By Peter Sacks
Read Part 1 here.
In examining the gulf between sticker price and real cost, let's consider the top 10 national universities as defined by U.S. News & World Report in its most recent rankings. Using U. S. Department of Education data, I compiled the average net prices that students from different family income groups would pay at the top 10 national universities combined.
Despite total sticker prices averaging more than $50,000 a year at these top 10 universities, net prices range from a low of $4,652, paid by students from poorest family income group, to a high of more than $35,000 paid by students from the richest category of family income.
These averages, however, mask the significant differences in net prices paid by poor and rich students at the individual institution. At Harvard, students from families in all income categories fare significantly better in terms of net price than they might at Harvard's competitors. Harvard's poorest students, whose parents earned $30,000 or less, paid net prices averaging just $2,170, significantly less than the average net price charged low-income students at all the Top 10 national universities.
At the same time, Harvard's net price for upper-income students, whose parents earn more than $110,000 a year, is also among the lowest of these universities, at about $32,000. Princeton, however, has the distinction of offering the lowest net price to upper-income students -about $24,000 a year. And then there's Duke. Duke not only charges low-income students the highest net price among its competitors, $9,200 a year, but Duke also charges upper- income students the highest net price among the top 10, more than $42,000 a year.
$50,000 off Sticker Price at Stanford
Another measure of a university's generosity to low-income students is the absolute dollar difference between what a given top 10 university charges low-income students and the full sticker price. On this measure, Stanford stands out, providing low-income students a discount of more than $50,000 a year off the sticker price. Duke, again, appears to be the least generous to low-income students among its competitors in the top 10, providing a discount off the sticker price of about $44,000.
Clearly, rich universities are able to provide steeper discounts - meaning more generous financial aid packages - than their less wealthy competitors. To assess whether an elite university is making any meaningful attempts to enroll lower-income students, one might argue that an institution ought to be gauged, not just on the difference it charges low-income students and high-income students, but also on how this price gap compares to its overall wealth.
I first compiled data from the U.S. Department of Education on the top 10 national universities in terms of their total endowment for each undergraduate student in 2008. The top 10 nationals fall into three distinct tiers: Harvard, Yale and Princeton, each having an endowment per undergraduate student in excess of $3 million, compose the richest grouping. The middle tier consists of the California Institute of Technology, Stanford and MIT. Among these ten elite universities, the "poor" group consists of Columbia, Dartmouth, Duke and Penn.
Next, I calculated the ratio of what each university charged its richest students compared to its poorest students. The greater the ratio, the more generous is the university's pricing structure for low-income students. In terms of the net- price ratio, the top 10 national universities again split into three fairly distinct groups. The universities with the highest ratios,10 and above, include Harvard, Stanford, MIT, and Cal Tech. A middle group, having ratios of between 8 and 9, includes Princeton, Columbia and Dartmouth. Then there is a final group with the lowest net price ratios, which includes Duke, Yale and Penn.
Where the Rubber Meets the Road
The nexus of rich student/poor student net price and a given university's financial capacity is where the rubber meets the road in terms of a university's interest in enrolling lower-income students. Duke and Yale, for example, share in common a relatively low net- price ratio. But Duke is among the least wealthy of the top 10 nationals. Yale, by contrast, is among the wealthiest, indicating that it has the financial capacity to be far more generous to lower-income students than is currently the case.
Besides Yale, the remaining top 10 nationals in the top tier of institutional wealth are Princeton and Harvard. While the financial capacity of the two universities is roughly similar, Harvard's net- price ratio of 15 - almost twice that of Princeton's -- indicates the two universities have two very different philosophies with respect to their interest in enrolling low-income students. Indeed, judging simply by Pell grant percentages, this seems to be the case. In 2008-2009, 13 percent of Harvard's undergraduates received Pell Grants compared to 10 percent at Princeton.
In contrast to Princeton, which enrolls lower-income students below its capacity to do so, Dartmouth and Columbia stand out among the top 10 nationals as doing more for lower-income students than their wealth would suggest. This is especially so for Dartmouth. Although its endowment per student is a third of Princeton's, Dartmouth's ratio of net price paid by rich students relative to poor ones is greater than Princeton's. The Pell grant numbers underscore these differences: 12 percent and 13 percent, respectively, of Dartmouth and Columbia's undergraduates received Pell grants in 2008-2009.
Liberal Arts Colleges
While the variation in net price among paid by students from different family backgrounds is significant at the top 10 national universities, the price gaps for different income groups at the top 10 liberal arts colleges are extreme. The following chart shows average net prices among the top 10 liberal arts colleges in 2008. Students at Carleton College, on average, paid the highest net price among these 10 colleges, more than $25,000 per year. In contrast, students at Williams paid average net prices of just $13,789.
But when family income is considered, the differences in net price at the top 10 liberal arts colleges are eye-opening. For instance, students at Pomona College whose parents earned no more than $30,000 per year paid net prices averaging just $344. Low-income students at Carleton, Middlebury and Haverford weren't nearly so fortunate, paying net prices of more than $7,300. For upper-income students, whose parents earned $110,000 and higher, Pomona was also the best deal, charging such students net prices averaging $27,942. Besides charging the poor students a hefty net price, Haverford was the most expensive for upper-income students as well, charging them net prices averaging more than $38,000.
Again, institutional wealth appears to be the decisive factor in explaining these large differences in net prices by family income. Pomona's wealth per undergraduate was nearly $1.2 million, significantly higher than all its peers, and dwarfing the wealth per student at the least wealthy competitor, Davidson College, whose endowment per student was $301,404.
