Resource Vampires and the Cost of College

In the US, college debt is a substantial burden on young adults. This is not new information. 70% of college students are taking out loans and then graduating with (on average) $37,000 in debt. While not the calamity often claimed (we’re talking basically the price of a new Camry), student debt hasn’t always been this high. Someone who graduated in 1984 had an average debt of $5470. If that amount had just increased with inflation, it would be $16,507 today. So debt’s going up at 2.25x the rate of inflation, which tracks pretty closely with the rate that tuition is increasing. In other words, the rising cost of college has doubled the inflation rate since Gen X started going to college. What has changed in the last 40 years? Are universities buying new things, spending more money on old things, or what? Where exactly is all this money going and why?

One would think that the faculty and the library are central to the mission of any university. But the money is not going there. According to one source, between 1987 and 2012, inflation doubled the cost of everything (2.02x) and the average faculty salary also doubled (2.15x). Another source says that for full-time faculty, inflation-adjusted salaries have only gone up 9.5% since 1970. So the extra money is not going to the faculty; we’ve been paid basically the same inflation-adjusted salary for over 50 years. A similar story is true about academic libraries: library spending has stayed fairly constant in absolute dollars, even as its percentage of overall university expenditures has decreased rapidly.

Bowen’s revenue theory of cost says that universities are spendthrifts: they raise as much money as they can and then spend it all as fast as possible. Costs are going up because colleges are making it rain. But if Bowen is right, then why are some universities swimming in cash and others are closing their doors? We should all just be living paycheck to paycheck.

Baumol’s cost disease says that the labor productivity of university personnel can’t really be increased as it can in other industries (one professor can teach only so many students), but wages must rise relative to growth elsewhere in the economy anyway (or colleges can’t attract any job candidates). The problem here is cost disease predicts that all the excess revenue—i.e. revenue above the inflation rate—should be going to big faculty raises. But it is not.

I think both the Bowen and Baumol explanations are wrong. The real cost driver is resource vampires.

There’s a lot of small purchases that seem like a good idea, don’t cost very much, and you have to pay for them on a regular basis. Then they add up. Take streaming services for example. Each one has different offerings and is not much money: DirecTV ($155/mo), Amazon Prime ($15/mo), Hulu ($19/mo), Netflix ($23/mo), Disney+ ($16/mo), Starz ($11/mo), ESPN+ ($12/mo), BritBox ($9/mo), HBO ($20/mo) and Showtime ($12/mo). You think, eh, I want to see the big game, what’s $12 to get ESPN+? The kids want to see The Mandalorian, and Disney+ is only $16 so let’s do that too. Plus my spouse loves those British detective shows, and BritBox is just $9. The price of lunch. The next thing you know you’ve signed up for everything. The problem is that all those trifling costs add up to over $3500 a year, and that’s actual money. Who wouldn’t want a $3500 annual raise? All of those little services are resource vampires—each bite only sucks a little blood from your bank account, but by the end it is a dry and hollow husk.

I think the right move is to ask what about higher education has changed since Gen X started college. Here’s some ideas of resource vampires that didn’t exist (or barely existed) when Gen X began college, and certainly didn’t exist in the Baby Boomer days. To be clear, I’m not saying any of the expenses I’m about to list is a bad idea. Maybe they are all good ideas. I like having Netflix, Starz, HBO, and Hulu too. The key point is that they are resource vampires: they add up, are annual expenses, and suck more and more blood. 

Resource vampires

Example 1: IT/tech/computers. In the early 1980’s, the most computing around was a few personal computers in a campus lab, and a mainframe at the research schools. No wi-fi, no broadband, no internet to connect to, no desktop or laptop for every faculty member, no 4-year replacement cycle for hundreds of computers, no large staff of IT people to make sure everything is running right, no campus-wide software licenses for a dozen expensive programs. No course management software, advising software, and similar things that annually cost six or seven figures per program per campus.

Either IT/computing increases productivity in a way that generates more revenue for universities, or it does not and it is a drain against the bottom line. If it increases revenue by making administrators more efficient, then colleges will need fewer administrators to accomplish the same tasks. This is the opposite of what has happened: between 1976-2018, student enrollment increased by 78%, but full-time administrators and other professionals employed in US higher education increased by 164% and 452%, respectively. If IT/computing increases revenue by making faculty teaching more efficient, then we should expect to see the student: faculty ratio increasing over time, as fewer professors are needed to teach more students. Again, this is not what has happened. During the same time period, the number of full-time faculty employed at colleges and universities in the US has increased by 92%, which is in line with the 78% increase in students. Moreover, because there are more part-time and adjunct faculty over the same period, the percentage of full-time faculty has gone down from 67% to 54%. In short, in a 40-year span colleges added a few more full-time instructional faculty, many more adjunct instructors, and vastly more staff and administrators, all of which is the opposite of what we should expect if IT/computing were creating cost-saving efficiencies.

