Q: How is college like owner-occupied housing?

In light of the rising costs of college tuition around the country, coupled with the staggering amount of debt students have accumulated, studies are being done to examine a possible causal connection between expanding federal aid programs and tuition hikes.

…A:  When loans to purchase it are subsidized, the price of it goes up!

I’m fascinated by this story that appeared in the Wall Street Journal this week (by Josh Mitchell), with the subtitle “As Student Debt Grows, Possible Link Seen Between Federal Aid and Rising Tuition.”  The article explains the link this way:

Rising student debt levels and fresh academic research have brought greater scrutiny to the question of whether the federal government’s expanding student-aid programs are driving up college tuition.

Studies of the relationship between increasing aid and climbing prices at nonprofit four-year colleges found mixed results, ranging from no link to a strong causal connection. But fresh academic research supports the idea that student aid in the form of grants leads to higher prices at for-profit schools, a small segment of postsecondary education.

The new evidence?  Continuing into the WSJ story, my emphasis added:

The new study found that tuition at for-profit schools where students receive federal aid was 75% higher than at comparable for-profit schools whose students don’t receive any aid. Aid-eligible institutions need to be accredited by the Education Department, licensed by the state and meet other standards such as a maximum rate of default by students on federal loans.

The tuition difference was roughly equal to the average $3,390 a year in federal grants that students in the first group received, according to the National Bureau of Economic Research working paper by Claudia Goldin of Harvard University and Stephanie Riegg Cellini of George Washington University…

The authors said their findings lent “credence to the…hypothesis that aid-eligible institutions raise tuition to maximize aid.“

I see two main problems with that bottom-line conclusion/suggestion–with the strong caveat that I have not read the actual NBER working paper but only this WSJ story on it (and would love to hear your comments, especially if any of you have read the academic paper).

First, it’s tough to do the “all else constant” experiment here, given that “aid-eligible institutions need to be accredited.” That means being aid-eligible is strongly correlated with whatever qualities make the schools qualify for accreditation.  The Education Department’s website explains how accreditation is done:

The goal of accreditation is to ensure that education provided by institutions of higher education meets acceptable levels of quality. Accrediting agencies, which are private educational associations of regional or national scope, develop evaluation criteria and conduct peer evaluations to assess whether or not those criteria are met. Institutions and/or programs that request an agency’s evaluation and that meet an agency’s criteria are then “accredited” by that agency.

In other words, how “comparable” can two for-profit schools really be with each other, if one (the one providing aid) qualifies as accredited and the other (the one without aid) not?  The way prices work in most markets (for all kinds of goods and services), is that higher quality goods and services command higher demand and hence higher prices.  Take houses, for example.  You could compare two houses that are alike in many measurable attributes–square footage, lot size, school district, age, number of bedrooms and bathrooms, etc–and yet there would be many unmeasured attributes (and some unmeasurable qualities) that might explain why one house sells for a much higher price than the other.  In the case of college tuition, whatever variables the researchers controlled for in terms of “comparability,” we know they couldn’t control for the underlying factors that determined accreditation of the schools–because any of those factors were indistinguishable (inseparable) from the characteristic of whether the school was aid-eligible or not.  I suspect that if we were shown any pair of “comparable” schools in this experiment, we would conclude the aid-eligible school was of obviously higher quality than the aid-ineligible school (it was obvious to the accrediting agency, after all), and that that difference in quality was a reasonable enough reason why the higher quality school charged higher tuition.  So theoretically, the aid doesn’t have to have anything to do with what causes the higher price, and it might be that the only reason why higher prices seem correlated with greater aid is because higher quality is correlated with higher prices.

Second, even if the subsidized loans do have at least something to do with higher tuition and fees, this is just due to the forces of supply and demand–not the ulterior motives of rent-seeking colleges. The WSJ article about the NBER paper suggests that the empirical research provides evidence that colleges take advantage of their aid eligibility to set higher tuition prices, and/or manipulate their tuition charges to increase their aid eligibility.  But I suspect the hard-to-measure true story doesn’t contain quite so much willful drama on the part of the actor playing the college.  Instead, the causation probably runs this way:  subsidized loans lower the out-of-pocket costs of college to students, the lower price increases demand (shifts out the demand curve), and higher demand means a higher market price.  (Microeconomics students learn that the government subsidy acts like a “wedge” between the price received by the seller (the college)–now higher–and the price paid by the consumer (the student)–now lower.)  The higher price received by colleges after the subsidized aid is  not because the suppliers just on their own decide to claim most or all of the benefits of the subsidy by simply setting the price of tuition to reflect a full (or otherwise whimsically-decided) mark-up.

