Bad Education: The Student Debt Complex

This piece is cross-posted from the journal N+1 where it originally appeared

The Project On Student Debt estimates that the average college senior in 2009 graduated with $24,000 in outstanding loans. Last August, student loans surpassed credit cards as the nation’s single largest source of debt, edging ever closer to $1 trillion. Yet for all the moralizing about American consumer debt by both parties, no one dares call higher education a bad investment. The nearly axiomatic good of a university degree in American society has allowed a higher education bubble to expand to the point of bursting.

Since 1978, the price of tuition at US colleges has increased over 900 percent, 650 points above inflation. To put that number in perspective, housing prices, the bubble that nearly burst the US economy,  then the global one, increased only fifty points above the Consumer Price Index during those years. But while college applicants’ faith in the value of higher education has only increased, employers’ has declined. According to Richard Rothstein at The Economic Policy Institute, wages for college-educated workers outside of the inflated finance industry have stagnated or diminished. Unemployment has hit recent graduates especially hard, nearly doubling in the post-2007 recession. The result is that the most indebted generation in history is without the dependable jobs it needs to escape debt.

What kind of incentives motivate lenders to continue awarding six-figure sums to teenagers facing both the worst youth unemployment rate in decades and an increasingly competitive global workforce?

During the expansion of the housing bubble, lenders felt protected because they could repackage risky loans as mortgage-backed securities, which sold briskly to a pious market that believed housing prices could only increase. By combining slices of regionally diverse loans and theoretically spreading the risk of default, lenders were able to convince independent rating agencies that the resulting financial products were safe bets. They weren’t. But since this wouldn’t be America if you couldn’t monetize your children’s futures, the education sector still has its equivalent: the Student Loan Asset-Backed Security (or, as they’re known in the industry, SLABS).

SLABS were invented by then-semi-public Sallie Mae in the early ’90s, and their trading grew as part of the larger asset-backed security wave that peaked in 2007. In 1990, there were $75.6 million of these securities in circulation; at their apex, the total stood at $2.67 trillion. The number of SLABS traded on the market grew from $200,000  in 1991 to near $250 billion by the fourth quarter of 2010. But while trading in securities backed by credit cards, auto loans, and home equity is down 50 percent or more across the board, SLABS have not suffered the same sort of drop. SLABS are still considered safe investments—the kind financial advisors market to pension funds and the elderly.

With the secondary market in such good shape, primary lenders have been eager to help students with out-of-control costs. In addition to the knowledge that they can move these loans off their balance sheets quickly, they have had another reason not to worry: federal guarantees. Under the just-ended Federal Family Education Loan Program (FFELP), the US Treasury backed private loans to college students. This meant that even if the secondary market collapsed and there were an anomalous wave of defaults, the federal government had already built a lender bailout into the law. And if that weren’t enough, in May 2008 President Bush signed the Ensuring Continued Access to Student Loans Act, which authorized the Department of Education to purchase FFELPloans outright if secondary demand dipped. In 2010, as a cost-offset attached to health reform legislation, President Obama ended the FFELP, but not before it had grown to a $60 billion-a-year operation.

Even with the Treasury no longer acting as co-signer on private loans, the flow of SLABSwon’t end any time soon. What analysts at Barclays Capital wrote of the securities in 2006 still rings true: “For this sector, we expect sustainable growth in new issuance volume as the growth in education costs continues to outpace increases in family incomes, grants, and federal loans.” The loans and costs are caught in the kind of dangerous loop that occurs when lending becomes both profitable and seemingly risk-free: high and increasing college costs mean students need to take out more loans, more loans mean more securities lenders can package and sell, more selling means lenders can offer more loans with the capital they raise, which means colleges can continue to raise costs. The result is over $800 billion in outstanding student debt, over 30 percent of it securitized, and the federal government directly or indirectly on the hook for almost all of it.

If this sounds familiar, it probably should, and the parallels with the pre-crisis housing market don’t end there. The most predatory and cynical subprime lending has its analogue in for-profit colleges. Inequalities in US primary and secondary education previously meant that a large slice of the working class never got a chance to take on the large debts associated with four-year degree programs. For-profits like The University of Phoenix or Kaplan are the market’s answer to this opportunity.

While the debt numbers for four-year programs look risky, for-profit two-year schools have apocalyptic figures: 96 percent of their students take on debt and within fifteen years 40 percent are in default. A Government Accountability Office sting operation in which agents posed as applicants found all fifteen approached institutions engaged in deceptive practices and four in straight-up fraud. For-profits were found to have paid their admissions officers on commission, falsely claimed accreditation, underrepresented costs, and encouraged applicants to lie on federal financial aid forms. Far from the bargain they portray themselves to be on daytime television, for-profit degree programs were found to be more expensive than the nonprofit alternatives nearly every time. These degrees are a tough sell, but for-profits sell tough. They spend an unseemly amount of money on advertising, a fact that probably hasn’t escaped the reader’s notice.

