Dollars, Cents, and Nonsense: The Harmful Effects Of Federal Student Aid

The federal system of higher education student financial assistance grew from good intentions but has become a disaster, a prime example of the law of unintended consequences. In this new study from the Center for College Affordability and Productivity, direct and circumstantial evidence leads to eight problems with the programs:

  • They have significantly contributed to tuition price inflation, the proceeds of which have helped fund an unproductive and costly academic arms race;
  • They have failed at achieving their primary goal of a larger proportion of college graduates from households with relatively low incomes;
  • They have succeeded in expanding enrollments, but this has contributed to the  large graduate underemployment problem;
  • The burden of the programs on borrowers and taxpayers is substantial, with loan delinquency rates that, if appropriately measured, approach 30 percent;
  • They, with their disdain for academic standards, have contributed to decline in quality  within American higher education;
  • The rising student loan burden has contributed to reduced household formation and lower birth rates;
  • Loan delinquencies reflect the lack of credit and academic standards, but also the fact that colleges face no consequences for non-payment: they have no “skin in the game;”
  • It is likely that the rise of loan financing has contributed to the significant decline in the national savings rate during the four decades that federal loan programs have existed.


The solution involves reducing the magnitude of these programs sharply, doing away with programs such as tuition tax credits and PLUS loans that are not exclusively targeted for those who require financial assistance to attend school. Academic and financial standards need to be tightened. Above all, colleges that cause much of the financial problems through their often inappropriate admission decisions should have “skin in the game” by facing significant adverse financial consequences for their decisions.


About the Authors

Richard Vedder is the director of the Center for College Affordability and Productivity, distinguished professor emeritus of economics at Ohio University, and a adjunct scholar at the American Enterprise Institute. He has authored Going Broke by Degree: Why College Costs Too Much (Washington, D.C.: AEI Press, 2004), and has written and lectured extensively on the cost of higher education. He received a BA from Northwestern University and an MA and PhD in economics from the University of Illinois.

Christopher Denhart is the Administrative Director of the Center for College Affordability and Productivity. He recently completed a BS in economics from Ohio University. He is the co-author of Why Are Recent College Graduates Underemployed? University Enrollments and Labor-Market Realities, a 2013 CCAP report.

Joseph Hartge is a Research Associate at the Center for College Affordability and Productivity. He is a senior in Ohio University’s Honors Tutorial College pursuing a BS in Economics.



The Dysfunctional System of Federal Student Financial Aid

The federal government was once almost nonexistent in higher education. Now, it plays an important role in financing attendance at America’s colleges and universities, and funds vast amounts of collegiate research. The intentions of those responsible for expanding federal involvement in higher education were admirable, although the Higher Education Act of 1965 was part of a Great Society effort to solidify and expand the progressive wing of the Democratic Party in determining national public policy.1 Beyond politics, however, most advocates of an expanded federal role wanted to use higher education to improve educational opportunity for those of modest financial means, thereby increasing economic opportunities and narrowing differences between the affluent and the poor.

After reviewing the various federal programs that evolved to assist college students, we conclude that they have largely failed. For example, the proportion of lower-income recent college graduates is lower than when these programs were in their infancy. The programs are complex and Byzantine, leading to forms such as the FAFSA (Free Application for Federal Student Aid), whose very complexity has reduced participation by low-income students. The law of unintended consequences has reared its ugly head.

On balance, these programs have increased the burden of attending college, leading to enormous student loan debts with unexpected effects such as reduced rates of marriage and childbearing and lower household formation, depressing the housing market. With their indifference and even hostility to academic excellence, these programs have contributed to grade inflation and mediocre learning outcomes. They have helped fuel a highly inefficient academic arms race and promoted the non-academic side of university life, complete with luxury housing, climbing walls, golf courses, and hedonistic living. They have led to massive university overinvestment, leading to financial hardship for tens of millions directly impacted, not to mention hundreds of millions living in families paying taxes. These programs have led to underemployment of college students, to credential inflation, to rent-seeking amongst members of university communities. In short, they have eroded the quality and integrity of higher education in today’s America.

Historical Federal Involvement in Higher Education

Colleges and Universities Before the Higher Education Act of 1965

I n the 378 years of American higher education, participation by governments in financing universities has been a significant factor for about half the period, beginning early in the 19th century. Direct federal government involvement in higher education begins in the Civil War era with the Morrill Act, although earlier legislation that predates the beginning of constitutional government, notably the Northwest Ordinance of 1787, provided a means for some indirect public support.2

the Servicemen’s Readjustment Act, better known as the GI Bill. That legislation is credited as being the impetus behind a major expansion of higher education in the United States. In the decade between the 1939-40 academic year (the last year before the beginning of serious U.S. preparation for World War II) and 1949-50, enrollments rose 78 percent, with over 80 percent of that increase occurring among men. The GI Bill provided essentially free college for veterans.3 One crude way to gauge the impact of the GI Bill is to assume that, in the absence of that legislation, male enrollment gains in the 1940s would have been the same as for females, few of whom were eligible for GI Bill funding. There would have been a reduction of more than 655,000 students in 1949 than actually were attending—a 25 percent reduction in enrollments.4 So there appears at first glance to be some truth to the notion that the GI Bill was a major factor in moving America from a relatively elitist collegiate environment where roughly 5 percent of Americans had at least a bachelor’s degree (1940) to one where college became a real option for a large portion of highschool graduates.

At the same time, supporters of the GI Bill exaggerate the transformative impact of it. The aboveenrollment estimate of the GI Bill’s impact probably overstates its importance; one estimate is that the GI Bill (and subsequent reauthorizations) increased postsecondary educational attainment of males 15-20 percent and that most of the incremental enrollment came from those of above-average incomes.5

Moreover, enrollments were growing robustly, except for periods of wartime or depression, well before the 1940s and the GI Bill (Figure 1). With the single exception of the Depression decade (1930s), enrollments grew about 50 percent or more in every decade from 1890 to 1940. In the boom decade most comparable to the 1940s, the 1920s, its enrollment growth of 84 percent exceeded the postwar boom.

Percent Growth in Enrollment by Decade, 1890 to 1950
Percent Growth in Enrollment by Decade, 1890 to 1950
Financial Burden: Tuition of Purdue University Students as a Percentage of Indiana per Capita Income, 1939 to 1955
Financial Burden: Tuition of Purdue University Students as a Percentage of Indiana per Capita Income, 1939 to 1955

Amidst today’s praise for the GI Bill, it is often forgotten that several prominent university presidents, such as Harvard’s James Bryant Conant and the University of Chicago’s Robert Maynard Hutchins, publicly opposed the legislation, concerned that unqualified individuals would populate the colleges and erode academic quality. Even before World War II, Conant opined that “there are too many rather than too few students attending the universities of this country.”6 After the adoption of the GI Bill, he said “we may find the least capable among the war generation is flooding the facilities for advanced education.”7 Hutchins went further, worrying that in the aftermath of the GI Bill, colleges would become “intellectual hobo jungles.”8 Prominent Harvard economist Seymour Harris opined about an excess supply of college graduates—the underemployment problem we have today.9 Rather than being myopic elitists as most modern-day scholars would likely characterize their opposition, educational leaders like Conant, Hutchins, and Harris were far-sighted visionaries who foresaw the ultimate consequences of the GI Bill and federal student financial assistance: overinvestment in higher education, dilution of academic standards, increased dropouts or elongated years of matriculation, credential inflation, higher costs, no real gains in educational opportunities for the poor, and, ultimately, severe underemployment of college graduates.

A key reason for the rising enrollment in American colleges and universities was that it was becoming more affordable. Incomes rose with economic growth and college tuition fees increased much less rapidly. Figure 2 looks at the period 1939-55 at one typical state land-grant school, Purdue University. In 1939, tuition fees at Purdue equaled over 20 percent the per capita income levels in Indiana; by 1955, those fees were under 10 percent of per capita income.

By the mid-1950s, the enrollment stimulus from the GI Bill was winding down (the Korean War having ended in 1953), yet in the years 1955 to 1965, before the Higher Education Act, enrollments continued to grow—more than doubling—with modest federal involvement. To be sure, as a consequence of the scare associated with the Soviet Union’s launching of Sputnik in 1957, there was some renewed federal higher education effort. In late 1958, the National Defense Education Act passed, including some funds for federal student loans. Those programs, however, were modest. In 1965, when the modern higher education law was passed, student loans under the National Defense Education Act were $159.2 million.10 Correcting for inflation, that is roughly $1 billion today—less than 1 percent of the current volume of student loans issued annually.11 These loans were about $30 for each student attending college in 1965, a relatively trivial amount. There were also smaller sums for fellowships to support graduate work in the sciences.

The federal student loan programs were partly motivated by the goal of equalizing educational opportunity. But there is a market failure argument often used to justify these programs: there are significant information costs involved in making loans to relatively inexperienced teenagers and young adults. There is little ability to measure the probability of the student repaying the loan, for example. These information costs, it is argued, would lead to a suboptimal amount of student lending if left to private market forces.12 Thus is the rationale for federal involvement in providing loans to students.

