MagicDiligence's  Instablog

Send Message
MagicDiligence provides useful, simple, and effective stock screening tools inspired by Joel Greenblatt's Magic Formula® Investing methodology. Our Spells give value and growth investors a list of great stock candidates every day, and our advanced Spell Caster lets you create the Magic-style... More
My company:
My blog:
  • Gainful Employment Rules Ignore Demographic Realities 5 comments
    Aug 24, 2010 12:25 PM | about stocks: APOL, COCO, CECO, GHC, DV, STRA

    One of the biggest risks facing publicly traded for-profit educational institutions is the Department of Education's (DoE) proposed "gainful employment" rule. Under the proposed rule, schools where less than 35% of students are repaying the principal on federal education loans would essentially become ineligible. Schools with 35-45% would face restrictions on their ability to receive these loans. There are some additional income stipulations, where if debt is 8% or less of income, then the above restrictions can be waived.

    Notably, the rule is set to only apply to for-profit institutions, not publicly funded 2 or 4-year schools.

    The rationale for these rules is noble. The government wants to insure their investment in education is resulting in employment where the student is making enough incremental salary to show the value of the education received.

    Private Sector Education Repayment Rates

    Shortly after announcing the proposed rules, the DoE released a report that detailed the repayment rates at over 8,400 institutions, both for-profit and not (download it here). This release caused the share prices of many for-profit companies to tank, as several of them fell well below the 35% threshold. The consolidated repayment rates and weighted average federal debt per student at some of the larger companies were:

    Corinthian Colleges (NASDAQ:COCO) 24% ($6,794)
    Strayer (NASDAQ:STRA) 25% ($14,908)
    Washington Post's Kaplan Division (WPO) 28% ($7,458)
    ITT Education (NYSE:ESI) 32% ($10,608)
    DeVry (NYSE:DV) 35% ($13,373)
    Career Education (NASDAQ:CECO) 36% ($10,775)
    Apollo Group (NASDAQ:APOL) 44% ($13,324)

    Those are unquestionably some poor numbers, leading to an explosion of reactionary journalism and "analysis" that immediately concluded that the bear case on for-profit educators, of being "churn and burn" marketing institutions, must be true. In fact, a New York Times article had a quote from Debbie Frankle Cochrane, program director at the Institute for College Access and Success, that read “I think it’s notable that the for-profits are the only type of school where the majority of students are unable to repay their loans”.

    With several of these companies a part of Magic Formula Investing, the issue is important to followers of the strategy (as well as investors at large). Is it really true that for-profit institutions are wholly unconcerned with educating their students? Is the problem just that the insatiable greed of capitalism is to blame for low repayment rates at private sector schools?

    All Students Are Not The Same

    For the purposes of this article, I'll put aside the objections raised by firms like Strayer that argue that consolidation loans, which are often structured as interest-only or graduated income repayment, are not considered in the repayment rates. That is fine, but if the methodology was the same for all schools, the net effect should be similar and bring down repayment rates for all schools.

    More of interest to me was raised on Corinthian's conference call last Friday. Corinthian raised a very salient point: these companies, particularly the 2-year focused ones like COCO, ITT, DV, and CECO, service a generally low income, minimally educated demographic. In fact, up until this year a quarter of Corinthian's enrollment was "Ability To Benefit" students, a program set up by the government to allow students with no high school diploma or equivalent to attend higher education. It is of little surprise that ATB students graduated at low rates, defaulted on loans frequently, and were hard to place in jobs (Corinthian is dropping them for 2011). Most programs at these 2-year schools are vocational, and targeted at either older students or younger students that do not possess the desire, SAT test scores, or GPA to be accepted into public universities.

    Could it be that the relatively low repayment rates at many of these institutions were due more to the demographic they served as opposed to the supposedly poor quality of education they delivered? I wanted to test the hypothesis.

    Is There A Demographic Pattern?

    One way to test the demographic theory on the DoE's data is to look at some 4-year public schools (the ones not ruined by capitalism). Data about average student incomes at public schools is obviously difficult to come by, but reputation is a good place to start. MagicDiligence operates from Baltimore, and I started there with a few inner-city and rural colleges that, by reputation, service low income demographics:

    Coppin State University 23% ($11,958)
    Morgan State University 23% ($14,765)
    University of Maryland, Eastern Shore 31% ($12,433)

    Hmmm... all 3 schools have similarly low debt repayment rates. We're off to a good start, but this is clearly not enough data. Let's stretch the example out some. Athletic conferences are built around schools with similar demographic populations, so let's look at the rest of the Mid-Eastern Athletic Conference (MEAC):

    Bethune-Cookman 15% ($12,890)
    South Carolina State 17% ($15,583)
    Savannah State 20% ($12,511)
    Delaware State 21% ($15,316)
    North Carolina Central 22% ($20,367)
    Norfolk State 24% ($13,325)
    North Carolina A&T 27% ($12,036)
    Howard 32% ($31,789)
    Florida A&M 32% ($18,804)
    Hampton 42% ($17,377)

    Well, this is interesting! The entire conference of schools have repayment rates below 45%, and only Hampton even clears the DoE's 35% restricted threshold. 7 schools have lower repayment rates than Corinthian, the worst of the for-profits (1 is the same). And several are sticking the federal government with more than double the debt of the 2-year schools (challenging another myth that for-profits are way more expensive than public schools).

