Looking for our new site?

Higher Ed Watch

A Blog from New America's Higher Education Initiative

< Back to the Education Policy Program

Gainful Employment Liveblog Day 2--Afternoon session

Published:  September 10, 2013

Due to the length of this morning's liveblog post, the afternoon session will be tracked here. You can read the post from the first day here.The summary of the regulations under consideration is here.

Here are links to liveblogs from Day 1Day 2 Morning, and Day 3.


1:25 p.m. Alternative accountability metrics

This area is open to negotiators, so there is nothing from the Department presented here.

Della Justice, from the Kentucky Attorney General's office, suggests that job placement rate be included as an additioanl metric. She notes it cannot stand alone and other metrics are necessary. She suggests a 70 percent job placement standard. She notes nationally accredited programs already have to meet this standard and many other accreditors require between 65 and 70 percent.

Editorial note: job placement rates are frequently what states attorneys general rely upon in their work overseeing colleges, so it's not surprising to see those suggested. Shorter programs already require a 70 percent job placement rate under the Higher Education Act.

Justice also brings up the idea of including the repayment rate.

Jack Warner from the South Dakota Board of Regents reiterates an argument he made yesterday that the repayment rate should be an initial gatekeeper metric, where those that perform well on it would be exempted from other considerations. Those that fall below certain levels could then be subject to other metrics or scrutiny.

Whitney Barkley from the MS Center for Justice calls for a high repayment rate threshold, then says you pass at least one debt-to-earnings ratio. If a program has a low repayment rate, she calls for three consequences: (1) post a letter of credit, (2) pass both debt-to-earnings metrics, and (3) an independent audit of the accuracy of the placement rate figures. Then if a program has an excessively low figure, she calls for it to lose eligibility.

Kevin Jensen from the College of Western Idaho talks about how his state requires placement rates to be done under a survey and aggregate data. Asks if it is workable. Justice says the methodology would have to be verifiable and reliable, but have to meet a definition for job placement the reflect in field and a period of time. Jensen also asks if meeting a single repayment rate would exempt an entire institution or if it would be at the program level. Barkley says she thinks it has to be at the program level to avoid good programs masking bad ones.

Sandra Kinney, from the Louisiana Community and Technical College System asks for a minimum percentage of students taking out loans, since a large chunk of her students do not borrow. She says compliance costs are high and would get passed on to students that would be unnecessary reporting due to few students taking out loans. She also raises concerns about the validity of survey response rates. She says unemployment insurance data isn't great either because federal employees are not including and private for-profit institutions are not included because the state won't share data with them.

Justice notes that placement rates have to be comparable or they are fraught for misconduct. She also says placement rates would help students make better comparisons between community colleges and for-profit institutions.

Editorial note: the National Center for Education Statistics did try to set up a placement rate definition through a technical review panel, but was unable to come to consensus. But what about defining what is not placement?

Kinney reiterates problems with placement rates for schools near borders or military placement rates, which would make things misleading.

The facilitator asks to move on to a discussion of other ideas before going deeper on the metric calculations.

Brian Jones from Strayer University asks about distinctions among institutions, pointing out that most of Strayer mostly offers degree programs for students who are already working. He asks also what gets done for graduate level programs. Justice says you would just document if you are in a related field of employment. Jones said it's still hard to define what the relevant program in the marketplace is.

Neil Harvison, who is from the American Occupational Therapy Association and represents the Association of Specialized and Professional Accreditors, points out that most accreditors do require placement rates, but tend to not set standards on them because graduated students have no obligation to provide information.

Margaret Reiter, who represents consumer advocacy organizations, suggests that a floor could be established that said the methodology at least had to have certain things in it.

Richard Heath from Anne Arundel CC asks about how a community college that fell into a zone of so-so repayment would provide a line of credit. Barkley says there could be a discussion of low borrowing institutions regardless of tax status to deal with that issue.

Jensen wonders if placement rates are overreaching, since they are trying to define gainful employment, not gainful employment in the field they were trained in. He says someone could not be in their relevant field but still repay their debt. He says the Social Security Administration will suggest if a student is employed and from those metrics could tell if they are gainfully employed. He does think placement rate data are valuablefor disclosures, just not accountability.

2:10 p.m. Program Level Cohort Default Rates?

John Kolotos from the Department says that Jack Buckley from the National Center for Education Statistics will talk more tomorrow about prior work on placement rates. He also says that requiring public institutions to submit a letter of credit is not acceptable.

Kolotos suggests an alternate metric of a program level cohort default rate. Says it would be the same thresholds as used in the statute for institutions, which is three years at 30 percent, one year at 40 percent. A chorus of "ooooohs" greet the idea in the room.

Editorial question: How would this metric work with the plan for failing two out of three years? Would the 30 percent rate not be considered failing? Or would that be a way of passing?

Kolotos says that the Department has not had a chance to figure out the results of this metric yet, but is throwing it out there as a suggestion.

