Every few years, I return to a consideration of ever-increasing college prices (2009, 2011). In these reports, I describe college pricing as an elephant because the issues around it are large, and depending on one's perspective, one can come to quite different conclusions as to what the beast looks like. In particular, I have looked at undergraduate prices (tuition, fees, etc) at private, non-profit four year institutions from three different perspectives: the public at large, students and parents, and the institutions themselves. I was stimulated to return to this issue by the recent release by the National Association of College and University Business Officers (NACUBO) of their 2013 Tuition Discounting Study, which provided me with some new thoughts about the institutional view of the beast. This year, I am adding a fourth perspective, that of government, since some interesting issues have arisen there recently.
THE PUBLIC'S VIEW OF THE ELEPHANT
In one way or another, almost all discussions of college pricing are influenced by the data shown in the first graph, which provides a 30 year, inflation adjusted comparison of family incomes to average published tuition and fees at private non-profit four year institutions. Mean household income data by quintiles and top 5% comes from the Census Bureau's Historical Income Tables, Table H.3, and the average published tuition comes from the College Board's Trends in College Pricing 2013.
As can be seen, mean household income for the bottom 80% of households (1st 4 quintiles, corresponding to household income less than approximately $100,000/year) has been relatively flat (inflation adjusted) over the last 30 years, even decreasing somewhat over the past 15 or so years. Household real income for the top 20%, and especially the top 5%, grew nicely for roughly the first 15 years of this period, then flattened out about 2000. However, the average inflation adjusted published price of higher education has increased by roughly 150% over this period. Although incomes of all groups have risen and fallen in response to broad economic conditions, published real (inflation adjusted) college prices have marched quite steadily upward at a rate of about 3.2% a year with very little coupling to general economic conditions or to the resources of its customers. This sort of sector-wide "tone-deaf" behaviour makes it easy for anyone who wants to ignore the messy complexity of actual price vs published price to find a ready audience.
The public also takes note of the widely publicised rapidly rising levels of student debt: the Federal Reserve of New York reported the 12/31/13 student debt level to be $1.08 trillion, up $114 billion in 2013 alone. This figure from Vanguard nicely captures the recent rapid increase in student loans compared to other debt (each type of loan indexed to 1.0 in 2007). An earlier Federal Reserve of New York analysis suggested that student debt was hitting the broader economy by depressing purchases of autos and first homes. A recent New York Times article reports studies showing that student debt also lowers the probability that the graduate will start a business of her own, and generally pushes graduates away from lower paying public-interest jobs. All of this discussion simply contributes to a very negative public image of the effects of apparently out-of-control tuition.
THE STUDENT'S AND PARENT'S VIEW OF THE ELEPHANT
The reality,of course, is that published price and actual price payed can be quite different. A considerable amount of non-loan financial aid is provided by Federal and State governments, and by the institutions themselves. This graph shows the main components of total undergraduate financial aid (excluding loans) over time. It is interesting to note that Federal and Institutional Grants were roughly equal from 1990 until the big economic downturn around 2008, when the Federal government increased grant aid significantly.
The variety of sources shown in this graph suggest the difficulty in talking about "average non-loan financial aid" for the "average student". Most of State Grants are directed at the public sector of higher education, most of the Institutional Grants are to be found in the private sector, educational tax benefits benefit different income groups differentially, merit and need are weighed differently by different institutions. Nevertheless, the College Board has taken a stab at calculating the average student aid in the private sector. This should be compared with the published price for tuition, fees, room and board (TFRB), since many grants also cover living expenses. The resulting picture for the net TFRB tells a different story than is suggested by the top graph. When grant aid is included, the inflation corrected net TFRB grew at about 1.6% per year over the period 1990-2008, about 1/2 as fast as the published TFRB. The big jump in aid during the last 3 years has led to an essentially flat net TFRB for that period, according to the College Board. A very good outcome, indeed, from the perspective of "average" students and parents.