An alternative suggestion would be that computing has allowed admin and faculty to take on more tasks, improving the overall quality of universities, and increasing the value of college per dollar spent. Even if that is right, though, it still doesn’t explain the cost increase. For example, when the iPhone debuted in 2007, the basic model was $499. Adjusted for inflation, that is $760 today. The cost of a brand-new base model iPhone 16—a device vastly superior to the original iPhone in every way— is $799, just barely over inflation. Technological improvement tends to drive down costs relative to quality, not increase them, and it would be a difficult argument to make that the quality of today’s universities is analogous to an iPhone 16 whereas 1980s universities were an iPhone model 1. This evidence suggests that IT/computing is not helping lower college costs and is therefore a resource vampire.

Example 2: Social services. It used to be that college health services amounted to not much more than high school nurse level. Society has cut back on social services, particularly mental health care, and those things have gotten crazy expensive to pay for privately. So colleges now have to have counseling centers, psychologists on staff, and more comprehensive urgent care, flu shots, etc. to make up the difference. Lower income students have more stresses in their lives; over 40% of US college students suffer from food insecurity. No surprise they need more mental health care as well. When there is little social safety net, they turn to us.

Example 3: Title IX. Title IX was passed in 1972, and states "No person in the United States shall, on the basis of sex, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any education program or activity receiving Federal financial assistance.” Since then, especially since the late 1980s, the federal interpretation of it has expanded and expanded and now every university has at least one and probably several Title IX officers, all of whom are getting paid solid admin money. In addition, it is an ill-kept secret that men’s college sports generally lose money, but women’s sports always lose money (if there is an exception to this rule, I don’t know about it). Title IX, however, requires comparable commitment on scholarships, sports opportunities, and so on. So even if the University of Oklahoma, say, is making a profit on men’s football and basketball, they wind up spending it on money-losing women’s sports.

Example 4: Remedial courses, student support, residential life, general hand-holding. It is not just generational grousing to observe that today’s students are less self-reliant and less capable of practical problem-solving. Students need more help with everything, from understanding how to register for classes, to finding tutorial services, to figuring out the courses they need for graduation. All that means more non-faculty staffing, none of which brings in any money. One might argue that these services increase retention and therefore help the bottom line (which is probably true), but if they were not needed in the past and we need them now, then the past was cheaper.

Example 5: Amenities arms race. Your average 18-year-old is impressed by fancy dorms, excellent food at the commons, a lovely quad, a climbing wall at the gym, and great sports teams. They are less impressed by (or aware of) brilliant faculty, a giant library, state-of-the-art labs and other academic things. The majority of universities accept most applicants, so they are in competition with each other for students. Even the universities at the top of the food chain are competing with each other, like pro sports teams fighting to attract the best players. The upshot is that students up and down the line are going to expect and want amenities. What’s different now from the past? Arms races are by definition less bad in the past and worse in the future. The food at the Average State U. commons now is much, much swankier than the food at Fancy Pants College 40 years ago. At the same time, I expect Fancy Pants College chow in the 1980s was better than Average State U. food in the 1980s. 

Example 6: Accountability. Accountability is the opposite of trust. Society has become increasingly distrustful of experts and deprofessionalized. The result is that at least at public universities, there is increased pressure to prove that we’re doing what we say we are doing. Accrediting bodies decided that was a good idea too, for some unknown reason. Thus the rise of outcomes assessment. None of that existed when Gen X started college—everyone assumed that if the profs said that a student knew tensor calculus or Plato’s Theory of Forms or the Krebs Cycle that they actually did. But now we need assessment officers at the VP level, and all kinds of time and resources devoted to filling out forms to produce “data.” VPs with their own fiefdom need underlings to prove their worth, and the costs go up without any corresponding revenue.

Example 7: Diversity, equity, inclusion. DEI enthusiasm has exploded in the past few years, so rapidly and comprehensively that it has generated a (predictable) pushback. When DEI is written into the foundational mission statements of universities, the money is sure to follow. After Texas passed anti-DEI legislation, the University of Texas at Austin sacked over 60 employees and closed an entire division. UT Austin boasts of 3000 teaching faculty for 170 undergraduate programs, which means they had about three and a half times as many DEI employees as their average academic program has faculty. The University of California has a DEI Vice-Provost whose salary and benefits exceed $400,000, and she is hardly alone.

