The Department of Education finally released its final gainful employment rule this morning. This is a high-profile attempt to address concerns about the quality of career education programs, particularly those offered by private, for-profit institutions. This version is the fourth major version of the text we’ve seen, after a proposed version back in March, as well as a proposed and final version that came out during a prior effort that was finalized in 2011 only to be struck down by a court. And while much of the rule looks the same as what we’ve seen all the way back to the first proposed version back in July 2010, there are still some changes.
Changed: One measure only
This final version rests on only one accountability measure: a ratio of the amount of debt graduates from a postsecondary program took on compared to their earnings. This debt-to-earnings rate, or DTE, has been in every version of the rule seen so far. And just as in those prior versions the DTE will be calculated two ways: an annual measure that compares debt to the overall earnings, and a discretionary measure that compares debt to earnings after subtracting 150 percent of the poverty line for a single person (about $17,000).
The single measure is surprising because the version of the rule released in March also included a second indicator that looked at the percentage of student loan borrowers that defaulted on their loans within three years of leaving school. This was the same as an existing cohort default rate measure institutions must address, only applied at the program level as well. This measure had value in that it included all students who borrowed for a program, whereas the debt-to-earnings rate only looks at graduates. This matters, since we know dropouts account for about 63 percent of student loan defaulters.
But the second measure has been the downfall of the rule in the past. The threshold on a student loan repayment rate included in the 2011 version of the rule was struck down by a judge as being insufficiently justified, leading the whole accountability part of the rule to be vacated. And Administration officials admitted to the New York Times that concerns about this rule’s legal prospects led them to drop the program cohort default rate.
Same: DTE thresholds
What did not change in this rule is the thresholds used on the debt-to-earnings rates. A program will still fail if its annual DTE exceeds 30 percent and its discretionary DTE exceeds 12 percent. These are the same failing levels included in the March rule as well as the 2011 final rule. But this version and the one in March are tougher than the 2011 final version in that it defines passing the DTE measures as being below 8 percent on the annual DTE or 20 percent on the discretionary one. Programs that fall in between passing and failing are in a “zone” status.
Same: Time to ineligibility
The consequences of failing or being in the zone are also the same as they were in the March version of the rule. Failing twice in a three-year period or spending four straight years in the zone will result in eligibility loss. The 2011 version of the rule had required a program to fail three times in a four-year period to lose eligibility and had no zone concept.
New: Interest rates for DTE
One of the biggest changes within the DTE calculation is the interest rate used to calculate annual debt payments. The March version of the rule would have used a six-year average of the Unsubsidized Stafford Loan. This version adjusts the rate based upon the type of program and whether it is undergraduate or graduate. Undergraduate certificates and associate degrees will use a three-year average of the Unsubsidized Stafford rate for undergraduates; bachelor’s degrees will use six years. Similarly, graduate certificates and master’s degrees will use a three-year average at the graduate Unsubsidized Stafford loan rate, while doctoral programs will use six years.
Same: Most other parts of the DTE calculation
Aside from the interest rates, little else in the DTE calculation is different from what we saw in March. Programs can still cap their debt at the lesser of the amount students take out or what the school charges for tuition, fees, books, supplies, and necessary equipment. It will also only look at programs that have at least 30 students who completed the program and received some form of federal student aid over either two or four years. And earnings will still be measured either three and four or six and seven years after graduation (add two extra years for smaller programs where four cohorts of data are used). The time period used to estimate how long students will repay their loans is also the same–10 years for associate and certificate programs; 15 for bachelor’s and master’s degrees and 20 year’s for doctoral and first professional degrees. The debt figures also still include federal loans, private loans the college knows about, and institutional credit extended to students.The debt figures will also drop the highest loan amount remaining for each student for which the Social Security Administration cannot obtain earnings information (e.g. if there are four non-matches, the four highest debts get excluded).
New: Longer transition period
One complaint from institutions about the rule was that they would not have enough time to adjust results and avoid failure. The March draft addressed this issue by including a transition period in which a program could replace the debt part of the DTE with newer figures for recent graduates for up to four years. The final version lengthens that time period substantially, allowing a one-year program to use a transition period for five years; a program between one and two years can use it for six years; and a program longer than two years can have seven years in the transition period.