Of course, in real life, obviously there will be some who do better than average, some who do worse than average. For example, the NCES 2011-2012 National Student Aid Study reports that almost 24% of students at private nonprofit four year institution receive no grant aid at all (Education Tax Benefits not included in this data): These students therefore are paying the rapidly rising published price. Roughly one-half of these full-pay students have parents whose income is in the lower four quintiles, the categories where average real income has not increased significantly in the past several decades.
A recent report analyzed net tuition as provided in required institutional submissions to the Department of Education. There are certainly problems with these data, but the results suggests the complexities of interpreting the data above for the "average" student. This report finds that net real increases in tuition at private universities over the past three years were greatest for students in the lowest income category, with middle income students having the lowest net increase. I will comment on why this might be the case below when considering the College's View.
It is also the case that an individual student may see a very different net tuition as she moves through her education. Overall, the amount of institutional aid often is somewhat higher for freshmen than for upperclass students. In addition, even if the percentage of sticker price that is covered by aid remains constant from year to year, the dollar cost of the portion that remains the student's responsibility will increase at the rate of increase of the published price. Thus, even students who are getting significant financial aid may belong to the group who see their actual costs increasing at the rate of the published price.
Of course, for all of the students in any category for whom the non-loan aid is not sufficient to cover the published price, there are loans. This, also is a key part of the Student's and Parent's View of the Elephant. Unfortunately, the College Board does not break down these total undergraduate loan data into public and private sector, so I cannot provide a figure similar to the one above on Net Tuition in the private sector to better understand the temporal evolution of the loan burden of the "average" student in the private sector. However, the NCES 2011-2012 Student Aid Study reported that 62% of students at four year private non profit insitutions have loans that average $8509, and 12% of their parents have Direct PLUS loans that average $14,861. Suffice it to say that loans are a significant part of the perception of students and parents of the elephant of pricing.
Finally, it is worth noting that a very interesting recent study by Monks has shown that net price has a relatively small impact on enrollment of low income students, while published price has a significant negative impact on enrollment of these students. Thus, for many low income applicants, the "good outcome" shown in the net cost graph above does not compensate for the bad image the the general public has of the tuition picture.
THE GOVERNMENT'S VIEW OF THE ELEPHANT
As the late Senator Everett Dirksen is reputed to have said (but probably didn't), " A billion here, a billion there, and pretty soon you're talking about real money." Federal aid for higher education has increased considerably in the past few years, and has reached a level that qualifies as "real money" for many in Congress. As can be seen in this chart, total annual Federal undergraduate aid, including loans, is now in the area of $130B, up from a bit less than $65B a decade ago. In a climate in which the goal for many has been to lower the Federal budget, this growth certainly has attracted attention. (It should be noted that Federal loans actually provide a positive cash flow for the budget, but that apparently has nothing to do with the ongoing discussion.) In particular, much mention has been made in Washington of what is sometimes called the "Bennett Hypothesis" after an article by William Bennett in the New York Times in 1987. This hypothesis suggests that the availability of student aid encourages colleges to increase tuition at a faster rate than would occur otherwise. This hypothesis increasingly finds supporters in government on all sides of the political spectrum, including President Obama.
A number of scholarly studies have failed to support this hypothesis because they find little correlation between changes in Federal financial aid and tution changes. My opinion is that the studies show nothing because they have asked the wrong question, and confused cause and effect. However, correct or not, the hypothesis has numerous supporters in both major political parties, and is key part of the discussion over the role of the Federal government in reducing future tuition increases. This argument suggests that the government can limit growth in the increasingly significant student aid budget line and simultaneously claim credit for limiting unpopular tuition increases. Talk about a political win-win!
THE COLLEGE'S VIEW OF THE ELEPHANT
As I have argued previously, most colleges are described by business models that produce a roughly 3% real (inflation adjusted) yearly increase in costs - a value that has been achieved on average each year for over three decades. It is enormously difficult to change business models in a significant way, since that requires a new rebalancing of mission, resources, procedures and customer expectations. Thus, serious consequences will occur for the current business model of higher education if annual revenues no longer increase at historic levels and thus can no longer balance the model- dependent built-in increases in costs. Net tuition is obviously a key component of annual revenues.
