measured skills in critical thinking, complex reasoning, and writing during their first two years of college. Even more disturbingly, almost half are demonstrating no appreciable gain in these skills between the beginning of their freshman year and the end of their sophomore year.
In addition to limited growth, learning in higher education is also unequal. Students from less educated families and racial/ethnic minority groups have low levels of skills in critical thinking, complex reasoning, and writing (as measured by the CLA) when they enter college. These inequalities are largely preserved—or in the case of African-American students, exacerbated—as students progress on their journey through higher education (Arum and Roksa, 2011, p. 54).
Some argue that rather than giving students knowledge about liberal arts specifically, teaching students in humanities teaches students how to think. It develops critical thinking skills generally through a process called transfer of learning. Caplan explains that over a century of study has gone into transfer of learning and psychologists’ chief discovery has been “Education is narrow. As a rule, students only learn the material you specifically teach them ...if you’re lucky” (Caplan, 2018, p. 50). Furthermore, according to psychologist Robert Haskell:
Despite the importance of transfer of learning, research findings over the past nine decades clearly show that as individuals, and as educational institutions, we have failed to achieve transfer of learning on any significant level (Haskell, 2001, xiii).
According to psychologist Douglas Detterman:
Transfer has been studied since the turn of the [twentieth] century. Still, there is very little empirical evidence showing meaningful transfer to occur and much less evidence for showing it under experimental control (Deterrman, 1993, p. 21).
Terry Hyland and Steve Johnson believe:
On the basis of the available evidence, however, drawn from many very different disciplines, we believe that the pursuit of [general transferable] skills is a chimera- hunt, an expensive and disastrous exercise in futility (Hyland and Johnson, 2006).
It appears that the claims made by many universities in their advertising are false or misleading. At the bare minimum, it would appear there is insufficient support for the empirical claims that are made. Universities have not done their due diligence to verify the veracity of their claims, but they continue to profit from misleading advertisements.
III. Poor Measures: Student Evaluations and Grades
Another moral failure of higher education is the use of inadequate and misleading measures. Colleges make decisions, including promotion and hiring, based on student evaluations of teaching (SETs) which they should know are not adequate measures of teaching proficiency. Nearly all universities use these tests (Emery, Kramer, and Tian, 2001, p. 37). If you are going to make important decisions that impact your employees’ lives, it is very important to make them based on legitimate criteria.
Student teaching evaluations present questions about the performance of a teacher with verbal descriptions. These verbal descriptions are converted to a number to be averaged across a class, but we don’t know exactly how to interpret these numerical values. Regardless, a recent meta-analysis concluded that “[S]tudents do not learn more from professors with higher student evaluation of teaching (SET) ratings” (Uttl, White, and Wong Gonzalez, 2017, p. 40). The authors write:
[W]hen [previous analyses that apparently showed a large effect size] include both multisection studies with and without prior learning/ability controls, the estimate SET/learning correlations are very weak with SET ratings accounting for up to 1% of variance in learning/achievement measures. . . . [W]hen only those multisection studies that controlled for prior learning/achievement are included in the analyses, the SET/learning correlations are not significantly different from zero (Uttl et al., 2017, p. 38).
In another comprehensive meta-analysis of forty-two studies that try to measure student learning and student evaluations, D. E. Clayson does find a positive correlation, but it was not statistically different from zero (Clayson, 2009, pp. 21-22).
As statistical sophistication has increased over time, the reported learning/SET relationship has generally become more negative. Almost 40 years have passed since the positive result in Sullivan and Skanes’ (1974) study was obtained. No study could be found after 1990 that showed a positive significant relationship between learning and the SET (Clayson, 2009, p. 28).
If SETs are not measuring teaching quality, then what are they measuring? In the “Dr. Fox” experiment, researchers had a charismatic lecturer discuss a nonsensical lecture entitled “Mathematical Game Theory and Its Application to Physical Education.” Participants rated the lecture highly (Naftulin, Ware, and Donnelly, 1973). Later research corroborates the idea that highly entertaining professors who tell jokes receive higher marks (Granzin and Painter, 1973). Unsurprisingly, charisma is an important factor in student evaluations (Murray, 1983). The evidence suggests that the SET is more about personality than it is about how well a professor is teaching the students.