American higher education grew robustly in the generations before federal student financial involvement. Vast numbers of new colleges and universities were established to accommodate rising student demand. Whereas at the beginning of the 20th century, universities with over 10,000 students were unknown, by 1965 there were many universities that did not even exist in 1900 with far more students than that. The notion that “federal student financial assistance programs created the establishment of wide-scale higher education for Americans of all incomes” is a myth.

The Modern Growth of Federal Student Financial Assistance

The story of the modern explosion of federal student financial assistance programs begins with the Higher Education Act of 1965, which provided the framework for ever-growing federal involvement. In contrast to bitter battles over other issues such as civil rights, the Higher Education Act was non-controversial. Only three senators voted against it in the Senate, and the House of Representatives passed it 368 to 22. During its signing, President Lyndon Johnson said “This act means the path to knowledge is open to all that have the determination to walk it.”13 The 1965 legislation has had numerous “reauthorizations,” most recently in 2008, each time expanding the scope of the legislation. From the start, the legislation provided for student financial assistance, although the initial efforts were modest.

Inflation Adjusted Yearly Pell Grant Payout, 1976 to 2013
Inflation Adjusted Yearly Pell Grant Payout, 1976 to 2013

In 1972, Senator Claiborne Pell proposed the Basic Education Opportunity Act which would establish a grant for the neediest students to pursue educational opportunities. These grants were renamed Pell Grants in 1978 and low-income families were awarded up to $3,000 per year to pursue higher education.14 The program has seen immense growth (Figure 3).

Putting this into context of other forms of student loans, we have seen explosive growth of federal involvement in higher education (Figure 4).

The Pell Grant program has been highly politicized, especially by President Jimmy Carter, who equated the volume of Pell Grant and similar programs to persons “academically minded” and dedicated to the future. A large bump in the funding came when President Carter signed the Education Amendments of 1978 and the Middle-Income Student Assistance Act:

The Middle Income Student Assistance Act, which I am also signing today, is similar to the G.I. bill as a landmark in the Federal commitment to aid families with college students…this bill provides more generous Basic Educational Opportunity Grant—Pell grants—to low-income students, and makes eligible students from families with income up to about $25,000. An additional 1.5 million students from middle-income families will be eligible for the Basic Grants program.15

Inflation-Adjusted Growth in the Three Largest Federal Aid Programs, 1997 to 2013
Inflation-Adjusted Growth in the Three Largest Federal Aid Programs, 1997 to 2013

Adjusting for inflation, this $25,000 amount is about $93,500, about four times today’s federal poverty line.16 That is not what the original intentions of the program were.

Federal involvement through loans began in 1958 with the National Defense Education Act, which was a direct loan system with U.S. Treasury-backed loan installments. This Direct Loan program was the forefather to what is now the Direct Loan Program (described in the appendix). In 1965 the Federal Family Education Loan (FFEL) program sought to expand aid to families by providing low interest loans. At this time, the FFEL program was set up as a guarantee on a private or third-party lender. In this way, the federal government did not have to show a 100 percent commitment of these funds in the budget for the year the loan was made, even though the loan would be paid back later with interest.

In 1992, under the George H.W. Bush administration, a report showed that a direct loan program is easier to administer and less costly over time. It was therefore founded in 1992 as a pilot, and in 1993, President Bill Clinton opted to phase it in as a way to reduce the deficit. In 1994, the Republican-dominant Congress hoped to phase out the Direct Loan program, and met resistance from university administrators that favored this program over the cumbersome bureaucratic process that the guaranteed program involved. Congress then passed a law prohibiting the Department of Education from encouraging or mandating a switch to the direct loan program. In effect, this allowed private lenders guaranteed by the FFEL program to gain a stronghold on the loan market, as they were not restrained in their recruiting abilities.

For this reason, the Direct Loan program began to dwindle, hitting a low in 2007. In 2008 the credit market crash had devastating effects on the guaranteed loan program, and many schools out of necessity switched to the direct loan program. In 2009, President Obama proposed ending the FFEL program altogether, and in, 2010 legislation was signed that eliminated the FFEL program. The loans were absorbed into the Direct Loan program. All loans as of July 2010 have been made under the Direct Loan Program.17

Federal aid in the form of tax benefits began in 1997 with the Taxpayer Relief Act which began the Hope Scholarship, Lifetime Learning Tax Credit, Education IRAs, income exclusion for $5,250 in employer education benefits, and tax deduction for up to $2,500 in student loan interest. This program was expanded in 2001 under the Economic Growth and Tax Relief Reconciliation Act of 2001. This EGTRRA bill enhanced education tax credits and renamed Education IRAs as “Coverdell accounts.” The American Recovery and Reinvestment Act of 2009, passed by the House and Senate on February 13, 2009, expanded many forms of student financial aid. The maximum for Pell Grants increased by $500, the Hope Scholarship tax credit increased from $1,800 to $2,500, and Federal Work-Study and AmeriCorps funding increased by $200 million each.18


Part 2: Assessing Federal Assistance

The State of Student Loan Debt in Context

The market for student loans has behaved unlike any other, mechanically and historically. Historically, it has followed its own course relative to other major forms of household debt in the United States. Figure 5 shows that student loans have grown at a compounded annual growth rate of 15.62 percent over the last decade. Therefore, student loan debt, previously relatively small, is now greater than credit card, auto, or home equity forms of loans. Auto and mortgage loans have grown at about 2.06 percent and 3.58 percent annually, respectively. As the burden of student debt has grown tremendously, it is beginning to affect young people’s economic decisions such as when to buy a car or house. Student debt hinders the borrower’s access to credit for a car or house, making the prospect of individual economic progress more daunting than a few generations ago.19

Total Debt Balances
Total Debt Balances
Salaries and Loan Activity by Field of Study
Salaries and Loan Activity by Field of Study

The federal student loan programs are fundamentally unique because any consideration of risk is largely ignored when deciding whether to make a loan. The decision of the course of study a student follows affects earnings and the capacity to repay a loan, yet the federal government lends to engineering, education, and art majors without any consideration for future ability to repay those debts.

Table 1 parses who is taking out loans by major and what students can expect to earn at the beginning and midpoint of their career. By combining data from the NCES’ National Postsecondary Study Aid Survey (NPSAS) and data provided by, we can get rough estimates of how much is being borrowed by major and at those majors’ ability to repay loans. Payscale reports the “Median Starting Salary” and “Median Mid-Career Salary” of 130 majors, and the NPSAS reports estimates of the proportion of students by 24 “Fields of Study” who are borrowing private and federal loans. By organizing the data for the 130 Pay Scale majors into the 24 NPSAS categories (we used only 18) and taking the median earnings of the majors in those categories, we can obtain the median of the median starting and mid-career salaries and look at the proportion of students in those fields who are borrowing. This gives a decent indication of the students’ ability to repay what they have borrowed from the taxpayers.

between borrowing habits of “Engineering and Engineering Technology” students and “Design and Applied Arts” or “Theology and Religious” students. About 7.4 percent of engineering students borrowed $8,500 or more in federal loans (the highest category available from NPSAS). At the same time, 15.6 percent of art students and 19.2 percent of theology students borrowed $8,500 or more—well over double the proportion. The median-median starting salary of engineering students ($59,900) is greater than the median-median mid-career salary of art students ($56,700) and theology students ($51,000). Also, over half of art students and theology students borrowed federal loans whereas 37.4 percent of engineering students did.

While religion and the arts play important roles in society, it is difficult to make an economic argument as to why the federal government should lend to those students on similar terms as highly productive majors such as engineering or computer science. Even President Obama raised the issue when he said “[A] lot of young people no longer see the trades and skilled manufacturing as a viable career. But I promise you, folks can make a lot more, potentially, with skilled manufacturing or the trades than they might with an art history degree.”20 Generally, we note that students in majors with lower earnings potential borrow more, not less, than other students.

Are Student Borrowers Good for It?

Not only have student loan balances grown more than any other major form of debt, but also borrowers have done a worse job of repaying than borrowers of any other form of debt. The amount of student loans that are “seriously delinquent” (90 or more days past due on payments) in the first quarter of 2014 are over five times higher than they were 11 years ago. The amount of serious delinquency in auto and credit card loans in early 2014 has fallen below the level 11 years ago. Furthermore, since the first quarter of 2009, as the country was almost out of the recession (it ended in June 2009) and mortgage delinquencies were at their apex, serious student loan delinquency has grown at a compound annual growth rate of about 3.04 percent while all other forms of seriously delinquent debt declined at almost twice that rate with a compounded annual growth rate of –5.85 percent.