Nassirian calls for waivers of places with low incidence of loan borrowing, but that it is important not to forget about places with high levels of debt. For example, he cites PhD's in counterterrorism that result in people working in the lobby of K street office buildings at $12 an hour.

Nassirian also strongly objects to the program level cohort default rate.

Barkely from MS Center for Justice also objects to the program-level cohort default rate because the repayment rate acts as an important mirror to protect against the gaming of cohort default rates.

Eileen Conner from the New York Legal Assistance Group proposes that dropout rates be taken into account as a way of capturing one of the things that had been picked up in the repayment rate. She says there could be a minimum threshold on it.

Marc Jerome from Monroe College objects to repayment rates, says that it does not reflect necessarily a good program. He cited the repayment rate by institution, which was released during the last negotiation session in 2010 that showed low repayment rates for many colleges that might be considered high quality. He brings up the demographics argument to see how like students at like institutions--e.g. poor/minority students at New York University compared to similar kids. He also notes that the initial repayment rate--which was measured after three to four years--did not match the 10-yer repayment timeframe.

Jerome also brings up the question about whether three years out is the right time to measure this, saying that those investing in their children's higher education would not look at results only three years out.

Kolotos says that the program-level cohort default rate would be a separate measure that would act as a kick out. So results on debt-to-earnings would not be done separately. This means that failure on a program-level cohort default rate would be another way to lose eligibility, but not a way to save a program. He also raises concerns about having too many thresholds.

Editorial comment/question: One argument for removing the repayment rate is its legal riskiness in light of the court's ruling on the threshold. But if the program level cohort default rate is completely separate from the rest of the regime, then how big a risk is the repayment rate back in by itself? Or could the cohort default rate at the program level be defined differently--such as not allowing deferment or forbearance to be used to avoid default.

ED provides more detail on the court's ruling on repayment rates. The court said there had to be more justification for why a given threshold was chosen and show that it was baed upon reasoned decision making.

Margaret Reiter says cohort default rates have a different goal from the purpose gainful employment targets. She asks if ED has considered alternative rationales for the repayment rate threshold? Kolotos says the Department has talked about it, but wants to hear from the community.

2:35 p.m. Completion rates?

ED asks if completion rates should be considered somewhere in this as a metric beyond disclosure?

Jerome says they should be, but the facilitator moves the discussion forward before more detail can be considered.

2:40 p.m. Back to the regulatory text--consequences of failure/zone

ED notes that this skips ahead of determining what final rates are, but this is about what consequences occur after the final rates come out.

If a program is set to lose eligibility in the next year (it failed once or has been in zone for three straight years) it has to give student warning to currently enrolled and prospective students. Currently enrolled have to be told in writing and in person what the options are for the student going forward. For a prospective student, they have to be told when they inquire that it could lose eligibility, a student may not get aid for the entire length of the program and there is a three-day wait period before he/she can enroll.

The other consequence is if a program fails two out of three years or four straight years in the zone, then the program loses eligibility for three straight years. The program would have to apply under the as yet undefined program approval standards. Ineligibility extends to either discontinued programs (either failing or in the zone). It also affects programs that are substantially similar to those that lost eligibility.

One year after being a failing program, the school cannot increase the number of students receiving Title IV aid in the program compared to what it was the year before.

Belle Wheelan representing accreditors from SACS asks if students might sue if they are deprived of aid they should receive. She also asks that there be another year of informational rates added in. Raymond Testa from Empire Education Group also asks for an additional year since the way the rates would be calculated differs from how the past informational rates were calculated.

Jerome asks if high performance on other measures could be used to lessen penalties for schools.

Helga Greenfield, from Spelman College, notes that minority-serving institutions are not really affected by gainful employment, but does support the additional year of informational rates.

Reiter calls for giving schools more time up front to ramp up, but once it does, the schools should be subject to the 8 percent and 20 percent accountability standards.

Testa says that the warnings do not make much sense for his programs, which tend to run less than two academic years, since they would finish before the threats could come do. He says it would be an inaccurate warning that would scare students. He also says you could frighten a student and chase them out the door with nothing.

Rory O'Sullivan says that there are some programs where it would be better to chase some students out the door because the payoff for students is not good.

Justice asks why a school that fails cannot just be required to refund tuition and fees.

ED says it is not sure that it could require that under the act, but it would basically require what the school is willing to do. If the program is not willing to do anything, then the student would have to take that into account when deciding whether to continue.

A lengthy back and forth between Wheelan and Justice occurs about whether the school even should owe the student anything if they do fail. Justice argues they should, Wheelan argues they should not.

ED clarifies that they are not out to close programs. The objective in the zone is that programs have a chance to fix themselves and do better--says you could describe it as probationary eligibility.

Conner from NY Legal Assistance Group pushes about what schools should be on the hook for after failing a metric and before it loses eligibility. She draws a comparison to when a school opens another campus and is found it should have sought approval first. ED responds that this is similar to the financial responsibility scores when a school is struggling there. In general, the Department is hands off in this area until an institution loses eligibility.

Snack break until 3:30 p.m.