    So, the MEAC should be shut down by the DoE and shorted by Steve Einsman. But is this an anomaly? What if we look at another conference of low income demographic public universities? Let's take the Southwest Athletic Conference (SWAC):

    Mississippi Valley State 8% ($19,143)
    Grambling State 12% ($17,111)
    Jackson State University 12% ($20,433)
    Alabama State 14% ($15,102)
    Arkansas Pine Bluff 14% ($13,559)
    Alcorn State 15% ($15,398)
    Southern 18% ($19,621)
    Prairie View A&M 20% ($20,833)
    Alabama A&M 21% ($18,186)
    Texas Southern 23% ($17,931)

    That's even worse than the MEAC! The entire conference, 10 publicly funded schools, with repayment rates well below Corinthian's (drastically below ITT and DeVry), and with more than double the repayment obligations.

    Clearly, Debbie Frankle Cochrane is dead wrong when she asserts that only for-profit schools have students that cannot repay their debt. But do we really believe that all of these schools are "churn and burn" institutions, uninterested in the outcomes for students? I think it is unlikely that is the case.


    While I realize that this is by no means an exhaustive analysis, I think it is more then enough to show that demographics matter to repayment rates - and they matter a lot. The "best and brightest", with well-to-do families that pay a significant portion of their bills, should not be the standard for which a student who went to work out of high school to support their family is.

    Furthermore, the "gainful employment" rule seems to do a very poor job at weeding out the bad apples in for-profit education - the ones that are not truly providing valuable instruction. In fact, for these low income students, the trade-based for-profit schools are a much better value than the 4-year publicly subsidized schools. They are providing jobs that allow higher repayment rates and leave students with much lower debt burdens, according to the DoE's own data!

    A better plan would be to simply do what has been done - limit federal aid based on cohort default rates. This system has been fairly successful, penalizing a school when cohort defaults exceed 25% for three consecutive years (or 40% for one year). Even lowering these rates by a few percentage points would help the bad apples to shape up. If the DoE is really serious about pushing "gainful employment", for which no positive outcome studies exist, a better idea is to include consolidation loans that are current. It is unfair to penalize any school (private or public) for loans that are current, regardless of the repayment schedule.

    I fear that, if passed, all this rule will succeed in is limiting educational opportunities for low income students looking to better their position in life. We need a better way to weed out what schools are providing value vs. those that are not.

    Disclosure: Steve owns APOL, COCO, ESI
Back To MagicDiligence's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (5)
Track new comments
  • Craigie
    , contributor
    Comments (24) | Send Message
    The version of the rule proposed by DoEd last month in fact does count consolidation loans that are paying down principal. What some of the large publicly-traded for-profits have been arguing is that the proposed formula be changed to incorporate an assumption that, once loans are consolidated, the full consolidated balances will be in the numerator, regardless of whether the loans are paying down are not. This is preposterous.


    On the other hand, it might be in the schools' best interests to propose changing the rule so that consolidation loans are not considered in the formula, either way. This would make it much easier for schools to challenge the data down the line. There is no way that a school or a lender can know how a consolidation loan is performing -- unless one of its staff happens to run into someone from the new lender at a cocktail party and they happen to discuss it -- which in and of itself would violate the privacy act. A consolidation loan is a completely new legal contract and promissory note. For DoEd to match up the old loan and the new loan and track behavior on both would chew up massive computing resources.


    There is all sorts of data on DoEd's site and elsewhere indicating that for-profit schools leave students with much higher cumulative debt levels than other types of schools. Of course, you have to compare the same length of time in school. Comparing a certificate program to a bachelors degree is a fallacious comparison. Under your "system," moreover, the schools with the highest drop-out rates would leave students with the least debt burden because they are spending less time and school (and thus less time signing up for new loans).


    Even the Wall Street Journal has recognized the importance of cutting through the confusing academic calendars advertised to students and putting things on an even measuring scale: "ITT's two-year associate degrees can cost as much as $47,000."




    24 Aug 2010, 11:52 PM Reply Like
  • MagicDiligence
    , contributor
    Comments (362) | Send Message
    Author’s reply » The DoE rule itself, as well as commentary from nearly all the private sector educators, indicate that the rule only counts those paying down principal. That excludes consolidation loans which are often structured as interest only or graduated income repayment. If those loans are current, they should not be counted against any school. The DoE is reviewing with Strayer, who was most vocal about this.


    I'm not sure how you conclude that the for-profits leave higher debt burdens. The argument has always been that they get 80-90% of revenues from Title IV loans, and the DoE's own spreadsheet show that the outstanding federal debt at payback time is very much in line with that from public schools, and much lower for the predominantly 2-year programs (as you would expect).


    By saying "as much as $47,000", the WSJ is as guilty as the for-profits outed by the GAO for saying that graduates can make "up to $60k". Neither figure is indicative of the average experience.


