Are construction costs driving up college tuition?
Andrew Martin has a very long, and not particularly illuminating, article about college indebtedness in today’s NYT. The title of the piece is “Colleges’ Debt Falls on Students After Construction Binges”, and it’s almost 3,000 words long, but somehow Martin fails to even hazard a guess as to the degree to which colleges’ debt is falling on students after construction binges. We’re certainly told that it’s happening:
A decade-long spending binge to build academic buildings, dormitories and recreational facilities — some of them inordinately lavish to attract students — has left colleges and universities saddled with large amounts of debt. Oftentimes, students are stuck picking up the bill…
Higher debt payments and other expenses have contributed to the runaway inflation of college costs, and the impact on students is real and often substantial.
How big are these bills? How substantial is the impact on students? Martin doesn’t hazard a guess: instead, he just says that “the costs are not easy to isolate”. But there are a few hard numbers, far down in the piece:
Outstanding debt at the 224 public universities rated by Moody’s grew to $122 billion in 2011, from $53 billion in inflation-adjusted dollars in 2000. At the 281 private universities rated by Moody’s, debt increased to $83 billion, from $40 billion, in that period. Rather than deplete their endowments, some colleges borrowed to help pay bills after the financial crisis, but most borrowing was for capital projects.
Since 2000, the amount paid in interest and principal has increased 67 percent at public institutions, to $9.3 billion in 2011, and it increased 62 percent at private ones, to $5 billion last year.
Martin doesn’t tell us what this works out at on a per-student basis, so let’s try. According to the Census Bureau (see Table 5), there are 20.4 million students enrolled in US colleges, split between 16.6 million undergrads and 3.8 million graduate students. According to Martin, using numbers from Moody’s, the amount of college-level debt being borne (in part) by those students has gone up by $112 billion, and the annual debt service has gone up by $5.6 billion. (These are numbers Martin could have just printed directly, but for whatever reason he chose not to; instead, you need to back them out of the numbers he cites.)
Students don’t bear all those extra costs: as Martin notes, “in some states, including New York, California and Connecticut, borrowing for public colleges and universities is mostly paid for by taxpayers, so students are not directly responsible for payments on the debt”. But for the sake of argument — and despite the fact that the University of California is the single biggest debtor, with SUNY at number 2 — let’s assume that all the extra costs are borne by students. In that case, we have 20.4 million students paying an extra $5.6 billion per year in interest, which comes to an annual cost of $274 per student.
Remember that $274 is a deliberate over-estimate, since a lot of the extra borrowing that Martin is writing about will get paid out of state budgets rather than out of students’ tuition fees. What’s more, the rise in interest payments coincides with a lot of universities shifting floating-rate debt to fixed-rate debt, which increases the interest payments but makes it them much less prone to rising unexpectedly.
Obviously, the increased costs will be higher at the universities with the most construction activity, and lower at more frugal colleges: the $274 is just an average. And I’m no fan of what Martin calls the Edifice Complex: I’ve been highly critical of capital projects at Harvard and NYU. But if it wants to make the case that students are paying “often substantial” sums as a result, the NYT is going to have to do better than this.
Indeed, if you want to criticize big capital projects, then “students end up paying a large part of the interest expense” is way down the list of good ways to do so. Interest expenses are generally small as a percentage of capital costs, because interest rates are low; what’s more, when you divide them between tens of thousands of students, the per-student cost becomes entirely manageable. The problem is more in the way that these projects force universities to lose a lot of flexibility in terms of their optimal size: it’s much easier to grow than to shrink, even as it’s hard to maintain quality when you’re growing too fast. The result, all too often, is shiny facilities, and a lower-quality education.
There’s a bigger lesson here, too, for the NYT. Martin says that the data underpinning this article was compiled for the NYT by Moody’s, which means that the NYT has access to a full and rich database. So why doesn’t it publish that data? Good data-driven journalism both publishes as much data as possible, and uses the data to drive conclusions, rather than simply dropping numbers into a foreordained article.
What should have happened here was for Martin to take a deep dive into Moody’s data, to try to work out which colleges saw the largest debt-service increase and whether there was any correlation between debt-service increases and tuition increases. Even if he didn’t have the appetite to do that work himself, at least if he published the data then the rest of us could do it. Instead, we just get an article which is very long on anecdote and very short on useful data. It’s a shame.