3:40 p.m. Consequences, continued

Ted Daywalt, from VetJobs, asks for a return to bigger picture and stopping to focus as much on consequences to schools and more on consequences for students. He talks about a need to think about predatory schools. He tells a story about a student who took six years to finish a credential, applied to 90 MBA programs and got turned down from all of them, eventually got a job as a GS 5 in the federal government because his degree was not recognized, but that he used Title IV dollars to pay for it.

Tom Tarantino from the Iraq and Afghanistan Veterans of America talks about how the theory of the regulations is to save tomorrow's students based upon the results of yesterday's students. But he says it is a staggering cognitive dissonance to talk about renumeration from those yesterday students. He says that since these students are aggrieved based on all the metrics, how that could not be addressed.

ED said it understands a proposal is being developed and says it is willing to look at it But it also notes that a discharge moves burden on to taxpayers..

Daywalt is getting more annoyed, asking how many times people have to get ripped off before someone steps in.

Nassirian says he is OK with extending the period of leniency, encouraging improvement, so long as there is no harm inflicted upon the student in the meantime. He returns to the idea of some kind of pro-forma upfront look to provide some protection and provide some relief to past students.

Nassirian asks if the Federal Trade Commission holder rule could be applied to suggest that students could use being in the zone or failing to the point of ineligibility as grounds for not repaying loans.

4:00 p.m. Data corrections and challenges

Kolotos outlines how schools can try and correct previously submitted data, the process of working with SSA, and further challenges a school can make in terms of calculating the rates. Key points to remember: ED does not get individual earnings, only a mean or median, as well as the number of students that could not be matched, but not who they are. Since ED does not get individual level earnings data, it cannot challenge the income figures, but it can challenge the loan information.

Jerome asks if the newer 2013 data could be identified by school. ED says anonymizing the data was intentional to keep the discussion at a high level. ED clarifies that when actual rates are generated, the schools would know what their rates are and could verify the people are correct. It's just for these draft figures that they cannot see them. Jerome is concerned about lack of visibility into earnings.

Editorial point: the last gainful employment regulations included all students, not just those who received Title IV aid. But focusing on only Title IV recipients means that each student would have had to fill out a FAFSA, which requires knowing a student's social security number. Since the non-Title IV recipients might not have needed their SSN verified, looking only at these students that got federal aid and filled out a FAFSA might improve the matching rate.

ED clarifies that the reason why some associate degree programs at community colleges were in the draft data was because some colleges reported programs they should not have and they were not removed when the figures were being put together.


4:20 p.m. What happens when a non-match occurs?

Conner asks why ED wants to drop the highest debt amount among borrowers for each match that does not occur. ED says it is because it does not know who did not match, so it was the fairest way to address that issue. Nassirian argues that a figure at the mean or median would be better. ED says it will take it back.

Alternative earnings

Programs can choose to use survey earnings instead of social security earnings to recalculate the debt-to-earnings rate and use those results to pass the measures or win that appeal. NCES will lay out its standards tomorrow. Among requirements is attestation from independent accountant and the school that methodology was sound and results can be relied upon.

Warner, from South Dakota, indicates that the survey does not make as much sense since it is unlikely to be better than SSA data. He suggests it would be better to have the school make some demonstration that the SSA data is missing something. But he cannot give an example of what that would look like.

Jerome says Monroe looked at the survey idea before and found a survey would be too expensive to be feasible.

ED explains it dropped using Bureau of Labor Statistics data as an alternative because it has no relationship to the students at those programs and is not interested in considering adding it back in. It also says that the state data do not cover enough people to be of sufficient quality. Said they are not viable and problematic, leaving survey data as the only alternative.

4:40 p.m. Reporting requirements

The blog demons ate some text here, but there were some questions about if schools have been willfully not reporting data for gainful employment programs that they should have. ED said some schools may have been unclear about what they should have reported, but didn't seen intentional exclusion of programs.

4:50 p.m. Wrapping up

There's a discussion of forming subcommittees on various topics, such as upfront eligibility, opportunity to immprove, treatment of grad students, ramping up requirements, placement rates, the repayment rate proposals put forward earlier in the day, etc. The committee will meet tomorrow a.m. to make a decision about which subcommittees to have.

ED says any new materials from subcommittees will need to come in by Sept. 30 to be ready in time for last meeting.

5:00 p.m. Public Comment

Mary Lyn Hammer from Champion College Services calls for a 30 day period of time to qualify for an exclusion since 60 days represent 17 percent of an earnings year. She also notes that some of the issues around gaming a cohort default rate are from servicer actions.

Whether subcommittees and caucuses are open to the public will be up to each committee. But default is to assume they are private unless there is consensus to allow the public.

That's it for today, back tomorrow at 9:15 a.m., following a private caucus to deterimine what subcommittees will be formed starting at 8:30 a.m.

Join the Conversation

Please log in below through Disqus, Twitter or Facebook to participate in the conversation. Your email address, which is required for a Disqus account, will not be publicly displayed. If you sign in with Twitter or Facebook, you have the option of publishing your comments in those streams as well.

Related Programs