Student Loan Debt and Cohort Default Rates at For-Profit Institutions

Last week, the U.S. Department of Education released updated institutional cohort default rates, a measure that tracks how many of a school’s borrowers default on their loans within two years of graduation. For the second consecutive year, the average rate for all institutions increased, going from 5.2 percent to 6.7 percent. In aggregate, 225,444 borrowers defaulted during this two-year measurement window, an increase of 22,478 from the previous year.

While cohort default rates have a number of flaws, they are one of the few measures of postsecondary education outcomes reported by the federal government. In fact, one could argue they provide a valuable snapshot of student workforce preparation—being unable to find a job post-graduation will make it very difficult to pay back a student loan.

Needless to say, whether or not high cohort default rates can rightly be attributed to institutional quality is a hotly debated topic, especially among schools that tend to do poorly on the measure. Proprietary institutions, for example, had a default rate of 11 percent, well above the 5.9 percent figure for public institutions or the 3.7 percent mark for private institutions. Adding fuel to that fire is a Government Accountability Office report released this week that used older data to show that proprietary school default rate increases from 8.6 percent to 16.7 percent when measured under a three-year window, and further increases to 23.3 percent under a four-year measurement.

Setting aside the third rail of institutional quality for the moment, do other factors appear to matter for cohort default rates? For example, if a school’s graduates have a high average student loan debt, are they more likely to default then their less indebted peers? There’s existing research that indicates that program completion plays a significant role in determining whether a student is likely to default on their loans, but does that still appear to hold true for this year’s batch of data?

Using average loan indebtedness and completion data from the Integrated Postsecondary Education Data System (IPEDS), I tried to take a look at both of these questions. Since it’s a lot of charts, I’m going to split the loan debt and graduation rate results into two separate posts. First up—loan indebtedness.

Intuitively, one can see why average student loan debt might matter for cohort default rates. Large loan balances require high monthly payments, which could be hard to make if a borrower is employed at a job that does not pay a great deal. But it can also be misleading—students attending top-tier private universities may graduate with a lot of debt, the name recognition and purported quality of their degree may decrease their likelihood of default. Moreover, the overall average default in the private, not-for-profit sector is so low that it could be at or around the minimum level it can hit, since unforeseen circumstances could always arise that lead to a default.

But what do the data show? Here for example, is a scatter plot showing the cohort default rate at four-year, for-profit institutions based on their average federal student loan debt:

The average default rate of the 103 institutions with above-average federal student debt levels is 9.91 percent—a bit above the overall sector average. This is because slightly more than half of schools with above-average debt (59) also have above-average default rates. This is somewhat noticeable in the data, as the top right quadrant has slightly more data points than the lower right one. At the same, there is a noticeable outlier—Brooks Institute in Santa Barbara, Calif., has the highest average loan debt in the sector at $14,594, but just three of its 840 borrowers defaulted on their loans (a rate of 0.36 percent).

For below average debt institutions, the story is reversed. Only 41 of those 96 schools have above average default rates, which explains why the below average debt schools’ cohort default rate is lower than the sector average.

All of that said though, the discrepancies certainly don’t appear to be overwhelming.

Here’s the plot of two-year for-profit institutions:

Just as with four-years, there is not apparent correlation between loan debt and default rate. The institutions with above-average loan debt are nearly evenly split between having a below- and above-average default rate. Overall, the default rate for above-average loan debt institutions is 10.79 percent—higher than the sector average, but not by much.

For institutions with below-average loan debt, the data also do not uncover much of interest. Slightly more than half of the institutions with below-average debt also have below-average default rates, but the split is not particularly large.

Here’s the plot for for-profit institutions that are less than two years:

Again, there is no clear correlation that emerges from the data. In fact, slightly more of the above-average debt institutions had cohort default rates lower than the sector average. The same is true of the institutions with below-average loan debt.

Having broken down the data for the three different for-profit sectors, it seems pretty clear that federal student loan debt does not have a significant effect on the default rate in those types of institutions. But it’s important to note the data’s limitations. There is no way to know the debt level of students who defaulted, versus those who do not. Thus, it is possible for a school to have a reasonable cumulative debt average, but also have lots of students with a lot of loans who end up defaulting. Without knowing the distribution of defaults relative to debt, it will be very difficult to definitively determine how much debt does matter for institutional default rates.

A few notes on the data:

• I excluded any institution with a repayment cohort of less than 30 borrowers, since those schools are not subject to the cohort default rate measure.
• The scatterplots exclude schools that had no average federal loan data. Institutions that had a federal loan average of zero are included. In each sector this never amounted to deleting more than about 15 schools.
• I have the data for other sectors, which I can post if anyone wants to see it.

Tagged as: Cohort Default Rates, For-Profit Education, Student Loans