New: Elimination of mitigating circumstances appeal
The March version of the rule contained a mitigating circumstances appeal in which a college could seek to avoid penalties on its DTE by showing that less than 50 percent of all its program graduates borrowed for their education. This allowed an institution where a relatively small percentage of its graduates received federal aid to avoid being judged. That provision is gone.
Same: Debt attribution
One issue that sometimes comes up is what to do with debt for a student that completes multiple programs at the same institution. As in the prior rule the Department will attribute all undergraduate debt to the highest undergraduate program someone completed. So someone who finishes a certificate and associate degree will have all their debt counted in the latter. A similar breakdown will be done for graduate programs. Someone who completes undergraduate and graduate programs will have their debt split and counted once in the undergraduate side and once in the graduate side.
New: What happens if no DTE calculated
The regulations contain a new provision for what happens if a program has a DTE rate one year but not the next. In this situation it will remain in whatever status it was in the prior year. So a program that fails in year one, has no rates in year two will stay as a failing program. But if a program has no rates calculated for four years then it will have its status reset.
What’s not clear about this is whether a program that fails in year one, has no rates in year two, and passes in year three will be considered to be a failing and ineligible program or not. UPDATE: In reading the full text what this means is that a program that failed in year one and has no rates in year two is treated as if they are still a program with one failure. So that means they still have to give student warnings but don’t lose eligibility.
Same: Disclosures for programs about to lose eligibility
Just as before, the only consequence for programs that fail or are in the zone is that the year before they might lose eligibility (after one failure or three years in the zone) they have to provide a warning disclosure to both enrolled and prospective students. The biggest difference is that e-mailed warnings must get some kind of electronic or written acknowledgement from the student that they got the message.
New: Disclosing determinations about meeting licensure requirements
Just as in the March version, institutions will have to note on their disclosure templates whether the program does or does not satisfy the requirements for licensure or certification in any state within their metropolitan area. But now it will have to also provide two other pieces of information: (1) whether it satisfies the same requirements for any other state in which it has made a determination as to whether it does or does not meet the requirements or (2) a statement indicating which states it has not made determinations about satisfying requirements in. This change will both capture more states where the institution has to note if the program meets requirements and at least warn students in places where the institution didn’t check that it cannot guarantee that it meets the requirements.
New: State authorization matters for certification
The final rule continues a requirement that was added for the March version that the heads of institutions will have to certify by the end of 2015 that all of their gainful employment programs are accredited, have programmatic accreditation if its required by their home state government or a federal entity, and if it meets the prerequisites for certification and/or licensure in the home state. But there is a slight change. The March version required institutions to certify that all programs met those same requirements for any other state within their metropolitan area (e.g. Missouri and Kansas for Kansas City). The requirement drops the metropolitan area requirement but says that the certification must also be made for any state where the institution needs state authorization to operate. It is unclear whether this is a stronger or weaker adjustment.
Same: Data on completers and dropouts
Though the only remaining accountability measure is the debt-to-earnings rate, institutions will still have to report data on all their students getting federal student aid. That’s because the Department is reserving the ability to still calculate default rates, earnings information, repayment rates, completion information, and a host of other data points on students who completed and withdrew. So dropouts are gone from the formal accountability system but still in the disclosure piece.
New: Fewer program levels
While the gainful employment rule has already separated out programs based upon their type and level (e.g. an associate degree in medical assisting is different from a certificate in medical assisting), the March version would have allowed for greater disaggregation. More specifically, it would have reported certificates of different lengths as separate programs. This would have allowed a college that offers a nine-month and 18-month certificate in the same field to have them show up as different programs. The Department moved away from this format in the final rule and is now treating all programs at the same level and type as one program. Colleges will, however, have to publish different disclosure templates based upon programs of different length.
New: Clarified poverty year
A small change, but the Department clarified that the poverty level used for calculating the discretionary debt-to-earnings rate will be for the calendar year following the year from which earnings data are obtained. In English, this means that if they pull data from Year X to calculate earnings, they will use poverty data from Year X+1. This is because the poverty levels for Year X+1 are established using data from Year X.
New: Mean and median earnings data will be reported
The final rule does not change the practice of using the higher of the mean or median earnings information, but the Department will now start indicating which one it used and reporting both figures. This will be a nice improvement to creating directly comparable figures.
[Cross-posted at Ed Central]