Grade point average (GPA) is another measure that suffers from similar problems. Some professors grade on a curve—a relative position compared to students—while some professors measure without regard to other students’ scores. But just like SETs, grades suffer from ambiguity in the measure. Different professors mean different things when they give a specific grade and different professors will evaluate students differently. Like students, professors will be subject to certain biases unrelated to the quality of the work. Jeffrey Schinske and Kimberly Tanner describe the research on consistency:
[Educational psychologist W. C.] Eells investigated the consistency of individual teachers’ grading by asking 61 teachers to grade the same history and geography papers twice—the second time 11 wk after the first. He concluded that “variability of grading is about as great in the same individual as in groups of different individuals” and that, after analysis of reliability coefficients, assignment of scores amounted to “little better than sheer guesses.” Similar problems in marking reliability have been observed in higher education environments, although the degree of reliability varies dramatically, likely due to differences in instructor training, assessment type, grading system, and specific topic assessed. Factors that occasionally influence an instructor’s scoring of written work include the penmanship of the author, sex of the author, ethnicity of the author, level of experience of the instructor, order in which the papers are reviewed, and even the attractiveness of the author. (Schinske and Tanner, 2014)
Students can receive or lose scholarships and jobs on the basis of grades that are not consistent in their meaning. Students attend different universities with different professors and take different classes with different levels of difficulty and are often graded on relative scales with different students. These different classes with different professors are then aggregated into one single number which is supposed to be representative of a student’s academic performance. Employment, scholarship and educational opportunities are made based on this measure.
A further complicating issue is that of grade inflation. The authors expected to find evidence of grading becoming easier, but that is not what they found. To prove there is grade inflation, one needs to show that GPAs have been increasing over time faster than the quality of students’ work. And yet, this has not been done. The authors are concerned that decisions about student quality are made based on a widely held belief that grade inflation is indeed happening, but there isn’t empirical evidence to support this.
IV. Disguised Self-Interest
There was a website called "Who’s Driving You? which documented safety incidents with Uber and Lyft drivers. It would appear that someone was documenting these incidents as a public awareness campaign out of an interest in the greater good. It wasn’t the case. It was run by a group called the Taxicab, Limousine, & Paratransit Association (TLPA). We can be skeptical of their motives especially given that they did not provide comparative data for taxis. It seems the goal was to preserve the economic interest of the entrenched businesses.
Kevin Simler and Robin Hanson explain in their book The Elephant in the Brain that humans have a strong tendency to engage in self-deception about their own motivations. Even in the case of charity, people often give opportunistically; they give when they will be recognized for their giving and in a less than optimal way because it makes them feel good. The effective altruism movement is a response to this widespread tendency to give less than optimally. It’s possible the TLPA was concerned about safety, but perhaps they were deluding themselves to justify their actions. Maybe they didn’t care about the truth at all.
Examples of self-interest disguising itself are everywhere. Brennan and Magness believe that there is a general education hustle; the required courses for undergraduate students are required because they benefit the departments teaching them rather than because they benefit the students taking the courses. For an illustration of the problem, imagine that the American Medical Association wants to reduce the widespread poor handwriting of doctors. A good justification would be that poor handwriting results in misreported vital statistics and incorrect dosages of drugs (Cheeseman and Boon, 2001; Lyons et al., 1998; Rodríguez-Vera et al.,1995; Goldsmith, 1976).
Imagine that the AMA requires a one-semester calligraphy course to obtain a medical license. As a result, universities begin offering calligraphy courses and opening calligraphy departments, creating a large demand for calligraphy professors. This large demand for calligraphy professors drives demand for MFA programs in calligraphy, which calligraphy professors lobby for in order to expand their departments and funding.
