In February 2021, the Veterans Education Project (the Project) released a report on the impact of closing the 90/10 loophole. The report’s authors were Jason Delisle, a visiting fellow at the American Enterprise Institute (AEI), and Cody Christensen, an AEI research assistant.

The central thesis of the report was that the 90/10 rule enacted in 1992 is “an antiquated and blunt ‘input’ test that is concerned entirely with revenues, not what students themselves earn from their education, whether they pay their loans, or whether they graduate. The 90/10 rule is clearly out of step with the movement toward more modern accountability policies in higher education that are based on advanced data systems and student outcomes.”

We believe that the Project’s report mischaracterizes the objective of the 90/10 rule, stating that it measures revenue, not student outcomes. The 90/10 rule was never intended as a measure of student outcomes. An 85 percent cap on revenue at for-profit schools from federal student aid funds (Title IV) was enacted in 1992 to address significant default rates by students attending such institutions.[1] By the early 1990’s, student default rates at for-profit schools were 41 percent compared to 22 percent for all schools. In 1998, the cap was raised to 90 percent, resulting in the so-called 90/10 rule.[2]

The idea behind the Title IV revenue cap was a market viability test: that taxpayers should not prop up low-quality schools that could not survive in the open market.[3] In short, the revenue test was viewed as a proxy for value. Institutions offering valuable education at a competitive price should be able to attract at least 10 percent of their revenue from employers, scholarship providers, or students willing to pay the tuition.[4] Currently, a loophole excludes revenue from military-connected students when determining compliance with the rule.

The Project’s report also argued that “hundreds” of public sector institutions, which are not subject to the 90 percent limit on revenue from Title IV, “report student outcomes as weak, or weaker, than those of the for-profit colleges failing a 90/10 rule that includes GI Bill and Department of Defense benefits.” This blog examines and identifies shortcomings in the methodology and findings of the Project’s report by evaluating the outcome metrics used and not used in its analysis. We focus on the 87 schools that the report concludes would fail a revised 90/10 rule if GI Bill and Defense Department benefits were included in the 90 percent cap on federal revenue, which we believe is the most likely scenario.


The Project’s report has three broad deficiencies.

Report’s methodology has shortcomings. The Project’s report (1) overstates the number of for-profit schools that that would be affected by closing the 90/10 loophole; and (2) doesn’t address the impact on its findings of a substantial proportion of the 87 for-profit schools that lack outcome data.

  • The report overstates student enrollment in for-profit schools, including military-connected students, by using old data (2016-17) even though more recent data showing a decline in for-profit school enrollment was available. In fact, at least nine schools included in the analysis have closed since 2016-17, such as schools owned by the Education Corporation of America, including Brightwood College and Virginia College, which enrolled roughly 20,000 students at the time the company collapsed.
  • Thirteen of the 87 schools in the report had no outcome data and one of the 10 featured schools, Sonoran Desert Institute, was missing data on three of the four outcome metrics.[5] Overall, about one quarter (26 percent) of 348 outcomes across the 87 for-profit schools had no data. Missing outcome data undercuts the report’s contention that outcomes at the 87 schools were “indistinguishable from most public institutions.”

Data used selectively to overstate for-profit schools’ outcomes. This Project’s report selectively uses metrics to suggest that for-profit schools affected by closing the 90/10 loophole have outcomes “on par with or superior to” those of selected public institutions. The report’s comparisons of outcomes across institutional sectors omits metrics that would undercut the report’s conclusions, including for-profit schools’ withdrawal rates, use of deferments and forbearance to manage default rates, and the proportion of students struggling with or making no progress in repaying their student loan debts. The report also excludes the outcomes of private nonprofit colleges, which have widely better outcomes than those in the for-profit sector, and compares for-profits only to public institutions (where 2- and 4-year schools’ outcomes are lumped together). The report substantially overstates the graduation rates of some of the 87 schools affected by closing the 90/10 loophole by excluding students—including many veterans—who enroll part-time or who had previously earned credits at another school. Focusing on completion rates for first-time, full-time students had long been criticized as a weakness in assessing student success, a weakness that the Department of Education addressed several years ago. The College Scorecard now provides graduation rates for all students.