Every college sets up a pool of money that it uses to provide grants to students to help in achieving the class it wants and needs. The pool typically includes money provided by gifts and endowment, and some portion of the tuition and fees that it receives. The fraction of the published tuition and fee rate that is "recycled" back to grants is called the discount rate. The higher the discount rate is, the less net tuition and fee income there is to spend on actually running the institution. Consequently, since tuition and fees are the dominant component of unrestricted income for most colleges, the discount rate is an important indicator of collegiate financial health.A second important indicator is the size of the student body. If the net tuition per student decreases due to an increase in the discount rate, the total net revenues can still increase if the number of students increases sufficiently. Alternatively, if the number of students fall, the likelihood that net revenues decrease is very high. The recent NACUBO 2013 Tuition Discounting Study (TDS) looks at both of these indicators for private non-profit colleges and universities.
TDS found very significant differences in how various sectors of the private non-profit sector were experiencing the elephant. Changes in student numbers suggest some of the sector differences. Almost 50% of all institutions in the survey reported that they lost or only maintained enrollment from Fall 2012 to Fall 2013; however, 85% of small institutions (total enrollment under 4000, mainly award only associate's and bachelor's) fell into that category, while only 11% of Comprehensive/Doctoral and 4% of Research institutions had similar problems. Overall, 17.2% of institutions saw a decrease of 10% or more in size of freshman class. A few institutions may have purposefully cut their student size, e.g. to increase selectivity, but a 10% plus decrease in freshman class certainly created a significant budget problem for most.
A similar effect is seen in the discount rate, as shown in this figure taken from TDS. Again, small institutions must discount their tuition much more than comprehensive or research institutions in order to fill their classes. Interestingly, however, the largest increase in discount rate over the past 13 years (the interval shown in this figure) has occurred in research universities.The rapid increase in discount rate after roughly 2007 corresponds timewise to the increase in slope of institutional support support shown in the graph of Total UG Grant Aid above. Unless offset by increases in student enrollment, this ever-increasing discount rate means ever-less net revenue that the colleges have to spend on operations.
Within these broad institutional categories, there are wide variations in the discount and enrollment trends. Generally, these variations reflect brand strength and size of non-tuition sources of revenue, e.g. endowment size. TDS notes an increasing sensitivity to price and a subsequent "flight to quality and size". This flight to quality with resulting increase in demand for high brand institutions has enabled some of these institutions to decrease their discount rates, and others to expand classes while maintaining quality.
In fact, the discount rate and how it is set is one of the primary tools that institutions use to maximize their net income. As noted above, the institution may lower its discount rate, hoping to attract enough additional students to increase net income. An institution can also change how it uses financial aid to attract more lucrative income demographics. For example, a potential full pay student who is tempted to attend a competitor might well be convinced to come to your institution for a relatively modest amount of "merit" aid. Thus a zero contribution from a full pay student who goes elsewhere is turned into, say, a 90% contribution from a 10% grant student who is actually at your institution. That grant comes out of the grant pool, of course, leaving less for someone further down the income distribution. Such calculations could lead to the type of results found in the report described in the Student and Family section above. Thus, determining discount rates and how they are achieved requires a balancing of need for resources and a variety of institutional values, such as desire for economic diversity.
Despite all that has been done through mechanisms such as these to increase critical net revenues per student,TDS ruefully concludes:
When average net tuition revenue dollars are adjusted for inflation,institutions have seen, on average, no growth over the last 13 years.
That is, even with an annual growth of more than 3% in the inflation adjusted published tuition over the last 13 years, the policy of "high tuition, high aid" has led to no real growth in institutional revenues!
Looking at the elephant of higher education pricing from a bit of a distance, we see that the strategy of raising published prices very rapidly and trying to mitigate resulting problems by providing aid - high tuition, high aid- doesn't work particularly well in the eyes of any constituency, and for some it appears to be a near catastrophe. NOBODY LIKES THE ELEPHANT! It may be time to begin to look at alternative pricing strategies.