However, after all of the penmanship courses, there is no discernible difference between physicians’ handwriting and the issue of poor penmanship persists. Some call for the calligraphy requirement to be cut, but the entire structure has already entrenched itself. Many professors and administrators have jobs that are contingent on the requirement, and so they exert a great deal of pressure to continue the program. In fact, they suggest that a single calligraphy course is insufficient. It is necessary to add another to see an improvement. These faculty would be engaging in what is called rent-seeking—they are trying to rig the rules of the game to give themselves a special advantage at the expense of others.
There are certain disciplines—notably English—in which the number of PhDs is growing faster than the academic market can absorb. Many graduates will be left without employment. Since administrators and faculty like having more people in their department and students want jobs, there is pressure to expand English departments and other departments with an excess of PhDs by creating more general education requirements. These requirements are supposed to be in the self-interest of the students, but they mainly serve the interest of the faculty.
Recall the study by sociologists Richard Arum and Joseph Roksa on student learning. They studied student learning during the first two years of college—the time in which the writing requirements are most commonly taken. The authors didn’t find a meaningful change in critical thinking, complex reasoning or writing skill. In these categories, the average student in the fall of 2005 showed a 7 percent increase by the end of the spring of 2007. The gains were negligible (Arum and Roksa, 2011, pp. 35– 36). A 2008 survey found “very little rigorous research analyzing [remedial college education’s] effectiveness” (Levin and Calgano, 2008).
Another example of a general education requirement is foreign language requirements. Unfortunately, among American adults, fewer than 2.5 percent claim that they can speak a foreign language “very well” or “well” because of their coursework (Caplan, 2018, 48). And it should be noted that people likely have an inflated sense of their ability to speak a foreign language (Caplan, 2018, p. 48).
Another argument would be that it is beneficial to expose students to topics because the students might take a serious interest in one of them and consider pursuing that field as a career. Because general education requirements are mostly humanities classes, the extent to which you think this is good might depend on how favorably you view a humanities major compared to science and technology majors. There are likely better ways of exposing students to a variety of subjects. For example, 32 single-week courses across different disciplines might be a more efficient way to expose students to a larger number of subjects. This approach may achieve similar ends with less of an opportunity cost in terms of money and time. Any time spent in an uninteresting and unessential general education course is time not spent in a student’s major or taking electives.
Faculty that benefit from general education requirements could be well-meaning and truly believe that they are doing some good. But for reasons of power, prestige, and money, they are incentivized to manipulate the system to get students into their department. When the courses fail to have the desired benefit, they can petition for more courses.
Just as departments want to grow with more undergraduate students, departments also want talented PhD students who are going to achieve tenure-track roles in academia. The individual PhD student is an odd case; they are typically very intelligent and hardworking people who want to take a huge chance for the ability to research what interests them. Many students are likely overly optimistic about their likelihood of getting a professorship. While graduate students often know the market is bad, it’s either not easy or straightforward to figure out exactly how bad the market is and estimate their chances would be of getting a professorship.
Many graduate departments do not keep, let alone publicize, data about how their students perform on the academic job market. They often list recent graduates who found a job, but fail to mention the students who did not. They often mention students’ current positions, but fail to mention how long it took them to get those full-time jobs. They often list where students got jobs, but neglect to mention whether such jobs are tenure-track, full-time but not tenure-track, short-term visiting, or mere adjuncting.
Faculty want to advise PhD students and spread their ideas through them. Teaching PhD students is more fun than undergraduates because they usually care much more about the topic and the professor likely finds teaching higher-level courses more interesting. Graduate students can assist professors with research and more annoying tasks like grading papers or holding office hours. Adding graduate programs can justify increasing salaries, hiring more faculty, and creating lecture positions that allow more time for research. Faculty has an incentive to keep their poor-placing departments alive and administrators benefit from larger departments in terms of budget and pay.
