Overstating the Case

by Andrew Gillen

Sara Goldrick-Rab overstates her case in this post, attacking Neal McCluskey for believing in the Bennett Hypothesis (the idea that financial aid leads to higher tuition):

the Bennett hypothesis keeps on rearing its ugly head. I think after more than 20 years of this nonsense it's time to call the idea what it is-- just plain stupid-- and stop giving ink to the people who repeat it.

She’s very angry that the Bennett hypothesis won’t die, and says the evidence shows the idea is "stupid." But there a slight problem with that – that’s not what the evidence says.

The story that emerges from the literature as a whole is that the evidence on the Bennett hypothesis is mixed and contradictory. There isn’t enough either for or against it to make a conclusive determination as to it’s validity. So attacking Neal for espousing a discredited theory that has not in fact been discredited is, well, wrong.

To the literature:

Sara quoted Bridget Terry Long

"Of the many studies that have tried to identify whether colleges react to federal financial aid, most find little to no response. While several studies do find a college price response, their overall results are mixed and often contradictory.”

To which I will add 3 more that I could dig up on short notice.

Rizzo and Ehrenberg

“Estimates of the size of this “Bennett Hypothesis” at public institutions range from negligible to a $50 increase in tuition for every $100 increase in aid.”

McPherson and Schapiro

“We found no evidence of the “Bennett Hypothesis,” [at private institutions]… We did, however, find that public four-year institutions tended to raise tuition by $50 for every $100 increase in federal student aid.”

Singell and Stone

“Previous studies with evidence pertinent to the Bennett hypothesis are suggestive. McPherson and Shapiro (1991), Turner (1997), Li (1999), Netz (1999), Acosta (2001), and Long (2002) all find evidence that tuition rises for at least some segments of the higher education market”

The conclusion to draw from these and other papers is that we just don't know if the Bennett hypothesis is valid or not. Personally, I disagree with Neal on this one - I don’t think the Bennett hypothesis provides an adequate explanation of what’s going on. But that is a far cry from being able to say it's "stupid." This is why I came up with a new and improved version (“nuanced” in Sara’s words. BTW, thanks for picking up on the difference, unlike some others who said “we’ve seen this kind of report before and it’s usually someone with an agenda and not necessarily a very insightful report”).

Importantly, my new theory offers a crucial take on why the evidence is all over the place. On page 18, I said

The second way in which this report differs from the Bennett Hypothesis is that it is explicit about when the effect occurs (and the types of aid likely to suffer from it). Specifically, aid will fuel increases in spending when it is given to students whose ability and willingness to pay is in excess of current costs at the school. Because costs and ability to pay vary by school, this implies that a much more nuanced view is warranted. The same aid program can have different effects based on the characteristics of the school and the students attending… Thus, lumping all federal aid together when analyzing its impact, or even all aid of a given type, is unlikely to yield accurate results. Unfortunately, public data on aid is generally only available in aggregate form (not student specific), which limits the extent to which we can analyze these issues.

Previous studies have mostly used aggregate figures (because finer data isn’t easily available), and some lump all types of aid or all types of a certain kind together (because the original Bennett Hypothesis didn't indicate any reason not to). If my new theory is correct, it would go a long ways towards explaining the mixed results, since at some schools, the aid is given to students who are already willing and able to pay current costs, and at some schools, it isn’t. Thus we would only expect to see higher costs (and presumably higher tuition) at some schools, not all – which is precisely what the literature keeps finding.

Provided I ever finish the current projects I’m working on, I intend to revisit this topic with some rather clever (if I may say so myself) ways to test my new theory more thoroughly (if anyone out there is interested in partnering with me on this, just get in touch).