Schools that engaged in problematic behavior featured in report’s analysis. Many of the largest for-profit schools featured in the report that enroll the most veterans and servicemembers have a troubled history with law enforcement for using misleading advertising and recruiting. For example, Ashford University, the University of Phoenix, and American Military University have settled federal or state lawsuits totaling $307 million for allegedly violating consumer protection laws, protections against aggressive and misleading advertising and recruiting, or improperly incentivizing recruiters—further evidence that the 90/10 rule should be strengthened. At least two other featured for-profit schools have been sanctioned by their accreditors—Spartan College of Aeronautics & Technology and Trident University International.


Shortcoming in Estimate of Schools Affected by Closing the 90/10 Loophole

The Project’s analysis identifying 87 for-profit schools that could be affected by closing the 90/10 loophole has a number of shortcomings, including overstating the number of affected schools. Moreover, the report fails to acknowledge the mechanisms schools use to maintain 90/10 compliance, which creates considerable uncertainty about the number of affected schools.

The 87 Schools in the Report Are Highly Dependent on Title IV Revenue, but Military-Connected Students Enroll in Just a Handful of Schools

In an accompanying excel file, the Project’s report identified 87 for-profit schools that could be affected by a revised 90/10 rule if they were unable to take steps to maintain compliance, such as attracting students not dependent on federal student aid.[6] The majority of these 87 schools are highly dependent on revenue from Title IV—53 schools receive 80 percent or more of their revenue from federal student aid. These 87 schools represent about 5 percent of the 1,700 for-profit schools that participate in Title IV, which the authors acknowledge is a small number, given the size of the for-profit sector.

Overall, we estimate that about 44 percent of the students at the 87 schools in the report have their tuition paid by VA and DOD.[7] However, most military-connected students were enrolled in just a handful of schools. Thus, 76 percent of GI Bill students and 99.8 percent of DOD students were attending four of the 87 schools—American Public University, Ashford University, Trident University International, and Sonoran Desert Institute. In fact, 24 of the 87 schools enrolled 10 or fewer GI Bill beneficiaries and 76 enrolled no DOD students at all. Maintaining compliance with a revised 90/10 rule should not pose a significant challenge for such schools.

Shortcomings in the Data Used to Develop the Project’s Estimates

We identified the following shortcomings in the methodology the Project’s report used to identify the 87 schools.

  • Missing data. The College Scorecard suppresses data for schools with small enrollment because of privacy concerns. Should 13 of the 87 for-profit schools with no outcome data have been included in the report’s analysis? One of the 10 featured for-profit schools, Sonoran Desert Institute, was missing data on three of the four outcome metrics. Overall, about one quarter (26 percent) of 348 outcomes across the 87 for-profit schools had no data. Notably, the outcome metric that most frequently had missing data was earnings, an important metric regarding student outcomes.
  • No context provided for several data elements. Although the analysis contains a glossary on data sources, it does not provide important context for the data elements included in the excel file. For example, the graduation rate metric excluded part-time students and students who had earned some credits prior to enrolling, which was not disclosed. For some schools, excluding these data resulted in higher graduation rates.[8]
  • Estimates include schools that have since closed. The Project’s report uses 2016-17 Education Department data on the percentage of revenue derived from Title IV when more recent data were available.[9] Since 2016, at least 9 of the 87 schools have closed but the report includes them in its list of institutions that would fail a revised 90/10 rule.[10]
  • Estimates overstate reliance on Title IV and veteran enrollment. For-profit school enrollment, including veteran enrollment, peaked in about 2012 and has steadily declined since then. As a result, the Project’s report overstates the impact of closing the 90/10 loophole by not using more recent data. For example, the Department of Veterans Affairs’ GI Bill Comparison Tool shows a decline of about 6,000 GI Bill beneficiaries at the University of Phoenix from 2016 to 2019.
  • Estimates treat for-profit conversions inconsistently. Ashford University is included even though it was purchased by the University of Arizona and has a pending application with the Education Department to convert to nonprofit status.[11] Conversely, the estimate excludes Kaplan University campuses, which were purchased by Purdue in 2018 and rebranded as an online, public institution—Purdue University Global.

Schools Carefully Manage 90/10 Ratios to Ensure Compliance

Many for-profit schools operate multiple campuses, each with its own six-digit Office of Postsecondary Education Identification (OPEID) number. Schools closely monitor their compliance with 90/10 and, according to The Institute for College Access and Success, schools combine the OPEID of a campus with a high 90/10 ratio with that of a campus having a low ratio in order to maintain compliance. Furthermore, few schools fail the 90/10 rule because they lose access to Title IV funds only if such revenue exceeds 90 percent for 2-consecutive years. In the Education Department’s most recent annual report on 90/10 compliance for academic year 2017-18, no for-profit school lost access to Title IV funds.