We can blame the students and say that it’s not the professor’s responsibility to ensure they get a job or know their chances. Or we can blame the faculty and administrators; students don’t know their chances and the faculty is supposed to mentor them. One could argue it is wrong to take on students who will almost certainly fail. Regardless of who is to blame, the system will perpetuate itself because of the poor incentives in place. Departments will continue trying to expand and get more graduate students, even if it is not worthwhile for the students.
V. Cheaters, Cheaters Everywhere
Undergraduates cheat often; probably more than you would expect. Most cheat at least once and many, about 20 to 40 percent, cheat repeatedly (McCabe, Butterfield, and Treviño, 2017, p. 58; Lang, 2013, p. 15–16). Some writers have attributed this to “consumer mentality” and “outcomes-oriented neoliberalism.” The authors don’t want to attribute things to these sorts of poltergeists. They believe that students cheat because “they’re bad at time management, because they expect it to work, because they reasonably expect to not get caught, and because they believe their peers also cheat.” When a student’s classmates cheat, it incentives them to cheat.
Since nowhere near all of the students who cheat are caught, a lot of data on cheating is self-reported. Surveys of this kind suffer from what is called “social desirability bias”—people will either lie or delude themselves into thinking that they cheat less than they do. This is another example of the elephant in the brain. We would expect that surveys would give a lower bound with the true number being far higher.
The most comprehensive and controlled surveys on cheating come from William Bowers’s 1962–1963 work, along with Donald McCabe and his coauthors’ subsequent replications. Here, we’ll present the summary McCabe, Butterfield, and Treviño offer in their most recent book (McCabe et al., 2017, p. 58).
Table 9.1 displays the percent of students at universities that lack a strong honor code system who admit to having engaged in various forms of academic dishonesty at least once in their undergraduate careers.
Students with lower GPAs and students with very high GPAs are more inclined to cheat (McCabe et al., 2017, pp. 83–84). Students in professional schools and preprofessional majors cheat more than students in liberal arts, social science or science majors (McCabe et al., 2017, pp. 161). What makes these students cheat? James Lang argues that there are four basic factors:
1. A heavy emphasis on performance: The more students are pressured to excel, the more they cheat.
2. High stakes: When individual projects, papers, and tests are worth a high percentage of a grade, students cheat more. Students cheat more when given one large assignment rather than many small assignments.
3. Extrinsic motivation: If students don’t care about the skill or knowledge being tested for its own sake, but instead just need the grade to pass the class, they cheat more. The more they care about grades for their own sake, the more they cheat.
4. Low expectation of success: Students who think they need to cheat to do well cheat more (Lang, 2013, p. 35).
It’s very tempting to cast judgment on students that cheat, but if almost everyone else is cheating, then going with the group appears rational and becomes more morally ambiguous. Of course, it depends on how grades are determined. Many classes grade on a curve. By not cheating—a student may end up making the grade unjust because others were afforded an advantage. From a game theory perspective, we could see how this could result in rampant cheating; one student cheats cause he thinks others are cheating, and so other students get the impression that students are cheating, and they decide to cheat as well, and so on.
To illustrate with the most dramatic example, imagine a course graded on a curve rather than an absolute scale. The curve turns grades into zero-sum competition among the students. One student’s grade can rise only if another student’s grade falls. Suppose Kevin has no choice but to take this course. (Perhaps the course is a general education requirement, or perhaps it is now too late for him to drop it.) Imagine that cheating is widespread in this particular class. In fact, every other student, except Kevin, cheats. The professor and the administration are unable or unwilling to stop the cheating. Suppose also that cheaters routinely do better gradewise than noncheaters.
Suppose Kevin knows that if he tries to keep his hands clean and just does his best without cheating, his grade will suffer. The cheaters— the kids with the crib sheets— have an advantage over the noncheaters, just as defectors in the prisoner’s dilemma get an advantage over the cooperators. Since the course is curved, the better others do, the worse Kevin does. The curve means their cheating does not simply improve their grades, it also lowers Kevin’s grade. Suppose in a classroom devoid of cheating, Kevin would earn a B, but in this cheater-inundated class, he will earn at most a C unless he also cheats.