Figure 6 and Figure 7 illustrate the issue of students repaying their loans. During the past 11 years, the number of seriously delinquent student loans has grown by about 15.41 percent per year on average, outpacing those loans that are merely delinquent (fewer than 90 days past due on payments) which averaged 13.54 percent annually. In other words, student loan debt is growing at an unsustainable pace, loan delinquency is a serious problem, and the proportion of delinquent loans that is seriously delinquent is growing. The level of seriously delinquent loan balances is approaching that of delinquent loans (Figure 8). There does not appear to be anything preventing seriously delinquent loan balances from surpassing delinquent loan balances. In fact, legislative provisions allowing for loan forgiveness after a certain number of years will incentivize borrowers not to pay down their debt.

In light of this evidence, some legislators have been pushing reforms that they feel would make debt more affordable and reduce the problems. For example, Senator Kirsten Gillibrand (D-NY) has introduced the Federal Student Loan Refinancing Act because she thinks “[student borrowers] should be able to refinance in the same way that our businesses and homeowners do.”21 She also claims that students’ inability to refinance their loans is “fundamentally unfair and bad for the economy.”22 At this writing, the proposal favored by President Barack Obama and many Democrats is that of Senator Elizabeth Warren (D-MA).23 She has proposed lowering interest rates charged on student loans to pre-2013 borrowers to under 4 percent. Student loans have behaved unlike any other form of debt largely because they have been treated unlike any other type of debt.

Levels of Seriously Delinquent* Debt Balances by Type
Levels of Seriously Delinquent* Debt Balances by Type


Has the Federal Student Loan Program Reduced Inequality?

Recall that the purpose of the federal programs of student financial assistance has been largely to aid poor students. Has the growth in federal loan programs during the past four decades been accompanied by greater attainment for students of modest incomes? No.

Figures 9 and 10 make the point, looking at changes in lower income families’ educational attainment during the history of federal student aid programs. Data was used from Postsecondary Education Opportunity which utilized Census Bureau data. Overall, in 1970 the bottom quartile of families by income accounted for 12 percent of total bachelor’s degrees received by age 24, but those families only accounted for 9.4 percent in 2010. Their share of bachelor’s degrees earned have declined during the past 40 years while the top quartile’s share has risen.

Amount of Delinquent Student Loans
Amount of Delinquent Student Loans
Delinquency Severity Ratio, Seriously Delinquent: Delinquent Balances
Delinquency Severity Ratio, Seriously Delinquent: Delinquent Balances
Share of Bachelor’s Degrees by Family Income Quartile
Share of Bachelor’s Degrees by Family Income Quartile
Bottom Quartile of Family Income: Share of Bachelor’s Degrees by Age 24
Bottom Quartile of Family Income: Share of Bachelor’s Degrees by Age 24


Problems and Solutions

Federal Collegiate Student Financial Assistance Programs: 8 Deficiencies

The evidence amassed above suggests that the federal student financial assistance programs in America have failed to achieve their objectives and are deficient on multiple grounds. Below, we expand upon the factual evidence and go further, documenting eight problems.

First, the Programs Have Contributed Significantly to Rising College Prices and Costs

Then-Education Secretary William J. Bennett declared in 1987 “If anything, increases in financial aid in recent years have enabled colleges and universities to raise their tuitions, confident that Federal loan subsidies would help cushion the increase.”24 The “Bennett Hypothesis” reasons that many students would simply forego going to college because of the costs except for their ability to borrow the money, on a non-commercial basis, from the federal government. Bennett continues to believe that the student loan/ tuition fee relationship holds.25

The Tuition and Enrollment Impact of Student Loans
The Tuition and Enrollment Impact of Student Loans

Before assessing the Bennett hypothesis empirically, consider Figure 11, a depiction of the determinants of tuition fees before and after a low-cost government student financial loan program is created. The preloan program demand for college is represented by curve D1—at lower tuition fees, more students want to attend college. The supply for college is represented by curve S1—the more people will pay, the higher the number of students accepted. The actual tuition fee and enrollment is indicated by point A.

Suppose a large federal student loan program is implemented. At any given tuition fee, more students wish to attend than previously, since financial constraints have temporarily eased. The demand curve shifts to D2. The actual fees and enrollment that will prevail are determined where the quantity demanded equals the quantity supplied, at point B. Tuition fees rise, as Bennett hypothesized, and enrollments also increase.

In the for-profit business sector, a sudden burst in demand might lead to higher sales of goods at higher prices, but soon private-seeking entrepreneurs will expand production, the supply curve will shift to the right, and prices will fall again—maybe even to below where they were originally. The tablet industry is an example. The iPad spawned imitators, somewhat lower prices (adjusting for quality), and greater sales.

In higher education, however, the lack of a profit motive dramatically reduces incentives for supply expansion. Moreover, some schools try to improve reputation by turning customers away, better known as “selective admissions.” When the demand for Harvard increases, enrollment remains the same because supply is fixed. It is possible there might be a small supply response (to S2 in Figure 11), but the new equilibrium price and quantity (point C) still represents higher prices than before the student loan programs began.

Do the data support the possibility that Figure 11 is correct? Yes—over time, tuition fees and enrollments have risen dramatically, although in part for reasons unrelated to federal student loan policies (increased population, rising incomes). Table 2 looks specifically at changes in real tuition fees over time.

Changes in Inflation Adjusted Tuition Fees at American Universities, 1939 to 2014
Changes in Inflation Adjusted Tuition Fees at American Universities, 1939 to 2014

In analyzing the table, a cautionary note is in order: before 1978, the U.S. Bureau of Labor Statistics did not publish annual tuition fee increase data. For the period before 1964, the authors took average tuition increases, with appropriate inflation adjustments, for a typical state university, Purdue University, for which they had data.26 For the 1964 to 1978 period, the authors used federal data reported in the Digest of Education Statistics. 27

Purdue Tuition Fees as a Percent of Indiana per Capita Income, 1939 to 2012
Purdue Tuition Fees as a Percent of Indiana per Capita Income, 1939 to 2012

The most striking thing to observe is that, not only have tuition fees risen after adjusting for inflation, but the rate of increase is rising since 1978.28 Federal involvement in providing student financial assistance is also growing over time. Showing correlation (tuition rate growth and federal student financial assistance increasing) does not show causation, as many other factors are changing during this time period, but the striking positive relationship between federal student aid and tuition growth enhances the probability that Bill Bennett was right.

Before 1978, increases in tuition fees after adjusting for overall inflation were roughly 1 percent a year. In the era of substantial federal student aid after 1978, inflation-adjusted tuition fee increases have ratcheted up to 3-4 percent a year. While this simplistic comparison does not control for other factors, it suggests the possibility that rising federal student financial aid was responsible for most of the increase.

While earlier tuition fee increases were less than the growth in real per-capita income, that would not be the case after 1978. For a while, tuition increases roughly equaled income increases. Figure 12 extends the data from Figure 2 to the present. While the Purdue University in-state tuition had fallen to around 10 percent of per capita income by the 1950s, in the past decade or so the proportion has exploded, going to 26 percent by 2013. The burden of going to Purdue has risen to a point where it was greater than in the Great Depression, 1939 (using the tuition to income ratio as the measure of burden).29

Let us characterize the era 1939 to 1978 as a period of 1 percent annual tuition increases after controlling for inflation. What if that rate of inflation-adjusted tuition increases had continued after 1978 instead of the 3 percent to 3.8 percent increases actually observed? What would tuition fees look like today? About 59 percent lower. State universities charging $10,000 in-state fees (a relatively common fee today) would have instead be charging a bit over $4,000. The elite private universities would have tuition levels around $20,000 instead of over $50,000. In such a world, there would be dramatically smaller need for student loans.

So is the Bennett hypothesis valid? External assessment is somewhat mixed. It should be noted that researchers into this issue mostly have a potentially severe conflict of interest: many of them receive their paychecks from universities, considerably financed by tuition fees.30 A thoughtful and thorough analysis of the matter by Andrew Gillen (a non-university researcher) concluded that, with some qualifications, Bennett was correct.31 Specifically, he quotes Bennett’s 1987 article, “Federal student aid policies do not cause college price inflation, but there is little doubt that they help make it possible.” He concludes “Those words remain just as true today as they were a quarter of a century ago.”32 Gillen refines the Bennett hypothesis and adds some nuances: for example, not all federal student aid has the same impact—Pell Grants have a different impact than guaranteed student loans, for example.

