Do Pell Grants “cause” tuition inflation?

I spoke about this, and about trends in college affordability, at a recent NASFAA meeting in Washington. You can see my remarks in the video linked at the bottom of this post.

But let’s address the question directly. The idea that federal subsidy is the root of all evil — in this case a major cause of price inflation in higher education — is well established in the popular writing about higher education. Former Education Secretary William Bennett put the case very directly in his famous NY Times op-ed piece titled, Our Greedy Colleges.┬áThe thinking behind this proposition is simple supply and demand. Government subsidy pushes up demand, so up goes the price. Stories like this make the economics professor in me cringe. People armed with nothing more than Econ 101 in their toolkit often think they understand quite a bit more than their slender training allows. Why would one presume that the simple model of perfect competition, which presumes perfect information, a homogeneous product, and small, profit-maximizing firms engaged in atomistic competition, fully describes the market for higher education? Got me.

This is a “market” dominated by heavily subsidized non-profit organizations that do not maximize profit, or even revenue. Selective schools leave tons of revenue uncollected because they actually care, and care deeply, about the composition of the freshman class. The most selective schools actually charge a net price that is lower than the schools a bit below them in selectivity. The admission process is not about attracting more and more customers. It’s as much about culling customers down to the “right” ones. In that sense, admission is a two way street. Imagine Burger King rejecting business. Universities do it all the time. But I digress.

At the level of the college or university, cost per student is roughly constant as the school adds enrollment. For smaller schools, the cost per FTE student actually would fall if the school enlarged a bit, so growth is not a force for rising tuition. At the industry level, there is no evidence that the long run supply curve of seats for students is upward sloping. Places at Princeton may be fixed, but the numbers of seats in the university system as a whole is fairly elastic. In other words, the long run supply curve is flat. This suggests that tuition increases over the last generation are not driven by rising demand. They are driven by large economic forces that are buffeting the entire national and global economy. We spend a lot of time talking about those forces in our book.

When the Federal government raises the maximum Pell grant, this creates a lot of good options for universities.

  • Allow the increased Pell support to meet a greater portion of needy students’ needs.

Many schools do not fully meet need, and this forces students who wish to attend those programs to borrow more. If this is what schools do in response to higher Pell maximums then access improves. Note that this choice would have no effect on the list price tuition the school sets. And it would reduce the net price to needy students.

  • Reduce the school’s own need-based aid, and plow the extra funds into endowment for the future.

If a school made this unfortunate choice, the added Pell support would not improve access at all. But again, it wouldn’t lead to higher tuition either.

  • Reduce the school’s own need based aid, and plow the extra resources into improved programming.

Again, this choice wouldn’t do anything much for access, but it wouldn’t cause tuition to go up. This choice might help retention and graduation, if the programmatic improvements were targeted toward those goals.

  • Lastly, schools might respond to an increase in the Pell Maximum by reducing their overall tuition discount rate. Since discounting is a major force propelling the “list” price upward, anything that reins in discounting would tend to reduce tuition.
The empirical evidence on the Bennett Hypothesis is inconclusive. Using mostly panel data, the studies of the link between Pell Grants and tuition are all over the map. Bob and I are the only researchers we know who have tested the proposition using Granger Causality. We find that changes to the Pell maximum do “cause” an effect on tuition, but the link is contrary to Bennett’s assertion. Increases in the Pell maximum tend to reduce tuition in the years that follow.
Here’s the video:


  1. Rick Gardner says:

    Just one last volley in our conversation. I wanted to make sure I communicated my thoughts accurately.
    1. My comparison to the real estate bubble was separate from the resulting financial crisis caused by the bursting of the bubble. The real estate bubble was caused by easy financing, which allowed people to borrow more, and hence pay more, for houses. I intended to compare this to the student’s ability to easily, usually in less than 5 minutes, borrow from $5,500 (freshman) to $7,500 (senior). I was not implying there is a parallel to the student loans and the defaults on real estate loans that resulted in the financial crisis. By the way, since the loans are guarenteed and not based on a students ability to repay, there is no reason to lie or withhold documentation.

    2. On the fast food analogy, I agree there are different classifications of schools, public, private, for profit, etc. and in hindsight, I should have used the term revenue, rather than profit. There is just as much pressure on public institutions to meet budget demands as any other type of institution. They may be driven by goals such as growing facilities, paying faculty, attracting students, etc. but all these endeavors have a cost. Institutions cannot meet budget without students, and I stand by my belief that they will maximize attendence (aka revenue) to meet their ambitions. More clearly stated, schools turn down many applicants, but will pull students off the waiting list when a space is available.