In this case, it’s not clear if we should blame or condemn Kevin should he choose to cheat. Through no fault of his own, he is stuck inside a corrupt academic system. He faces a prisoner’s dilemma where he knows ahead of time the other players are going to cheat (Table 9.3).
In this class, if Kevin cheats, he does not thereby disadvantage or take advantage of competing students. He does not impose harm on innocent people. Instead, he levels the playing field. Kevin cheats in self-defense. Kevin does not take unfair advantage of others. Rather, when he cheats, his cheating brings the world closer to the outcome that would have resulted if others had acted rightly. The other students threaten him with their cheating behavior, and by cheating, Kevin prevents them from harming him.
When others are playing by the rules, we have good moral reasons to play by the rules as well. However, if others are committed to not playing by the rules, and if playing by the rules then just makes us suckers who will be exploited, then we sometimes lose our obligation to play by the rules. We could not demand that Kevin remain honest when, through no fault of his own, he is stuck in a situation where honesty will just make him a victim of dishonest people. Kevin has a moral duty to be honest to those who deserve his honesty, not an absolute moral duty to be honest no matter what.
We would not go further and argue that Kevin has a duty to cheat, or that it would be wrong for him not to cheat. In fact, it may be admirable and praiseworthy for Kevin to take the high ground and refuse to cheat, despite the costs, simply because he is not the kind of person who cheats on a test. If Kevin takes a stand against cheating, bearing all the costs, he might deserve our admiration. Our point here is just that if he decides to cheat like the others, he does not deserve blame.
This is an extreme case, but the more that reality looks like Kevin’s case, the more excusable it is to cheat. The more that a grade is relative and others are cheating, the more it is like self-defense and not hurting other innocents. Some have critiqued curves for the reason that it incentivizes the failure of other students. We can’t be certain how like Kevin’s case the average student is. Many students claim that they do not cheat but believe that others engage in cheating at a rate 200% - 1000% higher (Fish and Hura 2013) than the students claim. Other studies have similar findings (Greene and Saxe 1992). Students should not have to worry about their colleagues cheating and professors should take more steps to prevent this sort of behavior.
VI. Who is to Blame?
Many attribute the problems of higher education to three powerful forces: the corporatization of higher education, neoliberal ideology and the threat of impending technological disruption. The corporatization narrative says that the bad outcomes in higher education are a result of profit-seeking behavior.
The theory of the corporate poltergeist purports to explain university spending priorities, hiring practices, student recruitment, curricular decisions, campus amenities, and above all administrative growth (Tuchman, 2009).
This is a misunderstanding of incentives and institutions. Universities function more like government agencies than corporations. The breakdown of the profit mechanism comes from three aspects of university operations (Buchanan and Devletoglou, 1970). First, the consumers are not the direct purchasers; there is public aid, scholarships, student loans, parents and subsidies. Students lack as much incentive to search for the best deal and the pricing is not so clear-cut regardless.
Second, the producers are not the sellers. The faculty depends on administrators for a great deal of the work including recruiting students, compliance with regulations, university finances and operating facilities. This lessens the incentive to prioritize student education because professors want to research rather than teach. Third, the university is not owned by faculty or administrators. Public universities are owned by the state and private colleges are run by the board of trustees. Rather than behaving like a profit-maximizing corporation, universities behave like government agencies with the same issues: “budgetary misallocation and waste, growing or bloated administrative staffs, redundant functions, inefficiency, and poor customer service.”
Another poltergeist is neoliberalism. One adherent to the viewpoint that neoliberalism is ruining education, critical theorist Henry Giroux, describes the problem:
Unapologetic in its implementation of austerity measures that cause massive amounts of human hardship and suffering, neoliberal capitalism consolidates class power on the backs of young people, workers, and others marginalized by class, race, and ethnicity. . . . [N]eoliberalism has wrested itself free of any regulatory controls while at the same time removing economics from any consideration of social costs, ethics, or social responsibility. Since the economic collapse of 2008– 2009, it has become increasingly evident that neoliberalism’s only imperatives are profits and growing investments in global power structures unmoored from any form of accountable, democratic governance (Giroux, 2014, p. 156).