Other studies are mixed regarding the Bennett hypothesis. For example, two highly distinguished scholars (one now the president of the Spencer Foundation and the other the president of Northwestern University) concluded “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.”33 Singell and Stone, however, reach roughly the opposite conclusion: “We find no evidence in support of the Bennett hypothesis among public or lower-ranked private universities… Among the best private universities, though, we find strong evidence of sharp increases in net tuition associated with increases in Pell aid.”34 Looking at for-profit institutions, two National Bureau of Economic Research scholars concluded “[aid-eligible] institutions may indeed raise tuition to capture the maximum grant aid available.”35

The most important and sophisticated study on financial aid policies is an NBER study using a general-equilibrium model.36 Generally supportive of the Bennett hypothesis, it says “private colleges game the federal financial aid system, strategically increasing tuition to increase student aid, and using the proceeds to spend more on educational resources and to compete for high-ability students.” Its analysis relates to three other consequences of federal loan programs: the impact on state university appropriations, the enhanced spending by colleges and universities (the “academic arms race”), and enrollment increases. Regarding the latter, the new NBER study concludes the aggregate enrollment effects of federal public assistance policies are actually quite modest, consistent with our earlier observation that enrollment gains were particularly robust in the pre-federal aid era.37

The gaming of the financial aid system, “using the proceeds to spend more on educational resources…” has helped fueled an academic arms race—ever more spending on a variety of things. Although an elaborate analysis is beyond this paper, over time colleges have enormously increased their staffs, particularly administrative personnel (roughly a doubling on an enrollment-adjusted basis since the 1970s); teaching loads of professors have declined, justified to support more research; facilities have been more luxurious, including fancy recreational centers and posh housing complexes; subsidies of intercollegiate athletics have risen.38

Second, the Programs Have Miserably Failed At Achieving Their Top Goal: Increasing the Proportion of Students Graduating From College from Low-Income Backgrounds

The Higher Education Act and related legislation were passed during a period when the nation’s president declared the War on Poverty. President Johnson, himself a relatively poor Texan first in his family to graduate from college, believed education held the key to economic opportunity and the American Dream.39 In 1965, about one in ten adult Americans had college degrees, and large numbers of low-income Americans had little exposure to higher education.

The reality is that college participation has grown more for students from middle- and upper-income backgrounds than lower-income students. The ultimate measure of success is not gauged by enrollments, but graduation rates. The financial payoff from college comes at the end, upon receipt of a college diploma. The diploma is a signaling device that informs employers that an individual is well-educated, disciplined, intelligent, and possesses other qualities desired in the workplace.

As indicated above, federal data on college participation by income quartiles carefully monitored by Post Secondary Educational Opportunity confirm that the percentage of recent graduates from the bottom quartile of the income distribution has declined since the early days of the student loan and grant programs (Figure 11).40 Recall that in 1970, over 12 percent of recent college graduates came from the bottom quartile of the income distribution, compared with less than 10 percent today.

In our opinion, there are three factors that explain this result. The first one has been suggested: the federal student financial assistance programs have pushed up tuition fees, and that has impacted low-income students more than affluent ones. Students from moderately prosperous backgrounds complain about higher tuition fees, but it deters few of them from college. Low-income students, by contrast, are price-sensitive, so high fees deter them from even applying to college.41 It is true that discounting tuition fees for lower-income students has grown with the rise in sticker prices, but there is evidence that students pay attention to sticker prices in making college decisions—particularly since the price amount of discounting is not known until an application is submitted.42

Second, while conceived as a way to allow poor students to overcome financial barriers, the expansion of federal aid programs has made them into middle-class entitlements. Two expansions in the program, education tax credits and PLUS loans, clearly benefit relatively affluent families. Additionally, eligibility requirements for federal assistance have been loosened to include middle-income students. While a modest 13 percent of federal and state grant aid in 2011-12 for dependent students went to those whose family incomes were above the median ($65,000), if institutional grant aid is included, the proportion rises to 25 percent.43 Institutions have adjusted their need-based aid downward in favor of merit-based scholarships, given the existence of governmental programs like Pell Grants; also, if the Bennett Hypothesis is correct, the ability of institutions to discount fees is contingent on federal student financial assistance. The tuition tax credits and PLUS loan total for the 2012-13 school year is estimated at over $37.5 billion, about 22 percent of all federal student financial assistance.44 It is an exaggeration to state that “federal financial aid programs are targeted almost entirely for low income families.”

Third, while the number of lower-income students who have applied for college has grown, these students have a greater dropout rate than ones from more-affluent families. For example, a study looking at a sample of colleges and universities found that, at flagship state universities, 40 percent of those in the bottom income quartile graduated in four years, compared with 58 percent for those from the top income quartile. Only 17 percent of the top quartile students failed to graduate within six years, compared with 30 percent of those from the bottom quartile.45 The goal of pioneer supporters of federal assistance was not to get poor kids into college, but to get them to graduate and use education to achieve financial success.

The failure of the financial assistance programs to promote educational equality has broader implications for income inequality. Not only has income inequality grown over time, but this has contributed to broader divisions between income classes, which increasingly fail to interact within one another. As Charles Murray so accurately puts it, America is “coming apart.”46

Third, These Programs Have Contributed to a Severe and Growing Problem of Underemployment and Unemployment among Young College Graduates

In 1970, it was rare for a taxi driver to have a college degree. Less than one in every 150 drivers had a diploma. Today, about one of every six holds a degree.47 The number of holders of bachelor’s degrees has grown far faster than the number of well-paying jobs in the managerial, professional, and technical fields where college graduates have traditionally obtained employment. Almost half (48.1 percent) of college graduates hold jobs that the Bureau of Labor Statistics estimate require less than a bachelor’s degree.48 We have, for example, over a million college graduates involved in retail sales, few of which are in positions where such an education is needed.

While the aggregate unemployment rate among college graduates is lower than that of the general population or those with lesser education, that statistic is somewhat suspect. Examination of the data shows that unemployment rates among college graduates under the age of 25 are higher than that of the general population. For example, while the aggregate unemployment rate in 2013 for all workers was 7.4 percent, it was 8.2 percent for college graduates under the age of 25.49 Moreover, as the number of college graduates soars, employers narrow their applicant pool by requiring college degrees for jobs that do not require college-level skills such as a bartender. As a consequence, this has pushed up the unemployment rate for high-school graduates, previously candidates for those positions.

Even if there is “underemployment” of college graduates, that does not mean the federal student financial assistance programs are the sole culprit. Yet virtually every observer would agree that the aggregate effects of the federal programs have been to increase enrollment and the number of graduates. That, in turn, has exacerbated the underemployment problem.

The ability to pay off college loans depends on the ratio of debt to income. The higher the ratio, the more difficult it is to pay off debt. If the Bennett hypothesis is at least partially correct as the evidence suggests, the expansion of federal assistance programs has increased debt because of higher tuition charges. But it may also have led to a reduction in incomes because of the underemployment problem, intensifying the problem that rising student loans have caused.

Fourth, the Federal Student Loan Programs Have Imposed Enormous Financial Burdens on Borrowers and Taxpayers

Earlier in this report, we presented evidence on the growth of various federal programs. The burden of that $1.2 trillion of federal student loan debt is unequally distributed, with the “average” debt (the arithmetic mean) roughly twice as large as the “median” debt (the maximum debt faced by the smallest 50 percent of borrowers). The plus side of that is that for 20 million borrowers, the size of the debt obligation is not large, with minimal annual costs of servicing the debt being no more than $100 or so a month.

Yet there are literally millions of borrowers for whom the debt obligation is above $30,000 a year, and a good proportion of them owe more than a year’s annual income, an amount that is relatively burdensome to service. This has contributed to the rise in loan delinquencies and overdue payments discussed above. Using a broad definition of delinquency (when loan balances are growing), about 30 percent of borrowers have several hundred billion dollars of unmet loan obligations—several thousand dollars per American household. The effort by the Obama administration and some members of Congress to promote loan forgiveness or other easing of credit terms (e.g. retroactively lowering interest rates) creates a moral hazard problem: if it is easy to obtain loan forgiveness or ever more generous repayment terms, there are large incentives not to make loan payments.

It is difficult to measure the outcomes of universities for a variety of reasons. There are multiple services universities provide, and even the core mission of advancing higher forms of learning is difficult to evaluate: do college seniors know more than freshmen, for example, or do they know more today than several decades ago?

That said, there is a fair amount of evidence that colleges are not doing a good job with respect to their core mission of educating Americans. A seminal study of collegiate critical reasoning and writing skills showed that seniors typically are marginally better than freshmen.50 A roughly decennial survey of adult literacy conducted by the National Center for Education Statistics indicates a decline in basic literacy among college graduates. In 1992, 40 percent of college graduates were proficient in prose literacy, but by 2003, the proportion fell to 31 percent.51 The Intercollegiate Studies Institute has given a 60-question multiple choice test of basic civic literacy to more than 28,000 undergraduates at over 80 colleges, and the average score was about 54 percent, with seniors at several elite universities (e.g. Yale, Princeton and Duke) showing lower scores than freshmen.52

Despite some evidence that modern-day students learn no more than earlier generations of students, today’s students receive far higher grades than their predecessors, with student grade point averages rising roughly 0.1 of a point each decade since the 1960s, or 0.5 over time. In 1960 at a wide variety of schools, 15 percent of grades were an “A,” but by 2009 that had risen to 43 percent.53 Moreover, the evidence is clear that students work far less in college than decades ago. As two authors put it, “Full-time students allocated 40 hours per week toward class and studying in 1961, whereas by 2003 they were investing about 27 hours per week.”54

Students “do less for more” in recent years. They receive far higher grades than their grandparents attending college a half-century ago, but learn no more (and probably less). They are studying far less. The record suggests reduced rigor, academic standards, and learning.

