Note: This post contains first impressions of the rule. For more straightforward analysis of what’s new and what’s different, go here.
Career education programs are going to have to care a great deal about the debt and earnings of their graduates if they want to stay in the federal student aid programs. But what that will mean for noncompletion and churn as institutions work to meet those targets is not totally clear. That’s the effect of the gainful employment final rule released this morning by the U.S. Department of Education.
The biggest change in the final gainful rule is that programs will only be subject to a debt-to-earnings measure. This metric looks at how much money graduates from programs are earning compared to their debt payments. A second measure that looked at the rate at which all student borrowers defaulted on their loans has been dropped.
The remaining debt-to-earnings measure is similar in structure to the one contained in March’s proposed rule. Programs will be grouped into three levels of performance: passing, fail, and zone, depending on whether the debt payments of their graduates are below 8 percent of annual earnings or 20 percent of discretionary earnings, above both 12 and 30 percent on those two measures, or somewhere in between. Any program that fails twice in a three-year period or spends four straight years in the zone will lose eligibility for federal aid.
This structure for debt-to-earnings rates is a tougher bar for programs than the one released by the Department in a 2011 attempt to define gainful employment that was ultimately struck down by the courts. In particular, we estimate using March data and definitions of pass, zone, and fail that the zone designation includes 832 programs and 121,682 completers, nearly double those that fail. (The Department says overall about 1,400 programs with 800,000 students are in non-passing programs.) Unfortunately since the Department changed its interest rate assumptions and did not release a new data set we cannot generate completely analogous figures.
Holding firm on the debt-to-earnings thresholds is a welcome surprise given that the last attempt at defining gainful employment saw a similar structure in the proposed rule get watered down to only the 12 and 30 percent levels by the time it was finalized.
The Administration also did not budge on the amount of allowed failures. A program that fails twice in a three year period or spends four straight years in the zone will lose eligibility. That’s faster than the final rule in 2011, which required three failures in four years and had no zone penalty. The one challenge is whether extremely generous transition periods that allow institutions to use more current debt data for between five and seven years will undermine the effectiveness of the thresholds.
What about dropouts?
Dropping the program cohort default rate creates a policy challenge. The debt-to-earnings rates only include students who graduated from a program. That makes sense, you want to look at whether the program paid off for people who did everything expected of them and finished. As such, this rate will do a good job addressing one of the major policy problems in this sector—programs that do not provide a good return for their students.
But there’s a second major policy issue—high rates of non-completion and churn. And here the rule is now silent. The March version addressed this issue by including a student loan default rate for all borrowers, dropouts and graduates. Its absence means programs are now only judged by the results for the completers.
That’s potentially a giant policy loophole. We know that dropouts, especially those with debt, are substantially more likely to default on their student loans, be unemployed and suffer other negative consequences. In fact, dropouts account for more than 60 percent of defaulters. Ignoring these issues could encourage colleges to be judicious about who they allow to graduate and could lead to tactics like giving retroactive scholarships to students who are about to graduate just so they can keep their debt balances down.
It also means that “lottery ticket” programs, where a few graduate with a decent payoff and many more do not, would still pass. Consider the fourth largest program with debt-to-earnings data—an associate degree in office management and supervision from the University of Phoenix. With typical earnings of $38,530, and a debt-to-earnings rate of 4.9 percent, it looks pretty good. But it has an extremely high default rate of 35.7. Why? Because it has a lot more students repaying loans than graduating. The program has 6,200 recorded graduates but more than 27,500 borrowers. It actually has more defaulters (9,835) than graduates. A program that produces 1.5 defaulters for every graduate will pass easily.
To be fair, the cohort default rate is not the perfect measure of student loan performance and has its own potential for gaming. Something like a repayment rate would have been much better. The issue is more the loss of any metric on non-completers, not this one in particular. Also, more programs and students appear to struggle with the performance zone than the program cohort default rate, so keeping the former will affect more programs than if the latter had remained instead.
Based upon the March data and definitions of pass, fail, and zone, the exclusion of program cohort default rates also probably won’t help struggling companies like ITT Technical Institute or Education Management Corporation, which, have 85 percent and 62 percent of their completers in failing or zone programs. It matters only for a potential sale, but Corinthian Colleges would benefit a bit, with 19 percent of its completers in programs that failed the default rate but now pass, though 39 percent are still in failing or zone programs. The change does, however, help Lincoln Educational Services, which had 41 percent of its borrowers in programs that failed due to the default rate, Apollo, which had 30 percent, and Career Education Corporation, which had 27 percent.
More broadly, March data and pass, fail, and zone thresholds suggest the exclusion of the cohort default rate means about 320 programs with 68,000 graduates and 115,000 borrowers that passed the debt-to-earnings but failed that measure can breathe easier. Another 359 programs with 22,000 borrowers that failed only the program cohort default rate will also not have to worry.
The End of the Beginning
On net the requirements programs have to meet for their graduates are unquestionably stronger than anything they’ve ever been held to in the past. The absence of any measures targeting non-completion, are, however a risk. The issue of churn and dropouts will need to be addressed by other efforts like continued audits and oversight by the Department, the Consumer Financial Protection Bureau, and the Securities and Exchange Commission. They’ll have to pay a lot more attention to this completion issue; otherwise it could become an easy loophole.
And of course there are the two major external challenges of a potential lawsuit and the need to implement this regulation beyond the end of the Obama Administration. All of that’s to say that while the regulatory text may be final, the gainful employment rule is not done.
Stay tuned for additional coverage from EdCentral.