So more subsidized loans means greater demand for the college educations those loans buy, and along with all the other reasons why demand for college is increasing (like, no one can get a job now anyway, but especially not those without college degrees), this raises the market price of college, and, yes, the profits of those for-profit institutions of higher education fortunate enough to be deemed good enough for (simultaneously) accreditation and government-subsidized aid.

So this reminds me of how the government subsidizes home mortgages, via the mortgage interest deduction.  (Actually, technically, it’s from the combination of the non-taxation of the imputed rental services from owner-occupied housing and the mortgage interest deduction.)  Economists universally understand that the mortgage interest deduction raises the price of owner-occupied housing, because the value of the mortgage interest subsidy gets capitalized into the price of houses everywhere–even the prices of houses that are not literally purchased by people who take out mortgages and live in them as owner occupiers.  The point is that many players in the market for housing will qualify for the subsidy, and market prices will reflect how much of the market is made up of those people.  On the issue of the mortgage interest deduction, economists often worry about what the economic effects of that subsidy are, and whether policymakers really intended for those effects.  We can say that homeownership is a good thing, and we can hope that this subsidy to homeownership actually increases homeownership.  But it is a subsidy not to homeownership per se, but to the costs of borrowing to purchase a home–and the larger and/or more expensive the house (and the bigger the loan), the bigger the value of the subsidy.  (And on top of that, the higher ones income, the bigger the subsidy–because the subsidy is run through a tax deduction that makes the subsidy an “upside down” one, with larger percentage subsidies given to people in higher income tax brackets.)  Economists have found evidence that the mortgage interest deduction definitely raises housing prices, but not as much evidence that the subsidy increases homeownership as much as encourages people to buy more expensive houses with larger mortgages.  (For microeconomics students out there, both the income and the substitution effects work in those directions.)  And then we get to the public policy concern:  is that government subsidy worth its cost?  What is the goal of the policy?  If there are social benefits to homeownership, are there even bigger social benefits when people buy bigger houses (even if by going into bigger personal debt)?

Now come back to the student loan story.  College is like owner-occupied housing in that if you subsidize the costs of borrowing for college, you will raise the market price of a college education–just like the mortgage interest deduction raises the market price of a house.  Further, you might hope that what you’re doing with the subsidy is making it possible for kids to go to college who otherwise wouldn’t be able to afford it.  That might be partly the case, just like there are surely some households at the cusp of the owning-vs-renting decision for which the mortgage interest deduction is the marginal factor that makes owning the winner.  But just like the mortgage interest deduction goes to a lot of people who would own homes anyway and might just opt for the more expensive home made possible by a bigger mortgage, subsidized student loans go to a lot of students who would have gone to college anyway but may now opt to go to more expensive colleges made possible by larger (but subsidized) student loans.  And some of us might wonder if that policy effect is worth the cost, because government-subsidized educational aid costs real money (it increases government spending and increases our public debt), just like government-subsidized mortgages do.  Do all of us really want to be partially paying for kids (not even our own) to go to expensive colleges?

By the way, I speak about this effect of subsidized aid on the demand for expensive colleges, not just conceptually, but from personal experience.  I have two daughters in college, one at Princeton and the other at Sarah Lawrence.  If you google their costs, you will see they certainly both qualify as “expensive.”  But a combination of need-based grant aid and subsidized loans have way narrowed the difference between the net cost to send my daughters to those schools and what the costs of sending them to the in-state (Virginia) public universities would be.  They would not be attending those schools were it not for the availability of such aid, but without that aid, they would have still gone to college–just to a cheaper (in-state) one.

Based on just my personal experience (thus far), I think that more expensive colleges tend to be worth their higher costs.  Certainly they must be worth their higher costs if they continue to attract students, and I know many students go to these expensive colleges without any subsidized aid–because they can afford it and decide it is worth spending their own money that way over other ways.  More expensive homes are worth their prices, too, given that (or to the extent that) there are people who demand such expensive homes.  The public policy question about whether we want to be subsidizing expensive homes, however, seems a more damning criticism than asking the same question about subsidized college aid, because there’s probably more value added–to society, broadly–in encouraging higher-quality educations that happen to come with higher prices, than in encouraging higher-priced homes that are made possible with larger mortgages.  If the bigger, more important, public policy goal regarding higher education is to make it possible for more students to attend any college, however, then the government’s subsidized student loan policy might score poorly compared with an alternative policy of increasing need-based grants–in the same way that the mortgage interest deduction scores poorly compared with alternative policies that subsidize affordable housing if the goal is to increase homeownership.