But what does this have to do with federal student financial assistance programs? Just because those programs have expanded almost exponentially with the passage of time as academic standards were slipping does not prove a causal relationship.

Every participant in the debate known to us agrees that, on balance, federal student financial assistance programs increase college enrollments and, ultimately, the proportion of Americans with college degrees. Indeed, proponents of these programs can point out that when they were in their infancy (1970), about 11 percent of adult Americans (25 years old or older) had bachelor’s degrees or more. Now, the proportion exceeds 30 percent. A majority of high-school graduates now attempt some form of postsecondary education.

“Higher” education is “higher” because the contents of it historically have required a high level of cognitive ability and discipline to master. As Charles Murray has put it, “For many years, the intellectual benchmark for dealing with college-level material was an IQ of around 115, which demarcates the top 16 percent of the distribution. That was in fact the mean IQ of college graduates during the 1950s. It cannot be nearly that high today…”55 To Murray, the rise in the relatively non-rigorous vocationally oriented university majors in recent decades is a consequence of the decline in the proportion of truly bright individuals attending college.

The dumbing down of the collegiate curriculum has other unintended consequences. Jackson Toby makes an excellent point: with college standards declining, high schools no longer require their students to take a rigorous set of courses.”56 He emphasizes a second reason why expanding federal student financial assistance programs have had a detrimental impact: unlike traditional student financial aid programs administered by institutions, the federal programs have no academic performance standards associated with them. Indeed, a student who takes six years to graduate because of poor academic performance gets roughly twice as much federal aid as one who works hard and graduates in three years. Academic performance and learning success is irrelevant to almost all federal programs, encouraging lower levels of academic effort.

Sixth, Indirectly, the Federal Student Assistance Programs Have Reduced Household Formation and Birth Rates, Enhancing the Aging of the American Population

The law of unintended consequences relates to the federal student assistance programs in many ways, but one that few would have predicted even a generation ago concerns demographic phenomena: household formation, birth rates, and the related issue of home purchases. There has been a marked decline in Americans in their late 20s buying homes with mortgages. Much of this is a consequence of the 2008 financial crisis. But the financial pressures posed by student loans have aggravated the decline (see Table 3).57

Percent of Americans 27 to 30 with Home-Secured Debt, 2003 and 2013
Percent of Americans 27 to 30 with Home-Secured Debt, 2003 and 2013

The decline in taking out mortgages to purchase homes was over 38 percent greater for individuals with student loans than for those without them. Over time, two things are responsible for the greater drop in home activity by student loan debtors than those without those obligations. First, the magnitude of the loans grew dramatically. Second, labor markets have deteriorated for college students, increasing the ratio of debt to income.

As indicated earlier, it is hard to overstate the growth in the proportion of non-mortgage debt that is student loans. It has more than doubled over the past decade; as of the first quarter of 2014, it constituted 31.9 percent of the total non-mortgage debt of $3.48 trillion, and almost 10 percent of all household debt including mortgages.58 It has surpassed credit card, car loan, and home equity loan debt.

The growing financial burden of student debt has led young Americans to defer marriage, buying houses, and having kids. While fertility rates have declined to new lows in general, the decline is more pronounced in younger people. Indeed, birth rates are rising among older women (Figure 13). Typical college graduates, those in their mid-to-late 20s, have had a decline in fertility rates by about 9 percent (from 116.5 to 106.5) in just seven years.59

It is true that the student loan financial crisis is not the main determinant of the fall in the aggregate U.S. fertility rate. Nonetheless, it contributes to the aging of American society and the long-run problems of financing a population where the proportion of individuals in the work force is relatively small and declining.

Change in Fertility Rates of American Women by Age, 2005 to 2012
Change in Fertility Rates of American Women by Age, 2005 to 2012


Seventh, The High Delinquency Rates in the Loan Program Reflect the Non-Market Based Nature of Loans, and the Failure of Colleges to Have “Skin in the Game.”

Student loan interest rates are determined politically, although there is a market-based adjustment process involved. Loan programs do not take the risk of borrowers into account. Much default risk can be predicted in advance based on applicant characteristics and academic major. Federal repayment policies favor students taking low-paying jobs, suggesting low economic productive value, since loan repayments are increasingly limited to a proportion of income and for a maximum number of years. Again, loan forgiveness provisions incentivize students to over borrow and to be laggards in repaying. Additionally, student loans originate in the admissions decisions of colleges and universities. Yet schools whose students do not repay their loans face virtually no adverse consequences. Colleges have no “skin in the game.”60

The federal government has belatedly shown some recognition for the problem, and proposed rules and regulations to impose some consequences on schools for accepting and graduating marginal students. The “gainful employment” rule is an example, removing federal assistance to students if a school has a poor job placement record. But those rules target mainly for-profit institutions and a few public career schools.61 Mediocre public institutions with low graduation rates (often below 20 percent) are excluded. Moreover, those rules do not deal with the root cause of the problem: the indiscriminate provision of federal financial aid to students regardless of student qualifications or performance.

The problems associated with college decisions on admitting and educating students are substantial. Take two schools, both with 1,000 students with federal loans. One, an elite private university, has 40 students who default on their loans, failing to repay on average $25,000 to the government, $1 million total. A second school has 250 students with the same average amount of student debt default, failing to repay over $6 million. If the colleges had to pay 25 percent of the value of the default, the school accepting students who fail disproportionately would have to pay the government $1.3 million more than the school with low loan defaults. With such a rule, schools would be more circumspect about whom they accept.

The U.S. Personal Savings Rate in Three Periods
The U.S. Personal Savings Rate in Three Periods

The federal government promotes access as a holy grail. Its loan and grant policies encourage ill-prepared students to attend that leads to poor performance, dropouts, and loan defaults. They assume large debts and obtain jobs no better than if they skipped college. Those disadvantaged individuals are hurt, not helped, by loose lending policies. Forcing colleges whose students participate in federal loan and grant programs to have “skin in the game” would dramatically reduce this problem.

Eighth, the Federal Aid Programs May Have Contributed to a Decline in the Rate of Savings in the U.S.

Families save for three expenditures in life that cannot be financed out of current earnings: a house, a retirement, and a college education. Rising college attendance should have been a motive for higher personal savings. The reality is that personal savings rates are lower today than in the era of a smaller proportion of students attending college and small or non-existent federal education programs.

Consistent data exists for the last 54 years, which we divide into three 18-year periods (Figure 14). The first period (1960-77) represents years where extensive federal financial aid programs were either non-existent or in embryonic form. The second period (1978-1995) represents years in which the federal programs, while growing rapidly, were moderate in size. The final period (1996-2013) represents years in which the federal programs were large and growing rapidly. The personal savings rate falls sharply over time.

Other factors—age composition of the population, interest rates, foreign financial capital flows, and taxes, among others—impact savings as well. Nonetheless, there is reason to believe the association has a causal nature. Without federal financial aid programs, students are forced to save for college expenses. With those programs, families shift from saving before the fact to students saving after to finance college. Yet expectations often do not pan out as planned. As over 40 percent of freshmen do not graduate in six years, and others obtain jobs paying less than anticipated, a systematic study of the federal student financial aid/ savings relationship is needed.

Reforming the System: Possibilities and Solutions

Six Conditions for Reform

First, changes must not be so disruptive that they impose serious hardships for individuals receiving federal student financial aid. Dramatically restricting loan eligibility so students lose assistance at the beginning of a school year with no warning is cruel and unfair. A new financial aid system must be phased in over several years to allow time for adjustment.

Second, any sensible reform lowers the burden on taxpayers. Some claim that the federal government makes hefty “profits” from student loans because the government borrows money at low interest rates (under 3 percent on long-term government bonds) and then lends out at higher rates.62 That is false.63 The government is not collecting interest due on portions of its loan portfolio (delinquencies, properly defined, are around 30 percent), and some loans are forgiven (e.g. for those entering public-service jobs). Moreover, tens of billions of dollars annually are expended through Pell Grants and other programs. Downsizing the federal financial assistance programs is desirable if the Bennett Hypothesis has any effect. If increased federal aid drives the tuition price explosion, decreases in that aid should dampen it and make college more affordable.

Third, any reform of the financial aid system needs to return to the initial goal of providing assistance to persons who, absent such aid, likely would not attempt college. Aid needs to be “progressive,” going to lower-income students. However, lower-income students typically have weaker educational training for college and have higher dropout rates. Too many people attend college already. The reality of the situation makes a strong case for the federal government to get out of the higher education assistance business. Yet it is politically infeasible to eliminate these programs, and helping the poor achieve is a worthy goal.

Fourth, any revision in financial reform needs to incentivize favorable outcomes and reduce incentives for bad outcomes. For students, that means adverse financial consequences for poor academic performance and severe behavioral problems. Time limits on student aid must be tightened. Bonuses need to be given for superior performance that saves taxpayers money (such as early graduation). While federal student aid needs to be primarily “need based,” an element of merit needs to be introduced to discourage mediocre academic performance.