Come on free market, do your thing

After I wrote the above, I contacted Justin Downs, an economist at the University of Buffalo. I wanted to understand why the market wasn’t fixing the cost problem.¹ Here’s what he said.

The fact that these goods and services are being produced by someone is not shocking; each provide value to some members of a university’s ecosystem, its students, faculty, staff, and administrators, so of course there is a demand for them. However, it is not obvious that the efficient way for these goods and services to be produced is by a university itself rather than by firms who then sell those goods and services in markets where university members can purchase them directly. The economics of markets for public goods and insurance may help us understand why this is the case.

When we think carefully about what is actually being provided in the examples of Title IX, DEI initiatives, IT services like free wifi, and outcomes assessment and other accountability measures, we see that each convey some benefit that can be enjoyed by any one member of a university without preventing any other member from enjoying it as well. In this respect they are like state parks, community swimming pools, public roads, and bicycle lanes; everyone has access.

Goods with this characteristic are called non-rivalrous, an essential characteristic of the “textbook” definition of a public good. In that same textbook, one will also find a central conclusion of the analysis of markets for public goods, that free markets tend to produce an inefficiently small quantity or amount of them. This inefficiency occurs in markets because profit-driven producers will find it difficult or impossible to charge multiple buyers for non-rivalrous goods, which undermines their incentive to produce them in the first place and leads to insufficient supply.

Now, when we consider the “industrial organization” of the provision of these goods and services, it becomes less puzzling why universities are bearing the costs. There is no market arrangement that will achieve the goals things like Title IX and DEI initiatives achieve, and accountability and accreditation processes provide public goods in the form of information people can freely use to make employment and educational decisions. Although IT services and internet access are things that are available for purchase in markets, their “public” provision within a university allows for organizational thinking to be done under the assumption that everyone has access, which is obviously not the case in private markets. With either no market or a highly inefficient market available to provide these goods, universities expensively do it themselves, leading to rising tuition and fees.

Applying the same exercise to some other examples, such as social services, student support, etc., we see that these services have characteristics in common with those that people typically buy insurance for. They involve idiosyncratic risks to the individual that may not be completely foreseeable, the likelihood that these services are needed is often the private information of the individual, and the individuals most likely to need the services are also the ones that it is most costly to serve. Need is also unpredictable; someone might need tutoring in English but not in chemistry. The obvious analogy is health insurance. 5% of the population accounts for nearly half of all US health spending, but everyone needs insurance because medical problems are often random.

In markets, these characteristics lead to what economists refer to as an adverse selection problem; setting a price based on the average cost of providing the service drives the least costly buyers out the market, which increases the average cost of the service, which leads more low-cost buyers to leave, etc.  This process is sometimes called a death spiral and can result in a complete collapse of the market.  When it comes to healthcare, most developed countries deal with this phenomenon by nationalizing the healthcare system and funding it through taxation, which avoids the death spiral, even if it may not entirely avoid high costs.

OK, it may be a bit dramatic to use the term death spiral when thinking about why universities often provide free or cheap tutoring, or advisors that guide students in selecting courses and majors.  However, the way universities have taken to centralizing the provision of general social and students support services resembles the nationalization and bureaucratization of healthcare; service is provided to all and the costs are baked into tuition, which makes sure everyone has access to services they may otherwise struggle to obtain in markets, but raises tuition for everyone as well.

I’m not offering a comprehensive solution to the burgeoning cost of college. Identifying the problem, as I have done here, is the first step. Still, when we find other resource vampires latched on to our bank accounts and sucking them dry, that’s when we can ask if they are worth the cost. I want some streaming services, sure, but when is the last time I really watched Hulu? And now that the championship fight is over, is ESPN+ still worth it to me? Similarly we should start to look at the resource vampires attached to universities and ask ourselves, “maybe this has merit, but is it enough that we should keep paying for it?” The value of many of the vampires I’ve identified is hard to measure, and to make them truly legible and calculable we run the risk of forcing those values into a Procrustean Bed. Everyone wants the One Weird Trick to fix complex issues like this one. Of course there’s no such thing. Nevertheless, if we are to get serious about reining in the cost of college, we may need some garlic and a wooden stake.

1. Justin and I originally pitched a version of this piece to a variety of legacy media outlets without success. A cynical take is that left-leaning places don’t want to hear that stuff they like such as DEI and Title IX costs money, and right-leaning places don’t want to hear that there’s problems the market cannot solve.