What is meant by neoliberalism isn’t always so clear. People don’t typically describe themselves as neoliberal and it seems to encompass an extremely large array of beliefs. If we don’t have a clear definition, it’s hard to rigorously study trends in neoliberal ideology. Brennan and Magnus argue that if neoliberal ideology is a major issue in higher education we should be able to do the following:
1. Identify who the neoliberals are.
2. Determine exactly when neoliberals first appeared and track their growth, especially in positions of power.
3. Show that bad events started to happen at higher rates when the neoliberals appeared or when neoliberal ideology took over.
4. Find some causal mechanism that explains how the neoliberals brought about those bad events.
It would appear there are more people decrying neoliberalism than there are people describing themselves as neoliberal. Academia is famously left-leaning and has been shifting leftward over the past twenty years. Since 2000, the proportion of conservatives and moderates has been declining.
One possible argument is that the influence is coming from the administration, but this lacks evidence as well. Evidence actually points to administrators being more left of center. In 2018, political scientist Samuel Abrams published an article in the New York Times entitled “Think Professors Are Liberal? Try School Administrators” in which he describes just how left-leaning the administration is:
Intrigued by this phenomenon, I recently surveyed a nationally representative sample of roughly 900 “student-facing” administrators — those whose work concerns the quality and character of a student’s experience on campus. I found that liberal staff members outnumber their conservative counterparts by the astonishing ratio of 12-to-one. Only 6 percent of campus administrators identified as conservative to some degree, while 71 percent classified themselves as liberal or very liberal. It’s no wonder so much of the nonacademic programming on college campuses is politically one-sided.
The 12-to-one ratio of liberal to conservative college administrators makes them the most left-leaning group on campus. In previous research, I found that academic faculty report a six-to-one ratio of liberal to conservative professors. Incoming first-year students, by contrast, reported less than a two-to-one ratio of liberals to conservatives, according to a 2016 finding by the Higher Education Research Institute. It appears that a fairly liberal student body is being taught by a very liberal professoriate — and socialized by an incredibly liberal group of administrators.
A further issue with the neoliberal claim is that university degrees aren’t optimized for industry jobs. The idea is that there is a disregard for the humanities in order to emphasize technology and science to cater to the profit-seeking mindset. Recall the previous discussion regarding the general education hustle; certain majors, especially humanities, are able to effectively lobby for general education requirements. In chapter 8 of the book, the authors explain how English has seen a growth in faculty despite a lack of growth in demand for the major.
Another poltergeists that the authors reject is the coming tide of technological change which will fundamentally restructure higher education. There is a great deal of educational materials available for free to almost anyone with an internet connection. In fact, there are online courses available that are often created by universities called massive open online courses (MOOCs), notably MIT open courseware. Why can’t you just get a degree from MIT without ever paying? Universities should be scared if they are proponents of the human capital model of education. People will be able to get content from the best universities completely free and so they will no longer need to attend university. But it’s likely that universities know that the credential is extremely important.
VII. Answering Taxpayers
A lot of money is being spent on higher education. Over $600 billion is spent on operations. It would be one thing if governments were a poorly run business, but they also receive government funding. In fact, a great deal of this is being covered by government funding. At the time of writing, colleges obtained about 37 percent of funding from the government.
This doesn’t account for all the public services that universities benefit from without paying taxes such as police protection, fire protection and road maintenance. It is important to use public funding effectively. A dollar that is not spent on education could be spent on “medicine, art, music, aid to women with dependent children, police protection, housing the homeless or some other cause.” Proponents of effective altruism will likely think of causes such as malaria nets, reducing existential risks or preventing extreme poverty.
For instance, left-wing economist Jeffrey Sachs argues in The End of Poverty that rich first world countries could rid the world of extreme poverty with an additional $60 billion to $130 billion in aid funding per year for the next decade or so (Sachs 2005). Maybe he’s wrong and it’s not that easy. But, if Sachs were right, we’d have to ask: Why not cut US spending on higher ed by $60 billion and instead use that money to end world poverty?
