By Michael Olenick, a research fellow at INSEAD who writes regularly at Olen on Economics
The New York Times ran an op-ed piece on student loan debt, The Student Debt Problem Is Worse Than We Imaginedby Ben Miller of the Center for American Progress. As Yves would say, Quelle Surprise: it’s a disaster. Using an Infographic derived from data in a Freedom Of Information Act (FOIA) request, for over 5,000 schools, the article shows student loan defaults will soar to about 15.5% after five years.
While the FOIA is a dramatic way to get the results, there’s an easier method: looking at investor reports. Like most asset backed securities, reporting data on student loan performance is readily available. Plus, as a public company, the one of the biggest student loan servicer, Navient, is required to disclose much of the information.
Let’s look at Student Loan Servicer Navient’s first quarter 2018 71-page investor deck, here. Page 10 clarifies that as of Sept. 30, 2017 there are $1.5 trillion in student loans. 22% are for $40,000 or more, 21% are $20K-40K, and 21% are $10K-20K, and 36% are under $10K. On page 20, Navient explains that average student loan payments have increased by $64.79, from $263.19 to $327.79 from 1999-2000 graduates to 2018 graduates. If true, that great; the 24.6% increase is well under inflation which itself is far below tuition increases over the same timeframe. If their repayment information is comparing the same loan terms, then keeping payments under control – especially in the education sector where tuition far outpaces inflation – is a remarkable feat.
Moving on, we find that borrowers who compete a degree are about three-times more likely to default than borrowers who do not finish a degree. That makes sense and is a reason that Obama’s Department of Education shuttered scam schools, a policy Betsy DeVos – a longtime proponent of scam schools (like her boss) – is working furiously to reverse.
On page 41 Navient cleanly lays out, in three bullets, why investors love and the general public loathes these loans. Quoting directly:
- Insurance or guarantee of underlying collateral insulates bondholders from most risk of loss of principal
- Typically non-dischargeable in bankruptcy
- Offer significantly higher yields than government agency securities with comparable risk profiles
Translating: this is effectively the same as US debt, since it is guaranteed by the US government and cannot be discharged, but at much higher interest rates. They’re all but bragging that it’s corporate welfare.
On page 53 we get to the part the default triangles, the historical default rates.
For overall undergraduate and graduate rates, the default rate peaked at 25.7% in repayment year 2008. You’d think that is caused by the financial crisis, but the rate was creeping up for years before that, at 24.6% for 2007, 23.7% for 2006, 22.7% for 2005. In 2015, the last year data is available shows a default rate of 10.4%. Either students are better managing their loans or servicers have done a better job of masking defaults through gaming the system.
Page after page of figures shows more interesting information. Students with a co-signer are about twice as likely to default as students without a co-signer. This occasionally has the unfortunate effect of student loan debt collectors calling grieving parents, after the death of their child, to make good on those co-signed student loans. Similarly, for-profit schools have about double the default rate as non-profit schools. People with FICO scores above 740 are about one-fifth as likely to default as those with scores below 670, though there is no information how much the student loans helped cause the low credit scores and resulting high cost of debt.
Besides the annual report there are also asset-backed security investor reports, here. These have more detailed information.
Since the New York Times piece focused on a five-year window I thought I’d take a ten-year one. Here is a linkto SLM Student Loan Trust 2008-9, chosen because it is about ten years old and it’s the first I clicked on. It’s the most recent investor report, with a distribution date (the day funds are paid to investors) of July 25, 2018. Navient is the servicer and Deutsche Bank the trustee. The figures reported are the first distribution date, Aug. 28, 2008 and the most recent reporting date, June 30, 2018.
The trust started with $4.056 billion of principal balance and now has $1.227 billion, so about three-fourths of the funds were either repaid, consolidated into other loans, or paid by the US Treasury as guarantor. Specifically, in Q2 2018 Navient collected $21.8 million from borrowers, $13.4 million from the guarantor, and $23.3 million from consolidations, and a smidgen more from other irrelevant income. Summarizing, after ten years less than 40 percent of the principal Navient collected for a big bundle of student loans came from students. Ouch.
Navient’s report shows 61.4% of the loans (by percent of principal) are current meaning that just under 40% are something other than current. About 14% are 31 or more days late, 15.7% are in forbearance, 8.1% are in deferment, and the rest in various other forms of student loan limbo. I’m not a loss expert but a bundle of ten-year-old loans only has 61.2% performing seems like a mess. 89.2% of the loans are for four-year schools, 9.4% for two-year schools and 1.4% for technical schools; for-profit status is not disclosed.
I almost feel like writing about student loans is taking a cheap shot. There is no other field – not even subprime debt at its worst – where we could find a rating like this. For Navient Student Loan Trust 2017-2, S&P projects a default rate of 42.5% – 47.5% on the $921.4 million of loans. They’ve rated that AA+, the same as the US debt that underlies it. In a roundabout way I suppose that the rating makes sense since the loans are guaranteed by the US government. But lending almost a billion dollars while projecting up-front that just under half will default – leading to ruined credit, stress, and an inability to purchase houses, cars, or start businesses – doesn’t seem like wise economic or public policy.