But despite the attention the for-profit sector has attracted (including congressional hearings), as in the housing crisis it’s hard to see where the bad apples stop and the barrel begins. For-profits have quickly tied themselves to traditional powers in education, politics, and media. Just a few examples: Richard C. Blum, University of California regent (and husband of California Sen. Dianne Feinstein), is also through his investment firm the majority stakeholder in two of the largest for-profit colleges. The Washington Post Co. owns Kaplan Higher Education, forcing the company’s flagship paper to print a steady stream of embarrassing parenthetical disclosures in articles on the subject of for-profits. Industry leader University of Phoenix has even developed an extensive partnership with GOOD magazine, sponsoring an education editor. Thanks to these connections, billions more in advertising, and nearly $9 million in combined lobbying and campaign contributions in 2010 alone, for-profits have become the fastest growing sector in American higher education.

If the comparative model is valid, then the lessons of the housing crash nag: What happens when the kids can’t pay? The federal government only uses data on students who default within the first two years of repayment, but its numbers have the default rate increasing every year since 2005. Analyst accounts have only 40 percent of the total outstanding debt in active repayment, the majority being either in deferment or default. Next year, the Department of Education will calculate default rates based on numbers three years after the beginning of repayment rather than two. The projected results are staggering: recorded defaults for the class of 2008 will nearly double, from 7 to 13.8 percent. With fewer and fewer students having the income necessary to pay back loans (except by taking on more consumer debt), a massive default looks closer to inevitable.

Unlike during the housing crisis, the government’s response to a national wave of defaults that could pop the higher-ed bubble is already written into law. In the event of foreclosure on a government-backed loan, the holder submits a request to what’s called a state guaranty agency, which then submits a claim to the feds. The federal disbursement rate is tied to the guaranty agency’s fiscal year default rate: for loans issued after October 1998, if the rate exceeds 5 percent, the disbursement drops to 85 percent of principal and interest accrued; if the rate exceeds 9 percent, the disbursement falls to 75 percent. But the guaranty agency rates are computed in such a way that they do not reflect the rate of default as students experience it; of all the guaranty agencies applying for federal reimbursement last year, none hit the 5 percent trigger rate.

With all of these protections in place, SLABS are a better investment than most housing-backed securities ever were. The advantage of a preemptive bailout is that it can make itself unnecessary: if investors know they’re insulated from risk, there’s less reason for them to get skittish if the securities dip, and a much lower chance of a speculative collapse. The worst-case scenario seems to involve the federal government paying for students to go to college, and aside from the enrichment of the parasitic private lenders and speculators, this might not look too bad if you believe in big government, free education, or even Keynesian fiscal stimulus. But until now, we have only examined one side of the exchange. When students agree to take out a loan, the fairness of the deal is premised on the value for the student of their borrowed dollars. If an 18-year-old takes out $200,000 in loans, he or she better be not only getting the full value, but investing it well too.

Higher education seems an unlikely site for this kind of speculative bubble. While housing prices are based on what competing buyers are willing to pay, postsecondary education’s price is supposedly linked to its costs (with the exception of the for-profits). But the rapid growth in tuition is mystifying in value terms; no one could argue convincingly the quality of instruction or the market value of a degree has increased ten-fold in the past four decades (though this hasn’t stopped some from trying). So why would universities raise tuition so high so quickly? “Because they can” answers this question for home-sellers out to get the biggest return on their investments, or for-profits out to grab as much Pell Grant money as possible, but it seems an awfully cynical answer when it comes to nonprofit education.

First, where the money hasn’t gone: instruction. As Marc Bousquet, a leading researcher into the changing structures of higher education, wrote in How The University Works(2008):

If you’re enrolled in four college classes right now, you have a pretty good chance that one of the four will be taught by someone who has earned a doctorate and whose teaching, scholarship, and service to the profession has undergone the intensive peer scrutiny associated with the tenure system. In your other three classes, however, you are likely to be taught by someone who has started a degree but not finished it; was hired by a manager, not professional peers; may never publish in the field she is teaching; got into the pool of persons being considered for the job because she was willing to work for wages around the official poverty line (often under the delusion that she could ‘work her way into’ a tenurable position); and does not plan to be working at your institution three years from now.

This is not an improvement; fewer than forty years ago, when the explosive growth in tuition began, these proportions were reversed. Highly represented among the new precarious teachers are graduate students; with so much available debt, universities can force graduate student workers to scrape by on sub-minimum-wage, making them a great source of cheap instructional labor. Fewer tenure-track jobs mean that recent PhDs, overwhelmed with debt,  have no choice but to accept insecure adjunct positions with wages kept down by the new crop of graduate student-workers. Rather than producing a better-trained, more professional teaching corps, increased tuition and debt have enabled the opposite.

