A political storm is brewing in Washington over the consequences of rising college costs, but few politicians are talking about the causes of the problem, particularly since a major perpetrator of it is Washington itself. Using his exquisitely honed rhetorical skills, President Obama panders to his political base by suggesting that the solution to ever-mounting student-loan debt lies in easing repayment terms; Senator Elizabeth Warren proposes lowering interest rates for many past borrowers.
The reality is, however, that the single biggest cause of this financial problem, and a contributor to many other weaknesses in our economy, is the dysfunctional, byzantine system of federal financial assistance for college students. College-aid programs have grown rapidly over the last several decades, with annual federal assistance in the form of grants, loans, tax credits, and work-study now totaling $170 billion a year, compared with about $10 billion (in today’s dollars) in 1970. These programs were designed to make college more affordable for all and to increase the proportion of college graduates coming from lower-income groups. On both counts, they have failed.
Let me discuss eight problems with our complex system of federal financial aid for students. A fuller elaboration can be found in Dollars, Cents, and Nonsense: The Harmful Effects of Federal Student Aid, a new study for the Center for College Affordability and Productivity that I co-authored with Christopher Denhart and Joseph Hartge. (It is viewable at CenterForCollegeAffordability.org.)
First, these programs have contributed significantly to rising prices and costs. In 1987, Education Secretary William Bennett correctly declared: “If anything, increases in financial aid . . . have enabled colleges and universities to raise their tuitions, confident that federal loan subsidies would help cushion the increase.” This is evident in the table below.
Education Statistics, College Board, Bureau of the Census
*Average for time period; 2014 is partially estimated
Before the late 1970s, federal financial-aid programs were modest in size or nonexistent. College-tuition fees usually rose roughly one percentage point annually more than overall inflation. Since 1978, in the era of rapidly growing federal financial-assistance programs, annual tuition increases have been 3 to 4 percent a year beyond the inflation rate. These numbers do not control for other contributing factors, but a fair amount of good research supports Bennett’s fundamental proposition.
Suppose tuition-fee inflation since 1978 had remained at the l percent rate. Where would tuition fees be today? About 59 percent lower. The state school charging nearly $10,000 would instead charge around $4,000. The elite private university with a $45,000 tuition fee instead would be charging a bit over $18,000. The burden of financing a college education would be dramatically lighter, and the need to borrow for college would be much smaller.
In other words, the student-loan “crisis” is mainly the result of the extraordinary growth in university fees. As universities have captured extra tuition money, they have engaged in a costly academic arms race: building expensive but under-utilized buildings (which are nearly empty much of the year), lowering teaching loads so that faculty can write more papers, hiring an army of administrators who are unrelated to the primary academic mission, and paying presidents million-dollar salaries.
Second, financial-aid programs have failed to increase the proportion of college graduates from low-income backgrounds. As is evident in the graph below, federal data on college graduates classified by income quartile show that a smaller proportion of recent graduates comes from the bottom quartile today (less than 10 percent) than did so in 1970 (more than 12 percent), before federal financial-aid programs became so large. As tuition fees have soared, middle- and upper-income kids complain but pay the fees, while lower-income kids are scared off by sticker shock and the complicated financial-aid forms that applicants for assistance must complete. Often they do not even apply to college.
Third, the federal programs have contributed to severe underemployment and unemployment of recent college graduates. Bureau of Labor Statistics data suggest that almost half of the nation’s college graduates are in jobs that do not require a bachelor’s degree. We have over a million retail salespersons with college degrees. The 2013 unemployment rate of recent graduates under age 25 exceeded that of the general population. The earnings differential between high-school and college graduates has fallen since 2006. Federal programs have been successful at one thing: increasing enrollment. We have too many people going to college for the limited number of jobs that require such education.
#page#Fourth, the federal programs have imposed enormous financial burdens on students, borrowers, and taxpayers. Student-loan programs have caused tuition to rise to an unprecedented fraction of income. At Purdue University in the depression year of 1939, tuition was $110, about 21 percent of per capita income in Indiana. Today it is about 26 percent. The burden of buying virtually everything else in society (excepting perhaps medical care) has fallen relative to income. Using conventional financial standards, the New York Federal Reserve Bank has found that close to 30 percent of student loans are now delinquent, and loan balances are growing. The notion that the federal government makes money on student loans (borrowing from, say, Chinese bondholders at 3 percent and lending money out at 5 to 6 percent) is an artifact of inappropriate government accounting. After correcting for risks — losses associated with large loan defaults or forgiveness provisions — these programs are a burden on taxpayers, as are the tens of billions spent annually on Pell Grants and other subsidies.
Fifth, financial assistance has lowered the quality of instruction. Good evidence exists that students typically improve their critical-reasoning and writing capacities only slightly while in college, and more college students are only marginally literate today than were a generation or more ago. Students study far less, but get much higher grades, than they did 50 years ago. Aid programs have induced more academically marginal students to attend college, leading to a dumbing-down of instructional content. The lack of academic-performance standards in disbursing federal assistance has exacerbated this problem.
Sixth, student aid has probably reduced household formation and birth rates. Birth rates have fallen sharply in recent years for Americans under 30, but not for other age groups. This is partially because the burden of college debt is leading to deferral of marriage and parenthood. Data from the Federal Reserve Bank of New York show a decline of more than 30 percent from 2003 to 2013 in the proportion of student-loan borrowers with home-secured debt, a bigger decline than for other categories of borrowers. Rising college costs have contributed at least modestly to sluggish home construction.
Seventh, federal lending programs for students suffer from high delinquency rates, partly because colleges have no “skin in the game.” Delinquency and default rates are higher for student loans than for other forms of lending, such as home-equity or car loans. Partly this reflects their virtually nonexistent credit standards, which allow students with a high probability of dropping out of college to borrow on the same terms as good students. Colleges are complicit in this; they push kids into borrowing, but face no financial consequences if large numbers default on loans.
Eighth, financial assistance for students may have contributed to declining personal savings in the U.S. Historically, people saved for college. So, since more kids go to college nowadays, Americans’ personal savings should presumably have increased. But have they? In the 1960s, Americans saved 8 percent or more of their income. Since the mid 1990s, with vastly expanded aid programs and more students in school, the savings rate has fallen to under 4 percent. Student-loan programs have reduced the need to save before college, probably contributing to the savings-rate decline.
In an ideal world, we would shut down these federal programs, but that is probably politically infeasible. Still, a large (say, 40 percent) reduction in the programs’ spending could be devised in a way that conservatives could love and liberals could like or at least tolerate: keep all the aid that helps truly low-income students, but wipe out assistance benefiting the more affluent, such as tuition tax credits and PLUS loans. Introduce academic-performance standards into the programs. Allow for innovative private financing arrangements, such as income-share agreements, in which students sell an equity interest in their earnings instead of taking on debt. Make colleges share the loss if students don’t pay their debts — this is hugely important. And, most of all, avoid costly short-run palliatives like the Obama-Warren schemes, which not only punish responsible borrowers but invite bigger loan-default problems in the future.
– Mr. Vedder directs the Center for College Affordability and Productivity, teaches at Ohio University, and is an adjunct scholar at the American Enterprise Institute.