    I also agree that Burger King is profit driven and does not have a $1B endowment. However, if customers, former customers, shareholders, etc. created a “Burger Endowment” specifically for the purpose of helping to pay for burgers… and funded it with $1B for the specific purpose of subsidizing costs for customers, I believe they would do so. They realize the same profits whether the burgers are paid for by customers or by the endowment fund. In fact, they may get more customers for this very reason, and the endowment fund may actually end up being their biggest revenue source.

    3. Increasing the speed at which students progress through the Higher Education pipeline will definitely increase the capacity of the pipeline. No doubt. But I do not know if this is actually happening, or if it will keep up with demand.

    Thanks again for your time and thoughtful response to my ramblings. I will now rest my pen and feel I have done my best to clairify my position. As I can see this topic degenerating into a dogmatic discussion of opposite views of the world, something I do not perceive helping anyone, I will “rest my case”.

  2. Mr. Gardner,

    You make some good points, but I have some further input.

    1. You ask if there are any similiarities between higher education and the real estate bubble. I believe there are some similiarities some of which you have pointed out, but some major differences as well. Colleges typically are not involved in the sort of speculative investing that major real estate investers were involved in. Also, most colleges are not involved in making “liar loans” utilizing false or nonexistent income verification documents for approval of loans, though there certainly is student loan fraud to be sure.

    2. You point out that there is a bit of a jump in the conclusion about maximizing profits. Point taken, but you would have to look at the whole fast food industry vs higher ed. Higher Ed is made up of many sectors which include private-for profit, private-nonprofit, public institions, etc. I don’t see a whole lot of variation between, McDonald’s, Wendy’s, BK, et al.

    Further, private endowments come into play. One wonders if Burger King had a Billion Dollar endowment whether they would subsidize the price of Burger’s? Since corporations take profit and declare dividends, I have to agree with Dr. Feldman.

    Also, while it is clear that the “product” is the student in higher ed (there may be others as well, such as public service or research) it is not clear to me what we are “counting” at Burger King. Is it burgers sold or paying customers? And do chicken nuggets count?

    Further, Some schools have dorms with empty beds and some schools can’t build dorms fast enough. When there are not enough beds more students will commute, but will still attend so we should not conflate enrollment with housing status…which increases cost.

    3. As far as “supply and demand” increasing seats is one way. However, there is a major conversation about “degree completion” which would factor in as well. If colleges are more efficient at progressing students through to degree completion, then increasing seats would be mitigated as a higher percentage already in the pipeline graduate.

    I believe Dr Feldmans point is a good one. The analysis is way too simplistic…and for the most part is being argued in sound bites…no pun intended…by folks not qualified to discuss in detail. I think I will buy the book.

  3. Of course Pell grants help increase tuition rates. Anytime anything is subsidized, the cost for that thing goes up.
    Look at healthcare as one great example. The more healthcare cost is paid for by “someone else” i.e., the taxpayers, the more providers raise their cost and the more healthcare is used by the consumer as they feel no need to limit their use of services. It ain’t rocket science…it’s just basic human behavior. Unfortunately, this behavior is unsustainable to the tune of a $16,000,000,000,000 deficit.

  4. Rick Gardner says:

    Great article. I have a couple comments:
    Do you see any similarities between the cost of higher education and the cost of housing during the real estate bubble ? Easy financing (student loans, Pell Grants, etc.) make paying for college easier, which I believe paralleled the easy financing of home mortgages. Since it is well documented that the financing opportunities drove up the cost of real estate, isn’t it possible that a similar situation exists in higher education?

    I agree that schools turn away applicants, but I do not make the logical jump to then assume they are not maximizing their revenues. They have capacity constraints. The Burger King restaurant is simply at capacity. If, on the other hand, schools were leaving beds empty, reserving them for the select perfect match students, this argument would carry more weight.

    Is there any precedent set to schools reducing their overall discount rate when the Pell Grant increased in the past? The maximum Pell Grant in 2000-2001 was $3,300 and it is now $5,550. I do not believe tuition has decreased over this period.

    As far as supply and demand go, you article indicates that the long term supply curve is flat, meaning that the university system is not adding more “seats” or capacity. As politicians attempt to make college affordable and boast “anybody who wants to go to college should have the right to attend”, this increases the students applying for college. If demand increases and supply is flat, prices will increase. Luckily, Virginia plans to add 100,000 additional graduates to the state higher education capacity, which I guess means that other states are decreasing their capacity by 100,000, assuming your flat supply curve is accurate.

    One thing that is absolutely accurate, is that if you increase the Pell Grant, you will absolutely decrease the remaining need for needy students who qualify. The Pell Grant alone is only a small part of the overall aid picture, but I do believe the impact of aid in general on the cost of higher education is debatable.

    Thanks for the research and thought put into your article.