Saturday, February 26, 2011

More Flexibility to Raise Tuition?

Central to debates over the New Badger Partnership is the question of whether additional flexibilities that make it possible to raise tuition are desirable.

Evidence can and must be used to make these decisions. A robust, evidence-based debate on our campus is obviously needed but to date has not occurred. Instead, to many of us outside Bascom it seems as though administrators have mostly relied on the input of a few economists and some other folks who work in higher education but are not scholars of higher education. It also seems like seeking advice from those mostly likely to agree with you. (Please--correct me if I'm wrong--very happy to be corrected with evidence on this point.)

It would be wonderful to see a more thorough review of existing evidence and the development of an evaluation plan that will assess positive and negative impacts of any new policy in ways that allow for the identification of policy effects-- not correlations. (Let's be clear: comparing enrollment of Pell recipients before and after the implementation of a policy like the MIU does not count.)

A few years ago I blogged about studies on the effects of tuition and financial aid on individual decision-making. To summarize-- effects of each are relatively small (especially when compared to effects of academic under-preparation, for example) but usually statistically significant. Also, what we call "small" reflects our value judgments, and we must recognize that.

Effects of "sticker shock" are thought to accrue early, such that the "shocked" students end up academically unprepared for college (for example don't even graduate high school) and thus are omitted from the eligible population of students on whom effects of aid and tuition are usually estimated. So hypotheses about sticker shock are very hard to test, partly because a good test requires measuring both the initial "shock" and the resulting behavior many years later (when college enrollment decisions are made).

There are other ways to think about these questions, beyond individual-level analyses. For example, we could contemplate possible effects of tuition hikes and aid increases on overall enrollment (which results from the aggregation of behaviors of many individuals). We could also look at evidence on how common it is for institutions like ours that hike tuition and raise aid to sustain the commitment to that aid over time.

Let's start down that path by examining one study that sheds light on the first of those questions. I will review more such studies in the coming days. My goal is to help facts and figures replace fear as the driving force behind our campus decisions.


In "Rising Tuition and Enrollment in Public Higher Education" Hemelt and Marcotte examine the relationships between tuition and aid on the one hand, and enrollment on the other. Essential to this discussion, for most of their analyses they disaggregate by type of institution, making it possible to isolate effects on universities comparable to UW-Madison.

Using national IPEDS data on public 4-year colleges and universities from 1991 to 2007, the authors find that on average a $100 increase in tuition and fees (in 2006 dollars) would lead to a decline in enrollment of a little more than 0.25 percent. Since we rarely raise tuition by $100, let's instead consider that a $1,000 increase in tuition would result in an enrollment decline of 2.5 percent.

But most relevant to this discussion, these economists find that the tuition elasticity of enrollment is largest at Research I universities-- and they specifically give the example of UW-Madison. According to these scholars, freshmen at universities like Madison's are "much more" affected by tuition increases than students at other kinds of institutions (for example, freshmen at UW-Stout). (The tuition elasticity is -0.24 at Research I's compared to -.107 on average). And, the average amount of aid received has the smallest effects for students at Research I universities, compared to other colleges (.06 on average, compared to .01 at Research 1's).

In plain English, what does this mean? The consequences of raising tuition are greatest for students at places like Madison, and the benefits of increasing aid are smallest.

Why is this? The authors consider the possibility that students at Madison are not weighing the price of Madison relative to the price of Stout or Eau Claire, nor the price of other Big 10 schools writ large, but rather the price of comparably elite Research I institutions. Restricting their analysis to the top 120 public universities in the country, then, they again find that these students are particularly price sensitive, and particularly aid insensitive.

A few words from the authors: "These patterns in price and aid sensitivity are consistent with students opting out of “top 120” schools for competitors as price rises, while finding a way to pay tuition bills at other state schools where students may have fewer options....The evidence...of higher price sensitivity but lower aid sensitivity at “top 120” and Research I institutions raises general questions about enrollment patterns at public four-year colleges and universities, beyond the implications of tuition on enrollment at single institutions. One implication may be a shift of students from higher income families to private institutions or public universities in other states, along with a shift of students from lower income families to less expensive public universities within the state. This would suggest a redistribution of students across public colleges and universities within a state, with those most financially able leaving the system, and others scaling back to enroll at more affordable
institutions. Obviously, student-level data are needed to test this."

Distributional consequences of tuition policies are too rarely considered, and are not addressed in the NBP.

Sure, consequences and benefits should be put into context-- for example considered against the consequences of not raising tuition. But this paper by respected economists clearly indicates that it is not appropriate to assert that increasing financial aid at institutions like UW-Madison will effectively hold students harmless from the negative effects of tuition increases. Enrollment will be affected, and distribution of enrollment across institutions may be particularly affected. Who will measure those effects? And who will care?