A further issue is federal student loans. Since students repay loans, it might not be fair to include them in the calculation. But there are opportunity costs to giving students loans. Even if it was economically beneficial, an unfortunate side-effect is that it leads to tuition increases. A recent study found a 60 cent increase in tuition cost for every one dollar increase in tuition, while another found that subsidized loans may crowd out enrollment that it wants to promote.
A possible justification for subsidizing a good is that it is a public good; it is non-rivalrous and non-excludable. Non-rivalrous means that consuming the good does not come at the expense of others and non-excludable means you cannot exclude people from using the good. The reality is that universities are neither of these things. A better argument would be that there are spillover effects—or “positive externalities”—to education that are greater than the cost of subsidizing it.
This kind of argument might well be sound. But whoever asserts it bears a burden of proof. That person would need to identify what the positive spillovers are, demonstrate that there really are such spillovers, identify how valuable those spillovers are, and then determine how much governments should spend to create those spillovers. “Universities are good for society, so let’s pay more” isn’t even the beginning of a serious argument. Maybe universities are good for society, but right now we’re spending $10 in subsidies for every additional $1 of social value. Maybe it’s the other way around. We’d need to check. There is an optimal level of public investment in higher education, and we don’t know a priori whether we’re below or above that level.
A generally good principle is that you bear the burden of presenting evidence if you want to divert taxpayer money toward some cause. There should be some evidence that whatever intervention you propose is going to work or is worth the cost. This is rarely what happens. People claim causal relationships without sufficient evidence and then use this as a justification for spending because it seems intuitively true.
For example, people believe that universities can foster an interest in high culture. But an important question is to what degree and at what cost. The argument for enlightenment from liberal arts is addressed in Bryan Caplan’s The Case Against Education. He argues that the evidence is weak that education is useful for preserving appreciation for high culture (Caplan, 2018, 238–261). Another example is the idea that universities are “museums of ideas”, but to justify spending on this basis you need to show it is worth the cost of maintaining this museum and this actually works in making people appreciate the ideas. There are lots of ways to expose people to ideas; avid Astral Codex Ten readers have been exposed to plenty of new and interesting ideas online without the help of a university.
Another potential justification for funding is that it provides a positive economic service by training future employees. The major issue is that a lot of information that is taught is not relevant for the job market, soft skills are not given to students, and there is little support for the idea of transfer of learning. If education is signaling, should signaling be subsidized to help employers find future employees? Likely not. Why should the government pay hundreds of billions to reduce employee search costs? Furthermore, subsidizing the signaling game makes it more destructive in terms of time and money. It even increases search costs.
To illustrate what we mean, imagine the manager at Chotchkies Restaurant wishes to assess which of his waiters are the most enthusiastic and perseverant. Suppose he can’t easily measure this directly, so he uses a proxy: the amount of “flair” his waiters wear. If the average waiter has three pieces of flair, then he can distinguish himself by wearing four pieces. But, because this is true for everyone, four pieces of flair will become the new normal, so waiters looking to demonstrate their superiority will wear five. And so on. Three years later, you find all the waiters wearing at least twenty pieces. The manager’s ability to distinguish the best from the rest hasn’t changed—it doesn’t matter to him whether the average number of pieces is three or twenty—but the waiters are engaging in progressively more expensive behavior to achieve the same distinction. Search costs have increased dramatically, but the system is no more efficient at separating the wheat from the chaff.
Aside from educating students, an essential function of education is knowledge creation. Universities definitely create knowledge, but to justify subsidizing research, we need to determine if it is worthwhile. Surely some forms of research are less useful. It’s difficult to justify expenditure on “twelfth-century, avant-garde Romanian poetry.” And even if we have some particularly worthwhile research, we need to evaluate whether professors are producing enough of it to justify their salary.