Fifth, colleges need to have “skin in the game.” Financial consequences should follow from admitting students who, statistically, will do poorly, fail to graduate, and default on loans. Again, tradeoffs are involved. Schools admitting poor students occasionally have success stories where a student receives a degree and excellent postgraduate employment. Tightening admission standards means that some deserving individuals will find college access more difficult. But the overall benefits outweigh these costs. A “skin in the game” rule could worsen the already fragile financial position of some universities. Indeed, it could put some of them out of business. We view “destructive innovation” (Clayton Christensen) or “creative destruction” (Joseph Schumpeter) as good—it makes higher education stronger and more efficient. This free market approach has served America well and should be spread to higher education.

Sixth, reduce the complexity of the current Byzantine-like system. There is some disagreement even as to the number of aid programs, although everyone agrees it is in the double digits. The federal government does three things: it lends students money, it grants them money, and it subsidizes work-study jobs. Eliminate the distinctions between subsidized and unsubsidized loans (all loans, in some sense, are subsidized), Perkins and Stafford loans, and federal work-study jobs. Doing so eliminates horizontalequity concerns about taxing similar people similarly. Simplification can reduce information barriers and confusion.

Limiting the Federal Scope

Many people added to federal student aid programs in recent years do not fall in the bottom quartile of the population. Money goes to promote those with incomes at or above the median, and universities can substitute federal aid in place of their need-based aid programs. Thus, the first target for aid reduction: eliminate programs that disproportionately benefit middle- or higher-income students.

Federal tax credits and PLUS loans account for 22 percent of federal student financial aid and are growing. Tax credits benefit parents, not the students themselves. First instituted on a large scale in the late 1990s, they benefit families with higher incomes. There are two tax credits, the American Opportunity and Lifetime Tax Credits, and two tax deductions, the Tuition and Fees and Loan Interest deductions. The American Opportunity Tax Credit is the most popular and lowers income taxes by $2,500 per student in college each year for up to four years. Since high-income families can claim the full credit, a large portion of the benefits go to parents whose children attend, regardless of the tax credit.

The PLUS loan program also directly assists parents, not students. It has grown dramatically, increasing at 10.8 percent annually in inflation-adjusted terms from the 2002-2003 to the 2012-2013 academic years.64 The program allows moderately affluent parents to borrow at below-market interest rates (currently 6.41 percent, but rising to 7.21 percent shortly) any amount up to the cost of attendance, which is as much as $60,000 a year at some elite private schools. Graduate students can borrow as well. As Janet Lorin of Bloomberg revealed, the terms of PLUS loans border on scandal.65 Parents can opt not to begin to repay the loans until six months after the student graduates. Why? According Lorin, just 45 percent of the $62 billion in active balances on these loans are currently being repaid. Moreover, the U.S. Department of Education plans to loosen credit standards and look at adverse credit histories over two years instead of the current five.

Over half of our goal to reduce federal financial commitment by 40 percent can be achieved by eliminating the tax credits and PLUS loan programs. The remainder can be eliminated by tightening eligibility standards and loan limits for other programs, including Pell Grants. Three types of restriction would be useful:

1. Time limits on aid. Although limits exist, it is hard to justify loan installments for longer than the advertised length of the degree, four years in the case of a bachelor’s degree. Students lingering at colleges for more than four years makes little academic or economic sense. No one should be able to borrow for more than four years for any program, including a Ph.D.

2. Tighter maximum amounts for loans. The federal government should not extend total aid to anyone at the undergraduate level beyond what the tuition levels are at a typical public university, about $8,000 a year. It is not the federal government’s role to finance education to private schools, although students attending such schools should be eligible for assistance up to the limits.

3. Minimal academic standards. Given grade inflation, it is difficult for a diligent student to “flunk out” of American institutions of higher education. Federal aid should not follow a student that transfers to a second school due to poor performance at the first. Similarly, students on Pell Grants, who graduate in three years, could receive a performance bonus of to reward good work and saving taxpayers the cost of a fourth Pell Grant.

Regarding “skin in the game,” numerous programs and criteria could be applied. One would be to tie aid to loan delinquency or student default rates. Suppose the national loan default rate for a four-year program is 12 percent. Schools that have a default rate of 10 percent or more above the national average (13.2 percent in this example) could face a penalty, 25 percent of the defaulted principal above the excluded amount. For example, a school whose students had a 20 percent default rate would have to pay one-fourth of the difference between 20 percent and 13.2 percent, or 1.7 percent of the total principal amount of the defaulted loans. If the school had 15,000 students and an average principal on defaulted loans of $20,000, the school would have to pay over $1 million annually in fines. That fee would be large enough to force schools to reconsider their admission practices, accepting students with higher chances of post-graduate success.

With respect to Pell Grants, a simple non-financial change could have significant effects. Most aid goes to student financial aid offices, basing the amounts on historical experience. The financial aid offices then distribute the grant awards to the students. Why not directly distribute grants to the students, making the Pell Grant into a student voucher? It empowers them by making it clear to the schools that they depend on students for their financial well-being. The voucher can vary in size with income and academic performance.

Finally, the complexity of the student financial aid system is unnecessary and drives students away from applying. Most financial aid decisions are related to a single metric, the student’s family income. Why not have a single need-based grant program, Pell Grants, and a single form of federal student loan?



The federal student financial aid system is failing. It contributes to skyrocketing costs, finances a wasteful academic arms race, weakens academic standards, lowers educational opportunity, and worsens the underemployment/overinvestment problem.

In a perfect world, all federal programs would be abolished. But democratic political processes are imperfect. A realistic solution would drastically downsize the federal programs, making them more progressive (helping lower-income students more than upper-income students) and smaller. By returning the programs to what they were early in their history—modest but useful financial support to truly needy students—it will moderate tuition price inflation and reverse the decline in academic standards. It will contribute to returning the nation to fiscal responsibility while improving the nation’s higher-education system. In short, it is the right thing to do.


Appendix: Federal Student Financial Assistance Programs


Federal Pell Grants: Usually awarded to undergraduate students who have not earned a bachelor’s or a professional degree. Those incarcerated in a federal or state penal institution or are subject to an involuntary civil commitment upon completion of a period of incarceration for a sexual offense are not eligible to receive a Pell Grant. The Pell Grant is based on financial need, cost of attendance, status as a full- or part-time student, plan to attend the full academic year or less, and the maximum award amount changes yearly, with 2014-2015 maximum awards set at $5,730. Funds are allocated through the university a student attends—they are not given directly to the student. Each school participating in the Pell Grant program receives enough funding yearly from the U.S. Department of Education to pay the Federal Pell Grant amounts to eligible students.66

Federal Supplemental Educational Opportunity Grants (FSEOG): FSEOG are grants based on financial need at the time of application, the amount of other aid received, the availability of funds at the applicant’s higher education institution, and range from $100 to $4,000 annual payments. Every year each participating institution is awarded a certain amount of funding to be disbursed to students. After the full amount of FSEOG funds have been awarded to students, no more can be awarded for that year. This is different from the Pell Grant program in that it does not allocate funds to every eligible applicant. Like Pell Grants, the FSEOG grants are institutionally, not individually based.

Teacher Education Assistance for College and Higher Education (TEACH) Grants: The TEACH Grant Program allocates up to $4,000 a year to students who are completing or will complete courses for a career in education. The grant is conditional: recipients will teach in a high-need field at an elementary school, secondary school, or educational service agency that serves students from low-income families for a minimum of four years within eight years after completing the course of study for which the grant is received. If those conditions are not met, the grant converts to a direct unsubsidized loan, with interest accumulating from the time the grant was disbursed. This grant program requires students be in the top 25 percentile on admissions tests and earn a minimum GPA of 3.25. This is a rare example of a federal program with explicit academic performance criteria.

Iraq and Afghanistan Service Grants: The Iraq and Afghanistan Service Grants serve those students who qualify for Pell Grants other than the expected Family Contribution (EFC) portion, but who lost a parent or guardian in the wars with Iraq or Afghanistan. Students must have been enrolled at the time of their parent’s passing and must meet the other Pell Grant requirements. The maximum payment for an Iraq and Afghanistan Service Grant is equal to the maximum Pell Grant award, which in 2014-15 will be $5,730. Due to sequestration, the payment has been reduced by 10 percent, translating to a $573 decline in the maximum award to $5,157. It is paid out in the same fashion as the Pell Grant program.



There are two separate federal student loan programs, the William D. Ford Federal Direct Loan Program and the Federal Perkins Loan Program.

William D. Ford Direct Loans: Broken into four subsets, the lender is the U.S. Department of Education. Since 2010 the Federal Family Education Loan program has been absorbed by the Direct Loan Program. All new loans made after July 2010 have been by the Direct Loan Program. The FFEL program included four programs: Stafford unsubsidized, Stafford subsidized, PLUS Loans and Consolidation loans. The primary difference in recent years was that loans made under the FFEL program were private or third-party lenders making federally guaranteed loans, while under the Direct Loan program the loans were made directly by the U.S. Department of Education.

