Last week, the New York Times’s Stanley Fish wrote an article arguing that there may not be a college cost crisis after all, citing the work of Robert Archibald and David Feldman. I’ll recap some of the basic arguments in a moment, but the readers basically called BS in the comments section, and loudly. Yesterday, Archibald and Feldman wrote a followup to address their critics. We get it, we get it. You’re trying to be contrarian, but quite frankly, I still call BS.
They raise three major points in their article (although presumably more in their book):
First, over time we have found ways to reduce the number of labor hours and kilowatts of power needed to produce most manufactured goods and agricultural products. By contrast, many services remain artisan-like…
Second, the upward trend of college cost has been accelerated by changes in income distribution over the last 30 years. People with high levels of education have seen big income gains…
Third, technology is a double-edged sword in many industries…We no longer employ typing pools. But other new techniques, like computer-aided design in architecture classes or pulsed lasers in physics labs, have increased cost.
Archibald and Feldman are claiming that the rise in tuition is entirely natural given the external (economic and technological) changes over the last 50 years, and that if you compare the cost of college to (say) dentists (which they do), you’ll notice a very similar trend. They also claim that the discount rate (university scholarships) absorbs some of the apparent rise in sticker price.
They’re right… but only to a point. That there are other exacerbating causes does not preclude the possibility that university tuition really has gone off the rails. In particular, their three main points are flawed.
- It’s not fair to compare the price of tuition to the CPI. After all, goods get cheaper, but labor doesn’t. They’re right, of course, but what they neglect to point out is that tuition (even net tuition after discounts) has increased dramatically when compared to other measures like median household income. A colleague of mine and I did a simple calculation based on the average sticker price of private universities, the median household income, and an increasing discount rate over the last 40 years.We found that in 1970, the effective average tuition was about 23% of a median household’s income, while it is 39% now.Archibald and Feldman don’t mention the comparison with median income explicitly. Instead, to argue affordability they make what seems to my mind (and apparently many of the other Times readers) to be a fairly unreasonable argument. They basically show that because real income increased over the last 20 years, even if University education costs more than ever, people still have more income after paying for it than ever before. Quite frankly, I find this absurd. In what context do we make our economic decisions on the purchase of one item based on the total amount of money we will have left over afterwards?
- Universities are labor intensive, and thus since university faculty are more educated than most, they will demand higher salaries over time. This is simply wrong. In real terms (compared to median income) faculty salaries have remained flat over 40 years. In 1970, the average faculty member at a US university earned $12,710 a year, approximate 146% the median household income. In 2010, the average faculty salary is $66,172, about 143% the median household income. In other words, properly normalized, salaries are flat.I’ll take this a step further. It’s not just the average over all ranks which is flat. The average Associate Professor salary is also flat (actually, somewhat decreasing), with the average Associate Professor earning 155% of the median in 1970, and 142% today.I’m not complaining about our salaries, but I will say that it is incorrect to point to skyrocketing faculty salaries as an explanation. If anything, there has been a definite trend away from tenured professors toward adjunct and teaching faculty over the last 20-30 years. Adjuncts instructors teach courses for far less (and teach far more of them per person) than those in the tenured line, which decreases the faculty expenses still further.Archibald and Feldman are right that there are a lot of highly paid people at universities. However, a better argument might be made that there has been an enormous proliferation of administration and staff over the past few decades, as well as a marked increase in recreational costs. My point is that these rapid increases do point to the possibility that there is an actual crisis in cost that can be dealt with.
- Are the costs technological? I have a $2000 desktop computer which I maintain myself, and which allows me to conduct my research, act as my own secretary, and communicate with colleagues far more cheaply than anyone could in a previous generation. This is fairly typical. As for teaching labs, the fraction of department teaching budgets which goes to lab equipment is incredibly small. Departmental teaching budgets as a whole represent the minority of university budgets. Further, experimental labs were taught 40 years ago and in real dollars, the costs were just as large. This argument is really a non-starter.
I’m troubled by the article and the argument because while it’s one thing to argued that tuition is a nuanced problem (and it is — I’ve looked into it, and there doesn’t seem to be a single cause or even a small group of them), I think it’s disingenuous to argue that there’s no problem at all.