Marginal Utility

by Tom Bozzo

Stephen Karlson picked up and ran with the implicit challenge from yesterday's post. A mathematical singularity prevents me from calculating the resulting percentage increase in elevation of the conservative position in the debate over Lileks's original commentary. The title here is meant to keep the level of discourse bounded.

Prof. Karlson notes that he

...has burned a lot of neutrons on this site in defense of thethe somewhat crazier idea that higher education for the masses ought to be, well, higher education.

No argument there, though we observe that Lileks's neuron prefers to score cheap points at the expense of areas of higher ed that are soft targets for satire.

Karlson continues, offering some on- and off-topic objections. The latter focus on academic freedom.

Surely Mr (sic) Bozzo is not encouraging loyalty oaths for state university employees?

Surely not. He is as free to be a traitor to his profession as I am to be a traitor to my income tax bracket. For the time being, at least.

He stays on the academic freedom bandwagon for a while.

Or ought someone at the American Economic Review have recommended against publication of W. Lee Hansen's "Income Distribution Effects of Higher Education" ...because the knowledge that "higher education for the masses" was the slogan and "regressive transfer" was the reality was not something a beneficiary of the idea ought be making public?

Lee Hansen can research anything he wants to research (*) with my blessing.

Now, the reference to Hansen brings us to Karlson's substantive response to my main question, which is that Lileks's bleat is isomorphic to the question, "Why should college cost more?" He suggests that a legitimate reason is precisely the Hansen-Weisbrod research that purports to show that public higher ed subsidies are regressive. This is a potentially powerful appeal, since Marginal Utility tends to like policies that don't make society's rampaging inequality worse. That's why, for instance, we favor policies such as continued taxation of large estates. (**)

The Hansen-Weisbrod research isn't totally compelling, though. Thomas Kane's calculations from his 1999 book on higher ed finance show that (1) low-income students receive larger subsidies, conditional on attendance, but (2) low-income students enroll at lower rates and don't stay in school as long, so (3) they end up with about half the gross subsidy of high-income students. Which is to say, unequal opportunities for low-income students are to blame rather than a deliberate reverse-Robin Hood design of the system.

Even as an unintended consequences story, Hansen-Weisbrod says less than you might think about the net distributional effects, because it doesn't account for the source of funds for the public subsidies. As it happens, my Educated Conjecture from A Guy With A PhD that cost shifts in higher education on net shift tax burden from higher-income parents to their children has support from real research, see here if so inclined. UVa's William Johnson shows that since the subsidies are funded by progressive taxes, they are on net distributionally neutral to mildly progressive. (***) This, in turn, implies that unless subsidy cuts are targeted at higher-income students, the effect of following Karlson's argument as a policy prescription would tend to be to price lower-income students out of the market.

Karlson's other reasons are more conjectural:

Two, Cold Spring Shops maintains that the subsidies encourage inefficiently optimistic enrollment leading to an intolerably low completion rate. Three, the Wall Street Journal from time to time suggests a principal-agent argument, in which students are less likely to gripe about classes and textbook prices as long as somebody else... is picking up the tab.

Johnson offers some evidence on the "optimistic enrollment" front, showing that subsidies are strongly directed to higher-ability students as measured by standardized test scores. ("Efficiency" is a much more difficult question for which I don't offer an answer.) I have no opinion on what would constitute a "tolerable completion rate." My social libertarian side says that by college age, students should be free to flunk out. And, obviously, raising the net cost of school won't help with economic causes of non-completion.

I also wonder about Karlson's last point vis-à-vis his original, and totally legitimate, concern of the higher-eddiness of higher ed. From my much more limited teaching (and studenting) experience, I would agree that it's at least a plausible stylized fact that Kids These Days are more strongly inclined to treat instructors as customer service agents than Kids Back In My Days. However, another stylized fact would be that Kids These Days pay (at least relatively) more, not less, for their experiences. You won't catch me arguing causality out of any bivariate correlation, but I can imagine a mechanism for it.

-------------------------------

(*) Hansen's present interests relate to speech codes, race-based admissions preferences, and academic freedom. My two cents is that the first and the last are not among the more interesting questions in economics. My next two cents is that those suggest a possible political coloration that should be taken into account in evaluating Hansen's research.

(**) Distributional equity of public policy is emphatically not James "Waiting for the 'Income Confiscation Act of 2017'"Lileks's primary concern in life. See, e.g., here from 2002, where Lileks is stupid and lies in the process of trying to explain how politicians who oppose estate tax repeal are "stupid... or lying."

(***) For the flip side, consider: is Social Security regressive because is funded by a regressive payroll tax?