In 2020, a series of articles in The Capitol Forum found that several years-worth of the Education Department’s 90/10 compliance reports did not include Colorado Technical University and American Intercontinental University (AIU), both owned by Perdoceo (formerly Career Education Corporation), including for 2016-17. The Department told the Forum that AIU’s 90/10 ratio was under audit because the school had delayed federal student aid disbursements to avoid exceeding the 90 percent cap on Title IV revenue. Perdoceo purchased one of the 87 schools identified in the Project’s report, Trident University International, and merged it with American InterContinental University (AIU) in 2020. The Capitol Forum’s investigation found that the motivation behind the purchase was AIU 90/10 compliance issues. Purchasing Trident lowered AIU’s 90/10 compliance score, which was close to the 90 percent cap on Title IV revenue; as shown in the Project’s report, Trident enrolls a large number of military-connected students, whose revenue is currently counted with the 10 percent of revenue schools must obtain from students who do not rely on federal student aid to pay tuition.

Outcomes Are Overstated for Schools that the Project’s Report Says Could Fail 90/10 if the Loophole Were Closed 

Using College Scorecard outcome data to make comparisons across institutional sectors is challenging both because of the multitude of metrics and the assumption that the metrics provide more certainty and precision than the data support.[12] We found that the Project’s report selectively uses College Scorecard metrics to suggest that for-profit schools affected by closing the 90/10 loophole have outcomes “on par with or superior to” those of selected public institutions. As a result, the report overstates the outcomes of for-profit of schools that would fail a revised 90/10 rule.

Average Inter-Sector Outcomes Were Mixed

With the exception of median earnings, average outcomes for the four metrics across the 87 schools that the Project’s report concludes could fail a modified 90/10 rule were poor and weaker than for for-profit schools (1,625) that passed a modified 90/10 rule (Table 1).[13] Moreover, the inter-sector results were mixed because the bottom 20 percent of less selective public institutions had worse graduation rates but similar or better outcomes on default (similar), repayment (better), and earnings (similar) than the comparison group of for-profit schools. In short, it is an exaggeration to conclude that the 87 for-profit schools’ performance was superior to the bottom 20 percent of less selective institutions.

WO Blog on AEI Paper Table 1


Project Report’s Graduation Rates Are Misleading

The graduation rates at for-profit schools that could fail a revised 90/10 rule are misleading because they (1) fail to take into account the predominant credentials offered by for-profit and public institutions; and (2) use a graduation rate metric that excludes students—including many veterans—who enroll part-time or who had previously earned credits at another school.

Predominant degree offered. Based on our analysis of predominant degree offered, it is not surprising that the Project’s report found an average 63 percent graduation rate for the 87 for-profit schools that could fail 90/10 if the loophole were closed; we found that the degree level for these schools is skewed toward shorter credentials compared to the degree programs offered by public institutions, which take longer to complete.

As shown in Table 2, it is not uncommon for an institution to offer multiple program levels, e.g., certificates and associate’s degrees or bachelor’s plus post-graduate degrees such as master’s and PhDs. Of the 2,565 for-profit institutions in College Scorecard as of academic year 2015-16, about two-thirds conferred only certificates; moreover, the majority of students who graduated from 2-year for-profit schools also earned certificates even though the institutions offer associate’s degrees.[14] In contrast, far fewer public institutions (330) confer just certificates and only 28 percent of graduates at 2-year publics leave with certificates.

WO Blog on AEI Paper Table 2


The 87 for-profit schools that could fail a 90/10 rule if the loophole were closed overwhelmingly offer degree programs that take less time to complete (see Table 3). Thus, 77 of the 87 schools are either (1) less than 2-year schools that only offer certificates (37) or (2) 2-year schools that offer associate’s degrees/certificates (40). Although the Project’s report does not identify the degree level of the two groups of public institutions with poorer outcomes, Table 2 suggests that the comparisons of institutions are apples-to-oranges.