The HERI survey also contains data about how many total pieces faculty have published over their careers. In fact, 63.2 percent have never published a book, 44.9 percent have never published a chapter in an edited volume, and 17.2 percent have never published a peer-reviewed article. Only about 21 percent have published 21 or more articles in their careers, while about 70 percent have published 10 or fewer articles in their careers. Basic arithmetic tells us that the 20 percent of faculty who have published at least 21 articles have, at bare minimum, published as a group (over their careers) almost twice what the remaining 80 percent of faculty have collectively published over their entire careers.
Although some research is unquestionably worth the return, we need to take a more critical look at the return on money spent subsidizing research to determine what is economically worthwhile to fund. Imagine that universities are not enlightening, good job training or producing enough worthwhile research. One could argue that they are nevertheless essential for democracy to function. One proponent of this is Martha Nussbaum who gives a wishlist of skills citizens need:
the ability to think well about issues affecting the nation, to examine, reflect, argue, and debate, deferring to neither tradition nor authority.
the ability to recognize fellow citizens as people with equal rights. . . .
the ability to have concern for the lives of others, to grasp what [effects policies have on others].
the ability to judge political leaders critically. . . .
the ability to think about the good of the nation as a whole. . . .
the ability to see one’s own nations . . . as part of a complicated world order in which issues of many kinds require intelligent transnational deliberation for their resolution (Nussbaum, 2016, 25–26).
We have reason to doubt that education imparts these skills to any substantial degree to citizens and that people will retain these skills for a very long time. Even if universities did give people these abilities, it wouldn’t justify all courses—only the ones that successfully imparted these abilities to students.
When is expenditure justified? Brennan and Magness propose a weak criterion called the bare minimum theory of just expenditure: “If a university takes government money for a particular purpose, society should generally expect to be better off, taking into account appropriate experimental risks.” As universities continue to increase in cost, it is becoming increasingly difficult to justify them. And yet people across the political spectrum want to fund them more. Government has a duty to spend money in a way that is not wasteful or socially harmful. It is a matter of justice. The authors can have the final word:
We’ve barely scratched the surface here. We’ve almost entirely ignored how faculty from all over the political spectrum use the university for political activism. We’ve ignored how some faculty use their status to sexually harass their students. We’ve ignored all the serious ethical problems that arise from universities running semi-professional sports teams. We’ve ignored whether universities act too paternalistically toward their students or, conversely, do too little to protect them from the dangers of college life. We’ve ignored whether admissions or hiring practices are fair or biased.
We admit these are important problems, many even more important. But we have two sets of reasons for avoiding these questions. First, some of them are inherently political, pitting the Left versus the Right. We wanted to discuss issues that people from all sides of the aisle can admit are real problems. Second, some of the other problems we just identified are, in a weird way, secondary rather than fundamental problems. Ethics issues plague college sports, but college sports are rather obviously a secondary function of the university at best. In contrast, gen eds are seen as a normal part of US and Canadian higher ed, but they seem to be mostly a way of exploiting students to the benefit of faculty in certain disciplines.
In addition, some of the problems we have not covered, though important, are relatively easy to solve. Consider this: Suppose, plausibly, that universities admit too few poor students. The reason: Poor students often had to work at low-paying jobs in high school and thus listed fewer or less impressive extracurricular activities on their résumés. Solution: We direct admissions officers to request income information and give extra weight to poorer students’ extracurriculars. But now consider the problem we discussed in Chapter 3: A liberal arts education presumes a false model of how students learn. How do you fix that?
Universities are a moral mess. The problem isn’t bad people. Universities are made up of decent people, if not angels. The problem is bad incentives. If you want to fix the problem, you need to change the incentives. Just how to change the incentives is itself a big problem— even we the authors have no incentive to change the bad incentives! But the bottom line is: You won’t fix what ails universities unless you fix the incentives. Good luck with that.
Subscriptions are deeply appreciated, and positive or substantive comments are welcome. You can subscribe for free by typing your email below or using RSS. Have a great day! 🙏🏻😁