Pomona uses its wealth to provide extraordinary discounts for lower-income students. Compared to its peers, Pomona's net price ratio of 81 is staggering, while Amherst ratio of 20 came in a distant second. Indeed, the differences among Amherst, Swarthmore and Williams, which form a second tier of institutional wealth after Pomona, are interesting. Amherst clearly stands out as being the most generous to low-income students among its comparable peers, with a net price ratio twice that of Williams and Swarthmore. As an institution of moderate wealth per student, Wellesley -- having a net price ratio equal to richer schools Williams and Swarthmore - also stands out in its relative generosity to low-income students.
The Affordability Gap
But even when colleges and universities provide relatively generous tuition discounts for lower-income students, affordability remains a major barrier to lower-income students. The best measure of affordability isn't net price alone, but net price as a percentage of family income. This information is not readily available for each institution in the top 10 lists. But I ran numbers for very selective institutions as a whole from the 2007-08 National Postsecondary Student Aid Study.
Consider students at very selective colleges and universities whose parents earn $36,000 a year or less. Some 66 percent of these students encounter net prices that are at least 41 percent of their family's entire income for a year. A family facing a college-cost burden of that magnitude means that parents must make a tradeoff between basic food and shelter and paying their child's net college costs.
By contrast, consider students whose parents earn at least $110,000. About one-third of these students at very selective colleges and universities face net prices that are just 10 percent of family income. Another 41 percent face net prices that are between 11 and 20 percent of family income.
Thus, even when wealthy institutions heavily discount tuition costs for lower-income students, the current subsidy levels are, with rare exceptions, not enough to make top-tier colleges and universities affordable for families who aren't relatively wealthy. This may partly explain why there are so few lower-income students among the ranks of America's elite colleges and universities. Indeed, studies have found that low-income students constitute as little as 3 percent of the enrollments of the nation's most selective institutions, while upper-income students make up about 75 percent of the enrollments at such institutions.
From a public policy perspective, at question is whether America's best colleges and universities, pre-occupied by status seeking behaviors that are rewarded in such rankings as U.S. News & World Report, are spending too much of their wealth on merit scholarships and other forms of tuition discounts on wealthy students whose college plans are not dependent on financial aid. As a consequence of excessive spending on the "arms race" for high-scoring students, are colleges spending insufficiently on need-based grants for students who would be unable to afford the college without support?
Consider how institutional merit-based grants are distributed to students throughout the higher-education system, depending on parent education and income.
For example, among students from families in the highest income group, with family incomes of $105,000 and higher, 34 percent received merit grants of $8,000 and higher. That was twice the rate that these affluent students received modest merit grants of between $1 and $1,499.
The size of merit grants is clearly related to family income: Among low-income students, whose families earned $36,000 or less, just 24 percent were provided the most-generous merit grants.
When I calculated the distribution of merit grants to college students based on parent education levels, the relationship became even more pronounced. Some 37 percent of students whose parents had doctoral degrees received the most generous merit awards. That's compared to 28 percent of students whose parents had bachelor's degrees who got the most generous merit awards.
By contrast, among students whose parents had only a high school diploma, about 33 percent received the least generous merit awards, of less than $1,499. Fewer than 15 percent of students whose parents had doctoral degrees received merit awards in this least -generous category.
As of March this year, there were few signs that the tuition-discount arms race in higher education industry was abating. Indeed, that's when the National Association of College and University Business Officers released its annual survey of tuition discounting at private colleges and universities, reporting that the average tuition-discount rate for full-time freshmen entering college in the fall of 2008 rose to 41.8 percent, up from 39.1 percent in the previous academic year.
Some 42 percent of the institutional grant awards were based on "merit," while 38 percent were based on financial need, the Chronicle of Higher Education reported.
"One of the unintended consequences of tuition discounting is that financial access to four-year colleges for lower income, financially needy students may generally be diminished," concluded Jerry Sheehan Davis in a 2005 report by the Lumina Foundation. "Lower-income students' access to grant aid is decreasing, while their cost of education is rising, making it increasingly difficult for them to afford college."
In terms of the public policy goal of increasing college enrollments among lower-income students, the tuition-discounting trend is indeed a double-edged sword.
The steep discounts that the richest colleges and universities afford low-income students are surely advantageous to the relatively few such students who are admitted to these elite schools. But less wealthy institutions might calculate that the bulk of their tuition discounts should be aimed at more affluent students who, colleges believe, are economically advantageous in at least two ways: Not only do such students require less financial aid, but their relatively high average SAT scores enable colleges to compete with their peers on college rankings, which in the long run can lead to growing endowments and more funds for tuition discounts to more students.
At least that's the theory.
Although individual colleges make these calculations, the systemwide outcomes suggest that such goals are rarely achieved, according to Davis's analysis, who found that widespread tuition discounting did not generally enhance their net tuition revenues, nor lead to higher average SAT scores.
"Tuition discounting works for some colleges," Davis writes. "We recognize that colleges are appropriately acting in their self-interest when using tuition discounting to try to achieve institutional goals. We also recognize that colleges individually and collectively millions of dollars in student aid awards to lower- and middle-income students to try to make themselves affordable to such students. But this report has shown that the actions by large numbers of individual colleges, when combined across all institutions, have produced some worrisome outcomes for students and for colleges in general."
Such findings, however, have largely failed to retard the tuition-discount gaming strategies of the individual colleges and universities hell bent on climbing up the college rankings ladder. The enrollment managers at such institutions might reason that, in the short run, pouring scarce institutional funds into steep discounts to affluent students might well harm the opportunities for needy students.
But in the long run, who knows? In the long run, there's always a chance that a Kenyon College could become the next Pomona. Or, that a Washington University could become the next Harvard.
In the long run, as the economist John Maynard Keynes once said, we are all dead.
Peter Sacks is a writer and economist. He is the author of Generation X Goes to College and Tearing Down the Gates: Confronting the Class Divide in American Education.