WO Blog on AEI Paper Table 3


Graduation rate metric. The Project’s report’s analysis of graduation rates at 87 for-profit schools that could fail a revised 90/10 rule excluded part-time students. The Project’s report presents College Scorecard graduation rate data on first-time, full-time students, which can be higher or lower than data that includes all students, regardless of their attendance intensity or prior enrollment at a different institution. Focusing on completion rates for first-time, full-time students had long been criticized as a weakness in assessing student success, a weakness that the Department of Education addressed by reporting graduation rates for all students in College Scorecard beginning in 2018.

According to our analysis, graduation rates were lower for 39 of the 87 schools when all students were included—sometimes substantially (a 30-40 percentage point difference) and sometimes slightly (a few percentage point difference).[15] For example, Ashford University’s graduation rate increases from 7.8 percent to 18 percent when all students are included in the calculation. However, American Public University Systems graduation rate drops from 42 percent to 14 percent because 95 percent of its students were part time and were excluded from the graduation rate that the Project reported. Overall, the average graduation rate for schools that would fail a revised 90/10 rule declined from 63 percent for first-time, part-time students to 54 percent when all enrolled students were included in the calculations.[16] The relatively small decline in average graduation rates across the sample is likely explained by the fact that 63 percent of the 87 for-profit schools that the Project reported could fail a revised 90/10 rule reported having no part-time students at all and by missing graduation rate data for all students.[17] Overall, average part-time enrollment for the 75 schools was only 13 percent. In some instances, however, this school reported data looks questionable. For example, Ashford University reports .01 percent part-time enrollment but all of the almost 5,000 servicemembers using Tuition Assistance are part-time students.

In general, students at for-profit schools are more likely than those enrolled in public institutions to be attending mostly full-time, another complicating factor in making valid inter-sector outcome comparisons. Graduation rates are higher at schools that enroll students mostly full-time in shorter programs compared to schools with a significant percentage of part-time students pursuing degrees that require more coursework. According to the Department of Education’s National Postsecondary Student Aid Study, a substantially larger proportion of independent students—the demographic targeted by for-profit schools—enrolled full-time (70.1 percent) at for-profits compared to the same demographic that enrolled in public and nonprofit institutions (see Table 4). In contrast, independent students attending public 2-year (71.6 percent) and public 4-year (50.8 percent) were substantially more likely to be enrolled part-time than full-time.

WO Blog on AEI Paper Table 4


Project’s Analysis Ignores Transfer and Withdrawal Rates

Graduation rates provide only a partial picture of quality because they do not capture students who transfer or who withdraw without earning a credential. According to an analysis by the National Student Clearinghouse based on a cohort of students who enrolled in 2011, 36.9 percent of students attending a public 2-year institution had transferred by 2017 compared to 19.8 percent of students enrolled in a 2-year for-profit school.[18] These enrollment patterns depress 2-year public institution graduation rates. The 2012 Senate HELP Committee report found that 54 percent of students that enrolled in 2008-09 at the 30 for-profit schools it studied had withdrawn by mid-2010, which it characterized as “churn.”[19]  Because many for-profit schools devote a significant portion of revenue to recruiting, they can readily replace Title IV dependent students with others who also relied on federal student aid to pay their tuition.

Although the Project’s report provides data only on graduation rates for the 87 schools, the College Scorecard discloses both graduation and withdrawal rates for schools that participate in Title IV. The average withdrawal rate 6 years after enrolling was 26 percent for the 49 of the 87 for-profit schools with data that could fail a revised 90/10 rule, ranging from 12 percent (Travel Institute of the Pacific) to 39 percent (Ashford University). Twenty of the 49 for-profit schools had withdrawal rates exceeding 30 percent.[20]

High Rate of Borrowing at For-Profit Sector Schools Provides Important Context for Understanding Default and Repayment Rates

The Project’s report fails to acknowledge the higher proportion of independent students, including veterans, who incur federal student loan debt at for-profit schools compared to public 2-year institutions. Such data provide important context for understanding default and repayment rates. Moreover, the default and repayment rates for the 87 for-profit schools that could fail 90/10 if the loophole were closed paints an incomplete picture of former students’ ability to repay federal student loans.