    The point of the article was to show that low repayment rates are very highly correlated to income demographics, and probably have little if anything to do with the quality of education being provided.
    25 Aug 2010, 09:48 AM Reply Like
  • Craigie
    , contributor
    Comments (24) | Send Message
    The reg language itself, as well as the reg preamble, the DoEd flowcharts, and the FAQs indicate that, in the final implementation of the rule, schools will get credit for any paydown of consolidation loans. 0% of FFELP consolidations are interest-only/neg am, because it is not allowed -- this would be costly in terms of lender subsidies (lenders aren't going to eat the lost interest), and the IRS won't share taxpayer records with banks or other govt agencies. Graduated repayment is not based on income, and is actually a fairly-ambitious repayment scheme from the borrower's viewpoint. It starts relatively-low but "gradates" every two years. Inherent in the plan is the assumption that most people will receive promotions at work at least every two years, and this the ability to repay will grow during the early years of the career. At least that was one of the underlying tenets of the plan when created 25 years ago. Thus, graduated repayment is in contrast to straight-line amortizations, which range from 10 years to 30 years based on amount borrowed. Income contingent repayment (ICR), available since 1995 and only in direct loan, does permit negative amortization and has a lifetime limit of 10% capitalization over the original amount borrowed. Originally designed to encourage public service among highly-educated, highly indebted graduates, it has been quite unpopular -- perhaps due to the numerous required interactions with the IRS -- and instead has become a route for FFELP lenders and guarantors to offload defaulted private loans into direct loan. IBR is a very new plan and its theoretical use is thus quite speculative. It is not likely to be much more popular than ICR. There is an initial three-year lender subsidy that could, under certain circumstances, simulate an interest-only period. This is essentially an extended economic hardship deferment, so it would not be expected to show up as a paydown of principal in any case. And of course all these repayment plans are available to Stafford loans not just consolidation loans.


    How could you not conclude that for-profits leave students with higher debt burdens? Among 2007-08 public four-year bachelor’s degree recipients, 38% graduated with no education debt, while 6% owed $40,000 or more. Among for-profit bachelor’s degree recipients, 4% had no education debt and 24% owed $40,000 or more. Among bachelor’s degree recipients who borrowed, median debt was about $17,700 at public institutions, $22,380 at private not-for-profit institutions, and $32,650 at for-profit institutions. When, in a given 12-month period, 90 percent of students at 2yr publics are not taking out any federal student loans, while at for-profits, only 12 percent are not taking out any federal student loans, while 77 percent are taking out both a subsidized and unsubsidized Stafford, 10 percent subsidized Stafford only and two percent unsubsidized Stafford only.


    As far as the annual and cumulative debt/cumulative borrowings data from the National Center for Education Statistics, it is a public data set from a peer-reviewed quadrennial study, so you can run the numbers yourself if you want. You can calculate averages, medians, centiles, etc. You will of course get lower average/median cumulative results if you include all students from that year, i.e., students who are a point in their postsecondary education at which they are not graduating that particular year.


    I am not saying anything here about quality, just that, when you re-arrange the intentionally opaque academic calendars of the for-profits onto the same template as a traditional academic year (or credit hour/semester hour), borrowing rates and amounts are closely correlated to the type of institution that the student has chosen.
    25 Aug 2010, 08:18 PM Reply Like
  • MagicDiligence
    , contributor
    Comments (362) | Send Message
    Author’s reply » Straight out of the proposal:


    "A loan would be counted as being repaid if the borrower (1) made loan payments during the most recent fiscal year that reduced the outstanding principal balance, (2) made qualifying payments on the loan under the Public Service Loan Forgiveness Program, as provided in 34 CFR 685.219(c), or (3) paid the loan in full. Other borrowers who are meeting their legal obligations but are not actively repaying their loans, such as those in deferment or forbearance, are not considered to be in repayment."


    Clearly, only those paying down principal are counted as repaying.
    25 Aug 2010, 08:49 PM Reply Like
  • Craigie
    , contributor
    Comments (24) | Send Message
    (1) includes former students who are paying down principal on their outstanding consolidation loans. (3) does not automatically include 100% of consolidations, although this seems to be the interpretation of some of the for-profits who have hit the cable business news networks aggressively. Payment in full means the original debt is paid-in-full. Strayer and others are clearly trying to pull the wool over investors' eyes by ginning up calculations which presume that all consolidation borrowers are competently repaying their loans on schedule. Some are in deferment, some are in forbearance, some are delinquent, some are defaulted, some have became disabled, some are wasting their time filing for bankruptcy, which can take many years (during which the borrower is not paying), and so on.
    25 Aug 2010, 10:37 PM Reply Like
Full index of posts »
Latest Followers


  • MagicDiligence re-launching as a screens and tools site! Useful, simple, effective screens - very low price.
    Jan 25, 2016
  • Why Polaris Is A Great Stock To Buy Right Now $PII $HOG $ACAT
    Nov 12, 2015
  • Funny to see GoPro $GPRO in value-based Magic Formula screens! Stock crashed back to earth, but still grew revs 43% last qtr, has no debt.
    Nov 6, 2015
More »

Latest Comments

Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.