Last year, in these pages, I reported that, on January 13, 2017 (just one week before the Obama administration exited DC), the Obama Department of Education released a memorandum confessing that it had “overstated student loan repayment rates at most colleges and trade schools.” The “updated numbers” provided by the Department were analyzed by the Wall Street Journal’s staff, who found that the new data “revealed that the Department previously had inflated the repayment rates for 99.8% of all colleges and trade schools in the country” (emphasis supplied).

As troubling as those revelations were, new data released last fall by the Trump Department of Education paints an even bleaker portrait of the lives of student loan borrowers. Education researcher Erin Dunlop Velez has just published a thoughtful study that analyzes the new student loan data now available. What she finds in the new statistics is still more alarming.

In the course of delivering her findings, she debunks three common myths regarding (1) how long it takes students to repay their loans, (2) how long it takes the average student borrower to default, and (3) the relation between default rates and the amount a student borrows.

The first myth called into question by the longer-term data is the common belief that most students pay back their loans within ten years (ten years is the standard repayment period for federal student loans). Until now, higher education analysts were forced to try to analyze student loan repayment in the absence of the long-term data they needed. Now that the Department of Education has made this data available, it appears that, in fact, the average student loan borrower takes longer than ten years to repay his/her loans. Velez’s research finds that “even among those who borrowed only for their undergraduate education . . .  only half of students had paid off all their federal student loans 20 years after beginning college in 1995–96.” Instead, average borrowers in “this group still owed approximately $10,000 in principle and interest, about half of what was originally borrowed, 20 years after beginning college” (emphasis in original).

The second myth that Velez finds debunked by the new data is the view that “Most defaults occur in the first few years after students leave college.” This assumption appeared credible heretofore due to the fact that recent graduates can find themselves hard-pressed to find well-paying jobs at the outset of their careers, thereby limiting their ability to repay their student loans. More important, “current federal default statistics measure default only in the first three years after students exit college.” However, Velez’s analysis of the new, longer-term data finds that, “among 1995–96 beginning students . . . more than half of students who defaulted within 20 years of beginning college were in repayment for more than three years before they defaulted. The average time between students entering repayment and their first default was 4.9 years” (emphasis in original).

Because “more than half of defaults occur outside the [time] window covered by current federal default statistics, overall default rates are much higher than previously thought.” How much higher? Velez goes on to argue that, “among students who began college in 1995–96, one-quarter defaulted on a federal student loan within 20 years of beginning postsecondary education.” However, “among students who began college in 2003–04 . . . one-quarter defaulted in just 12 years after beginning college. Time will tell what their default rate will be at the 20-year mark.” Given the number of defaults of 2003-04 college entrants in only their 12th year after starting college, it is reasonable to expect their 20-year default rate to be still higher.

The third and final myth pounced on by Velez’s analysis of the new student loan data is the belief that “increasing default rates are caused by increases in student debt.” Given that the rate of student loan defaults has been growing over time, it is understandable that it has been widely assumed that a “major cause of this trend is increasing tuition prices, which have led to increased student loan debt.” That is to say, it appears to make sense to “assume that the more debt a student takes on, the harder it will be to repay.” But the new data belies this assumption: “those who borrow the most are not the most likely to default.” Why?

Although additional borrowing can make it harder to keep up on one’s student loan payments, “students who complete degrees also borrow more than students who drop out, and degree completion is a major factor related to default.” It stands to reason that students who complete a bachelor’s degree borrow more than those completing an associate degree or a certificate. Moreover, students who fail to complete college “borrow less, on average, than those who attain a bachelor’s or associate’s degree.”

But Velez demonstrates that, although those completing bachelor’s degrees “borrow more, they have significantly lower default rates, whereas certificate recipients borrow far less, “but are considerably more likely to default.” From this Velez rightly concludes that we need to look more closely at degree completion and degree type, rather than solely at the amount borrowed, to better explain student loan defaults.  These data suggest that whether a degree is completed, and what type of degree is completed, may be more important factors related to the increasing default rate than the amount students borrow.

Velez promises that the statistics now available are only “the tip of the iceberg of the analyses that can be conducted with this new data source.” Her “hope is that the data can trigger a deeper understanding of loan repayment.”

I share her hope, for the misleading repayment and debt numbers with which we have had to deal in the past have served only to exacerbate the growing student-loan debt crisis. Flying blind is never a sound strategy.