1. The Direct Subsidized Loan is available to students who demonstrate financial need, and do not accumulate interest (Interest payments are paid by the U.S. Department of Education) until the borrower is outside the six-month grace period allowed after college attendance.

2. The Direct Unsubsidized Loans are available to undergraduate, as well as graduate and professional students, but not contingent on financial need. Unsubsidized Loans accumulate interest at all periods, and if it is unpaid while in school, they will be added to the principal amount of the loan.

3. Direct PLUS Loans are loans made to graduate or professional students, and parents of dependent undergraduate students. These are designed to help pay for education expenses that are not covered by other forms of financial aid. To receive a PLUS loan, the borrower, and in the case of an undergraduate dependent, the student must meet the eligibility requirements for financial aid as determined by the FAFSA. They must also not have adverse credit history. The maximum amount that one can borrow is set at the cost of attendance, as determined by the school, minus any other financial assistance received. These loans face a 4.288 percent origination fee and a 6.41 percent interest rate (rising slightly soon). Interest is collected from parents after the loan is fully disbursed. However, they can request deferment while the student is enrolled at least half-time, and for the same six-month grace period that the subsidized loan faces.

4. Direct Consolidation Loan: This loan is a consolidation of multiple federal education loans into one. This results in a single monthly payment instead of multiple payments. This loan program allows simplified repayment plans; however, they will typically last much longer. The benefits of the individual loans are lost when they are consolidated, including lowered interest rates, rebates, cancellation benefits, which can reduce the cost associated with repaying the loan. Loans eligible to consolidate include:

a. Direct Subsidized Loans

b. Direct Unsubsidized Loans

c. Subsidized Federal Stafford Loans

d. Unsubsidized Federal Stafford Loans

e. Direct PLUS Loans

f. PLUS loans from the Federal Family Education Loan (FFEL) Program

g. Supplemental Loans for Students (SLS)

h. Federal Perkins Loans i. Federal Nursing Loans

j. Health Education Assistance Loans

k. Some existing consolidation loans.

The Federal Perkins Loan Program provides low-interest (5 percent this academic year) student loans for undergraduate and graduate students with “exceptional” financial need. Perkins loans are made available through the school, although not all schools participate in the program. Payment is made back to the school or its loan servicer. Funds depend on their availability at the college (each school gets an allocation from the federal government). The Perkins Loan Program provides up to $5,500 with a maximum borrowing amount of $27,500 as an undergraduate student. Graduate or professional students are eligible to receive up to $8,000 per year, with a maximum of $60,000 including any Perkins loan amounts borrowed as an undergraduate. While in school at least half-time, students have nine months after they graduate, leave school, or drop below half-time status before repayment must being.


Federal Work-Study Jobs

Federal Work-Study provides part-time jobs for undergraduate and graduate students with financial need. The program gives students an opportunity to pay for various education expenses. Work-study award is set at the federal minimum wage, with opportunities to earn more depending on the type of work. The award depends on the time of application, level of need, and level of school funding. Students completing a Federal Work-Study job are limited in the number of hours worked by the school’s financial aid office. They are paid directly by the schools at least monthly. Undergraduate students are paid on an hourly wage schedule, while graduate or professional students are paid by the hour or on a salary basis. This is the one federal program that has had very modest growth over time—the notion that students can finance their education in significant part by working their way through school has fallen out of favor.


Tax Benefits

Two separate tax credits help to offset the costs of college or career school by reducing the amount of income tax owed. The American Opportunity Credit allows up to $2,500 per student per year for the first four years as the student works toward a degree or similar credential. The Lifetime Learning Credit allows up to $2,000 per student per year for any college or career school tuition and fees, as well as for books, supplies, and equipment required for courses that had to be purchased from the school. Those benefits are claimed by parents and are phased out at high levels of income.


Notes and references:

1. There are several good biographies of Lyndon B. Johnson and his political machinations to achieve adoption of the Great Society. See especially Robert Caro’s monumental multi-volume The Years of Lyndon Johnson, the fifth (and most relevant volume here) of which is in preparation. See, however, volume four, The Passage of Power (New York: Alfred A. Knopf, 2012). See also Robert Dallek, Flawed Giant: Lyndon Johnson and His Times, 1961-1973 (New York: Oxford University Press, 1998); a contemporaneous account is Rowland Evans and Robert Novak, Lyndon B. Johnson: The Exercise of Power: A Political Biography (New York: New American Library, 1966)

2. See Daniel L. Bennett, “Myth Busting: The Laissez Faire Origins of American Higher Education,” The Independent Review 18 (Spring 2014), pp. 503-526 for a more elaborate historical discussion.

3. A much smaller, but enduring, program was the Fulbright program, named after Senator J. William Fulbright. That program involving international exchanges of students and scholars, continues to this day, but is relatively small in terms of its impact on student aid.

4. The calculations here are from the National Center for Education Statistics, Digest of Education Statistics, published annually. There is a table at the beginning of Chapter 3 in each year’s Digest with a historical summary of key higher education variables going back to the 1869-70 school year.

5. Marcus Stanley, “College Education and the Midcentury GI Bills,” Quarterly Journal of Economics 118 (May 2003): 671-708.

6. James B. Conant, “The Future of Our Higher Education,” Harper’s, May 1938, p. 563.

7. See Mark David Van Ells, To Hear Only Thunder Again: America’s World War II Veterans Come Home (Lanham, MD: Lexington Books, 2001), p. 135. Conant later issued a more favorable assessment of the GI Bill.

8. Robert M. Hutchins, “The Threat to American Education,”Collier’s 114 (December 30, 1944), p. 21.

9. Seymour E. Harris, The Market for College Graduates (Cambridge, MA: Harvard University Press, 1949). We are indebted to retired Rutgers sociology professor Jackson Toby for bringing this to our attention.

10. Statistical Abstract of the United States: 1970 (Washington, D.C.: Government Printing Office, 1969), p. 133.

11. “Trends in Higher Education,” The College Board: Federal Aid per Recipient by Program in Current and Constant Dollars over Time,

12. See Robert E. Martin and Andrew Gillen, How College Pricing Undermines Financial Aid (Washington, D.C.: Center for College Affordability and Productivity, March 2011) for a good discussion of several of the issues relating to financial aid.

13. Lyndon B. Johnson: “Remarks at Southwest Texas State College Upon Signing the Higher Education Act of 1965,” November 8, 1965. Online by Gerhard Peters and John T. Woolley, The American Presidency Project, http://www.

14. “Federal Pell Grant Program.” U.S. Department of Education.

15. Jimmy Carter: “Education Amendments of 1978 and the Middle Income Student Assistance Act Statement on Signing H.R. 15 and S. 2539 Into Law.” November 1, 1978. Online by Gerhard Peters and John T. Woolley, The American Presidency Project,

16. The Federal Poverty line is contingent upon size of household. For an average family of four, the poverty line is $23,850 in 2014 dollars. For each additional member of the household add $4,060. U.S. Department of Health and Human Services, 2014 Poverty Guidelines.

17. For more analysis, visit New America Foundation, Federal Education Budget Project, Federal Student Loan Programs—History at:

18. “History of Student Financial Aid.” FinAid! The SmartStudent Guide to Financial Aid, educators/history.phtml.

19. Meta Brown, Donghoon Lee, Wilbert van der Klaauw, Andrew Haughwout, and Joelle Scally. “Measuring Student Debt and Its Performance,” Federal Reserve Bank of New York Staff Reports, 668, April 2014.

20. Scott Jaschik “Obama vs. Art History.” Inside Higher Ed. January 31, 2014. Accessed at http://www.

21. Eric Anderson “Gillibrand: End Heavy Student Loan Burden,” Albany Times Union. February 18, 2014, accessed at

22. Eric Anderson “Student Loan Bill Attracts Interest,” Albany Times Union, May 13, 2014, accessed at http://www.

23. “Obama to Endorse Elizabeth Warren’s Student Loan Proposal,” Huffington Post, June 6, 2014, accessed at: http://

24. William J. Bennett, “Our Greedy Colleges,” New York Times, February 18, 1987.

25. William J. Bennett and David Wilezol, Is College Worth It? (Nashville, TN: Thomas Nelson, 2013)

26. The Purdue data used prior to 1964 tracts closely with tuition fee increases for three private schools (Princeton, Chicago and Vanderbilt) over the period 1948 to 1965. Average tuition fees rose 215.9 percent at the three private schools, compared to 200 percent at Purdue. See William G. Bowen, The Economics of the Major Private Universities (Berkeley, CA: Carnegie Commission on Higher Education,, 1968), p. 65 for the private school data; the Purdue data were provided to the authors by Purdue University. Bowen anticipates the Bennett Hypothesis some 20 years earlier. First noting that tuition fees had risen a lot in the 1950s and early 1960s, Bowen says” “Will tuition continue to increase at this rate? The answer depends in part –and probably in large part –on the future of student aid programs.” Ibid., p. 36.