If overfed teachers aren’t the causes or beneficiaries of increased tuition (as they’ve been depicted of late), then perhaps it’s worth looking up the food chain. As faculty jobs have become increasingly contingent and precarious, administration has become anything but. Formerly, administrators were more or less teachers with added responsibilities; nowadays, they function more like standard corporate managers—and they’re paid like them too. Once a few entrepreneurial schools made this switch, market pressures compelled the rest to follow the high-revenue model, which leads directly to high salaries for in-demand administrators. Even at nonprofit schools, top-level administrators and financial managers pull down six- and seven-figure salaries, more on par with their industry counterparts than with their fellow faculty members. And while the proportion of tenure-track teaching faculty has dwindled, the number of managers has skyrocketed in both relative and absolute terms. If current trends continue, the Department of Education estimates that by 2014 there will be more administrators than instructors at American four-year nonprofit colleges. A bigger administration also consumes a larger portion of available funds, so it’s unsurprising that budget shares for instruction and student services have dipped over the past fifteen years.

When you hire corporate managers, you get managed like a corporation, and the race for tuition dollars and grants from government and private partnerships has become the driving objective of the contemporary university administration. The goal for large state universities and elite private colleges alike has ceased to be (if it ever was) building well-educated citizens; now they hardly even bother to prepare students to assume their places among the ruling class. Instead we have, in Bousquet’s words, “the entrepreneurial urges, vanity, and hobbyhorses of administrators: Digitize the curriculum! Build the best pool/golf course/stadium in the state! Bring more souls to God! Win the all-conference championship!” These expensive projects are all part of another cycle: corporate universities must be competitive in recruiting students who may become rich alumni, so they have to spend on attractive extras, which means they need more revenue, so they need more students paying higher tuition. For-profits aren’t the only ones consumed with selling product. And if a humanities program can’t demonstrate its economic utility to its institution (which can’t afford to haul “dead weight”) and students (who understand the need for marketable degrees), then it faces cuts, the neoliberal management technique par excellence. Students apparently have received the message loud and clear, as business has quickly become the nation’s most popular major. 

When President Obama spoke in the State of the Union of the need to send more Americans to college, it was in the context of economic competition with China, phrased as if we ought to produce graduates like steel. As the near-ubiquitous unpaid internship for credit (in which students pay tuition in order to work for free) replaces class time, the bourgeois trade school supplants the academy. Parents understandably worried about their children make sure they never forget about the importance of an attractive résumé. It was easier for students to believe a college education was priceless when it wasn’t bought and sold from every angle.

If tuition has increased astronomically and the portion of money spent on instruction and student services has fallen, if the (at very least comparative) market value of a degree has dipped and most students can no longer afford to enjoy college as a period of intellectual adventure, then at least one more thing is clear: higher education, for-profit or not, has increasingly become a scam.

We know the consequences of default for lenders, investors, and their backers at the Treasury, but what of the defaulters? Homeowners who found themselves with negative equity (owing more on their houses than the houses were worth) could always walk away. Students aren’t as lucky: graduates can’t ditch their degrees, even if they borrowed more money than their accredited labor power can command on the market. Americans overwhelmed with normal consumer debt (like credit card debt) have the option of bankruptcy, and although it’s an arduous and credit-score-killing process, not having ready access to thousands in pre-approved cash is not always such a bad thing.  But students don’t have that option either. Before 2005, students could use bankruptcy to escape education loans that weren’t provided directly by the federal government, but the facetiously named “Bankruptcy Abuse Prevention and Consumer Protection Act” extended non-dischargeability to all education loans, even credit cards used to pay school bills.

Today, student debt is an exceptionally punishing kind to have. Not only is it inescapable through bankruptcy, but student loans have no expiration date and collectors can garnish wages, social security payments, and even unemployment benefits. When a borrower defaults and the guaranty agency collects from the federal government, the agency gets a cut of whatever it’s able to recover from then on (even though they have already been compensated for the losses), giving agencies a financial incentive to dog former students to the grave.

When the housing bubble collapsed, the results (relatively good for most investors, bad for the government, worse for homeowners) were predictable but not foreordained. With the student-loan bubble, the resolution is much the same, and it’s decided in advance.

In addition to the billions colleges have spent on advertising, sports programs, campus aesthetics, and marketable luxuries, they’ve benefited from a public discourse that depicts higher education as an unmitigated social good. Since the Baby Boomers gave birth, the college degree has seemed a panacea for social ills, a metaphor for a special kind of deserved success. We still tell fairy tales about escapes from the ghetto to the classroom or the short path from graduation to lifelong satisfaction, not to mention America’s collective college success story: The G.I. Bill. But these narratives are not inspiring true-life models, they’re advertising copy, and they come complete with loan forms.

###

This essay appear in Shareable's paperback Share or Die published by New Society, available from Amazon. Share or Die is also available for Kindle, iPad, and other e-readers. 

74

Don't miss a Shareable thing!

Get Shareable in your inbox once a week

.