One of the very first government-created disasters that I wrote about on this blog was the student loan mess. The blog opened for business on November 6, 2012; and already on November 29, 2012 I had a post titled “Out Of Control Student Loans.” That post noted that, from a beginning in 1966, the federally-guaranteed student loan program had just reached a milestone of $1 trillion of student loan debt outstanding, with no signs of an end to the ongoing increase in outstanding balances. Also, at that time, the so-called “delinquency” rate for student loan borrowers had just suddenly soared from about 8.5% to 11% within less than a year. And, the post pointed out, the 11% delinquency rate was in fact quite deceptive, because very large numbers of borrowers — about half — are in some kind of deferral or forbearance status at any given moment, which leads them not to be counted as “delinquent,” even though they are out of school and not paying anything. That meant that if delinquencies had been measured as a percentage of borrowers expected to be paying, rather than as a percentage of all borrowers, the delinquency rate would have suddenly doubled to more like 22%. The post concluded: “Get ready to lose half or so of the trillion.”
Could things possibly have gotten even worse since then? Of course! After all, this is a pile of free federal money there for the taking. Who is going to pass it up? As long as nobody is really paying much attention, this problem will just continue to explode out of control. And really, what with the “civility” crisis, the Elizabeth Warren Cherokee scam, the Kavanaugh confirmation, the midterm elections, and whatever, who has time to pay attention to a mere trillion dollar problem?
Which is exactly why the problem continues to grow. A few months ago (June) CNN reported on the new aggregate of outstanding student loan balances, and it had just reached the $1.5 trillion milestone. So, 46 years to go from zero to the first trillion, and just five and a half years to add the next half trillion.
You won’t be surprised to learn that if you look for official government information on this subject, you will find data that are aggressively spun to minimize the apparent extent of the problem. Searching for the latest from the government (DOE) on student loan default rates, I find this press release from September 26, with the headline “National Student Loan Cohort Default Rate Falls!” OK, I added the exclamation point. But, can you tell by reading that headline whether overall student loan default rates are actually going up or down? Or what the long-term trend is? Or how much of the $1.5 trillion is ever likely to be repaid? And do you have any idea what they mean by that term “cohort default rate”? It turns out that this press release is only about a very small piece of the puzzle, namely a comparison of the percentage of defaults prior to September 30, 2017 among the group of student loan borrowers who entered repayment status between October 1, 2014 and September 30, 2015 to the percentage of defaults prior to September 30, 2016 among the borrowers who entered repayment status between October 1, 2013 and September 30, 2014. Got that? And, in that comparison of those two “cohorts,” and for this brief two-year period after entering repayment, the defaults for the FY 2015 cohort were 10.8%, while the defaults for the FY 2014 cohort were 11.5%. Defaults went down! But is that particular comparison something that is even significant in the big picture? I certainly can’t tell from anything here, and I’ll bet you can’t either.
For information on the big picture, you’ll just have to look elsewhere. A pretty good place to start would be this study released in January by Judith Scott-Clayton of the Brookings Institution, titled “The looming student loan default crisis is worse than we thought.” Already, it sounds ominous. Some key points:
Brookings decided to take a couple of cohorts (the term “cohort” here referring to that group of borrowers who entered repayment status in a given government FY) and see how defaults develop not just in the first couple of years, but over the long pull. The two cohorts selected are the 1995-96 and 2003-04 groups. For the 1995-96 group they have a full 20 years of data; and for the 2003-04 group, 12 years. It turns out that for both groups defaults continue to rise at a steady pace throughout the repayment period. The earlier 1995-96 group actually had slightly higher default rates in the first couple of years (perhaps you are now getting a little more dubious about DOE’s press release above), but over time the defaults for the later 2003-04 group prove to be much higher. Twelve years out, the default rate for the 95-96 group had reached about 16%; and twenty years out, it had reached 25%. For the 03-04 group, twelve years out the default rate has already reached 25%, and Brookings projects that by the time year 20 arrives, it will be over 40%. Conclusion: “Applying these trends to the 2004 entry cohort suggests that nearly 40 percent of borrowers may default on their student loans by 2023.”
Oh, the default rates for these two cohorts studied by Brookings (95-96 and 03-04) in the two years after entering repayment status were both in the range of 3%. (See Figure 1 of the Brookings study.) Compare that to the 11.5% and 10.8% two-year default rates for the 13-14 and 14-15 cohorts trumpeted by DOE in its September press release above. DOE screams that the default rate is “falling,” based on a slight year-over-year decline in the two-year rate, without mentioning that the two-year rate has multiplied by a factor of almost four over the past twelve years. What does that mean for the expectation of cumulative default rates for the most recent group of borrowers over the next twenty-plus years? You don’t even want to know. If you just project out trends on the same slopes as for the 95-96 and 03-04 cohorts, you would conclude that the large majority of borrowers will ultimately be defaulting. Say good-bye to half or more of the $1.5 trillion.
To its credit, Brookings looks specifically at the situation for black student borrowers, to see how the student loan situation is affecting them. The picture is ugly. “Debt and default among black college students is at crisis levels, and even a bachelor’s degree is no guarantee of security: black BA graduates default at five times the rate of white BA graduates (21 versus 4 percent), and are more likely to default than white dropouts.”
I can’t figure out how the term “predatory lending” does not regularly get applied to this situation. I seem to have a clear recollection of Obama-era prosecutors going aggressively after private housing lenders, mostly banks, said to be “predatory,” who had made loans to enable purchases of homes by people of low income and bad credit, and then sought to foreclose after default. OK, but the people did get to live in the houses, and after default the debt could be discharge by sale of the house, or in bankruptcy. Here, the government lends large amounts to young kids with no financial experience, often for nearly-worthless degrees (or to kids who never finish the degree), and then the debt is not dischargeable in bankruptcy. Low income borrowers have their financial lives ruined and non-dischargeable debt hanging over them for decades. And this hammer falls with great disproportion on black borrowers. How is this not a scandal of predatory conduct by the government? Got me.
So what is the solution? Certainly requiring borrowers to have a credit history is not going to work; it’s in the nature of student loans that nearly all borrowers have no prior credit history. Neither will it work to exclude all blacks from the program on the ground that too many of them are likely to default. So far, the only big idea for reform that has gotten anywhere has been for the government to cut off student all loan privileges from those institutions with persistently ridiculously high default rates. The DOE press release linked above says that they have cut off 12 schools from the program. Great! But there are over 6000 in the program, so this is a drop in the bucket.
There actually is a rather obvious solution that would have real impact, namely, charge back defaults to the school attended by the defaulter. I would say that 100% of defaults should be charged back. But we could start lower, say 50%. This could quickly straighten out this very messed up situation.
But as of now, no one is paying much attention. Next time we look in, it is highly likely that the $1.5 trillion problem will have gone to $2 trillion.