27. National Center for Education Statistics, Digest of Education Statistics: 1994 (Washington, D.C.: U.S. Government Printing Office, 1994), Table 304. We used a average of private and public tuition fee increases weighted by enrollment

28. It appears from data for Purdue University, that inflation-adjusted tuition fees actually fell from 1939 to about the mid-1950s, and then rose robustly in the decade from 1954-64.

29. Purdue seems fairly typical of American public universities in this regard. Current Purdue president Mitch Daniels seems very aware of the rising burden of college, and has aggressively dealt with it. The class of 2016, entering in 2012, will be the first in decades to have seen no tuition increases the entire four years of school.

30. The authors of this study are not dependent on universities for their income, with one trivial exception (the senior author receives about 1 percent of his income from part-time teaching, and a small additional amount in honoraria from occasional lectures given at universities). Federal student aid did not significantly contribute to the finances of the other authors.

31. Andrew Gillen, Introducing Bennett Hypothesis 2.0 (Washington, D.C.: Center for College Affordability and Productivity, February 2012), accessible at . See also, Martin and Gillen, “How College Pricing Undermines Financial Aid.”

32. Ibid., p. 26.

33. Michael S. McPherson and Morton Owen Schapiro, The Student Aid Game: Meeting Need and Rewarding Talent in American Higher Education (Princeton, NJ: Princeton University Press, 1998).

34. Larry D. Singell and Joe A. Stone, “For Whom the Pell Tolls: Market Power, Tuition Discrimination and the Bennett Hypothesis,” University of Oregon Economics Working Paper No. 2003-12, April 2003.

35. Stephanie Riegg Cellini and Claudia Goldin, “Does Federal Student Aid Raise tuition: New Evidence on For-Proift Colleges,” National Bureau of Economic Research Working Paper 17827, February 2012.

36. Dennis Epple, Richard Romano, Sinan Sarpca, and Holger Sieg, “The U.S. Market for Higher Education: A General Equilibrium Analysis of State and Private Colleges and Public Funding Policies,” National Bureau of Economic Research Working Paper No. 19298, August 2013.

37. In a much quoted book, The Race Between Education and Technology (Cambridge, MA: Harvard University Press,2008) Claudia Goldin and Lawrence F. Katz argue that the slowdown in enrollment growth and college degrees created after about 1975 has led to slower U.S. economic growth and greater inequality. It is interesting that the slowdown roughly coincides with the vast expansion of federal student financial aid programs. While we are dubious that Goldin and Katz are correct in their overall conclusion, the ineffectiveness of the federal aid programs in expanding enrollments might be viewed as a contributing factor in declining U.S. growth and rising income inequality.

38. See Richard K. Vedder, Going Broke By Degree: Why College Costs Too Much (Washington, D.C.: AEI Press, 2004) for some evidence.

39. Johnson’s maternal great-grandfather did attend college (the University of Alabama), but never graduated.

40. Data provided by Post Secondary Educational Opportunity

41. Some of the evidence in the Epple, Romano, Scarpa and Sieg NBER study cited above is consistent with this; the overall enrollment effects of student assistance programs are probably moderately positive, but low-income students are more likely to be turned off by higher fees. The differential price elasticity of demand by income group is very explicitly noted by Michael S. McPherson and Morton Owen Schapiro. See their “Does Student Aid Affect College Enrollment? New Evidence on a Persistent Controversy, American Economic Review, 81 (March 1991), pp. 309-318.

42. Hoxby, Caroline and Avery, Christopher. “The Missing ‘One-Offs’: The Hidden Supply of High-Achieving, Low-Income Students.” Brookings Institute, Brookings Papers on Economic Activity, Spring 2013 http://www.brookings. edu/~/media/projects/bpea/spring%202013/2013a_hoxby.pdf. Hoxby and Avery show that Low-Income students “Slightly disfavor schools with higher sticker prices and do not have a preference for net costs that is statistically different from zero.”

43. College Board, Trends in Student Financial Aid: 2013 (New York: College Board, 2013), p. 30, and authors’ calculations..

44. Ibid., p. 10.

45. William G. Bowen, Matthew M. Chingos and Michael S. McPherson, Crossing the Finish Line: Completing College at America’s Public Universities (Princeton, NJ: Princeton University Press, 2009), p. 39.

46. Charles A. Murray, Coming Apart: The State of White America, 1960-2010 (New York: Crown Forum Books, 2012).

47. Richard Vedder, Christopher Denhart, and Jonathan Robe, Why Are Recent College Graduates Underemployed? University Enrollments and Labor-Market Realities (Washington, D.C.: Center for College Affordability and Productivity, January 2013, p.24.

48. Ibid. p. 12.

49. We thank our colleague Daniel Garrett for his assistance in obtaining age and education specific unemployment rates from the Bureau of Labor Statistics data base. The numbers here are based on Heidi Shierholz, Alyssa Davis and Will Kimball, “The Class of 2014: The Weak Economic Is Idling Too Many Young Graduates (Washington, D.C.: Economic Policies Institute, May 1, 2014).

50. Richard Arum and Josipa Roksa, Academic Adrift: Limited Learning on College Campuses (Chicago, IL: University of Chicago Press, 2011).

51. U.S. Department of Education, A Test of Leadership: Charting the Future of U.S. Higher Education: A Report of the Commission Appointed by Secretary of Education Margaret Spellings (Washington, D.C., 2006), p. 14.

52. Intercollegiate Studies Institute, Enlightened Citizenship: How Civic Knowledge Trumps a College Degree in Promoting Active Civic Engagement (Wilmington, DL:, 2011), accessed on May 20, 2014 at http://www.

53. Studies are voluminous on this. See, for example, Stuart Rojstaczer and Christopher Healy,” Where A Is Ordinary: The Evolution of American College and University Grading, 1940-2009,” Teachers College Record 114 (July 2012), pp. 1-23. Another excellent article is Philip Babcock, “Real Costs of Nominal Grade Inflation: New Evidence from Student Course Evaluations, Economic Inquiry 48 (October 2010), pp. 983-996.

54. Philip S. Babcock and Mindy Marks, “The Falling Time Cost of College: Evidence from Half a Century of Time Use Data,” National Bureau of Economic Research, NBER Working Paper No.15954, issued April 2010.

55. Charles Murray, Real Education: Four Simple Truths for Bringing America’s Schools Back to Reality (New York: Crown Forum, 2008), p. 69.

56. Jackson Toby, The Lowering of Higher Education in America: Why Financial Aid Should Be Based on Student Performance (Santa Barbara, CA: Praeger, 2010).

57. The authority on this issue is the Federal Reserve Bank of New York. See, for example, their Quarterly Report of Household Debt and Credit (New York, February 2014). For an excellent summary of the findings, see Jonnelle Marte, “That’s a Drag: Debt Weighs Down Graduates,” Washington Post, May 16, 2014, p. A15.

58. See, for example, the aforementioned Quarterly Report of the New York Fed. A paper with lots of graphical evidence, see Mark Huelsman and Sean McElwee, “You Can Blame Student Debt for America’s Inequality and Shrinking Middle Class,”Quartz, February 20, 2014, accessed on that date at .

59. See U.S. Department of Health and Human Services, National Center for Health Statistics, National Vital Statistics Reports, December 30, 2013, Births: Final Data for 2012

60. Alex J. Pollack has written cogently about this. See his “Fixing Student Loans: Let’s Give Colleges Some ‘Skin in the Game,” The American, January 26, 2012, accessed on August 21, 2014 at january/fixing-student-loans-lets-give-the-colleges-some-skin -in-the-game/

61. These rules have been proposed and changed several times, partly in response to successful judicial challenges of them. The narrow nature of them is reflected in the title of one recent U.S. Department of Education news release “Obama Administration Takes Action to Protect Americans from Predatory, Poor-Performing Career Colleges,” issued on March 14, 2014 and accessed by us on May 22, 2014 at . The anti-business attitude of the rules is revealed in the title: somehow, some for-profit career colleges are “predatory,” but state universities with very low graduation rates (e.g., Chicago State University, University of Texas at San Antonio) are not

62. Ralph Atkins and Michael MacKenzie, “US Government Borrowing Costs Fall to Fresh Lows,” Financial Times. May 28, 2014. Accessed at:

63. “Fair-Value Estimates of the Cost of Federal Credit Programs in 2013” Congressional Budget Office. Washington D.C. June 2012.

64. Authors’ calculations from The College Board, Trends in Student Aid:2013, p. 10.

65. Janet Lorin, “Alarm Raised by Plan to Ease Credit Norms on U.S. Parent Loans,” Bloomberg, May 27, 2014, accessed on June 6, 2014 at . 66. The information in this primer comes from Federal Student Aid within the U.S. Department of Education: https://