Student Loan Debt. An important factor in understanding default rates is the fact that greater proportions of independent students attending for-profit schools borrow and incur higher average student loan debt than students attending public 2-year institutions, which have substantially lower tuition rates.[21] In short, students at for-profit schools are more likely to take out loans to begin with because of the sector’s high tuition.[22] For example, 62 percent of independent for-profit students took on loans in 2015-16 averaging $8,000 compared to 14 percent of community college students whose average loan amount was about $5,300.[23]

Robert Shireman, a former Deputy Undersecretary of Education and now a Senior Fellow at The Century Foundation has suggested a test for the argument that the excessive use of financial aid is a reflection of the low-income students they serve. In a January 2020 paper, he proposed two tests to differentiate schools that are preying on the poor from those dedicated to serving them. For-profit schools fail both of these tests. First, does the school have an oversight board without a financial conflict of interest that vouches for the mission and has final say over the school’s pricing and spending decisions? Second, does the school invest significantly more to educate its students than the tuition it collects. Our 2019 report on instructional spending found that many for-profit schools that received more than $100,000 in revenue from GI Bill students in 2017 spent less than 20 percent of their revenue on instruction, including schools featured in the Project’s report—University of Phoenix (15.3 percent), Trident University International (19.5 percent), The Refrigeration School (12.6 percent), and Ocean Corporation (19.1 percent).

Default rates. According to the Project’s report, the average 3-year cohort default rate for the 71 of the 87 for-profit schools with data was 16.4 percent.[24] However, 32 of these 71 schools had higher default rates ranging from 17.8 percent (about the average default rate for the bottom 20 percent of public institutions) to 48.3 percent, and 12 of these 32 schools had default rates higher than 25 percent.[25]

However, 3-year rates may not be an accurate measure of the proportion of students who default on their federal loans because of steps some schools take to manage their default rates. Currently, the Department of Education monitors only 3-year default rates, imposing sanctions on schools with high default rates.[26] To manage their default rates, some schools encourage borrowers to use repayment options known as deferments or forbearance for the first 3 years after entering repayment, allowing them to stop their payments without becoming delinquent or defaulting.”[27] The College Scorecard shows that in 2018-19, 40 percent of former American Military University students were in forbearance and deferment 2 years after entering repayment.

Five-year cohort default rates, which carry no sanctions, paint a less sanguine picture of the proportion of students who default. In a 2018 report, we found that the 5-year default rate for all but a few of the 22 for-profit schools examined was notably higher. The Department of Education does not publish 5-year cohort default rates, which were obtained through a Freedom of Information Act request filed by the Center for American Progress.

The College Scorecard recently began reporting default rates separately for students who graduated vs. those who withdrew. The data is revealing and shows that the default rate for graduates of American Military University 2 years after entering repayment was 2 percent but 17 percent for those who left without a credential. Similarly, default rates for Ashford University were 4 percent for graduates but 9 percent for those who left without a credential.

Repayment rates. High graduation rates can be misleading if graduates’ earnings after completion are not sufficient to pay down any student loan debt.[28] The College Scorecard repayment metric used by the Project’s report shows only if those entering repayment had paid down at least $1 of their federal student loans, not the proportion of former students who are struggling to repay their federal student loans. For example, American Military University System’s 3-year repayment rate is 23 percent, but 21 percent of former students have made no progress, that is, paid down at least $1 of their loan debt 2 years after entering repayment. Similarly, Ashford University’s 3-year repayment rate was 24.6 percent, but 28 percent of former students had made no progress, that is, paid down at least $1 of their loan debt 2 years after entering repayment. Difficulty in making progress in paying down student loan debt may be related to the high tuition charged, the proportion of students who borrow, the amount of the student loan debt, and the debt burden on those who left without a credential.

Median Earnings Metric Has Limitations but Outcomes at For-Profit Schools Were Poor

Although earnings data are a key College Scorecard metric, a 2016 Brookings report pointed out several important limitations in the data. The College Scorecard metric (1) only includes students using Title IV, and therefore excludes many students attending public institutions where the share of students dependent on federal student aid is lower than at for-profit schools;[29] and (2) doesn’t factor in the contribution of students’ backgrounds to future salaries, suggesting that earnings are solely a result of value added by the institution. The latter is particularly relevant to veterans who acquire meaningful skills while serving in the military.

According to the Project’s report, former students from the bottom 20 percent of less selective public institutions who were working 6 years after enrolling had median earnings that were slightly higher ($23,400) than those of students who had attended (1) the 87 for-profit schools ($23,296) that could fail a revised 90/10 rule (37 had no earnings data), and (2) the 1,625 for-profit schools ($22,045) that would pass a revised rule. For 50 of the 87 schools with earnings data, only 9 had median earnings 6 years after enrolling that were higher than those of a high school graduate. According to the College Scorecard, high school graduates’ median earnings are about $28,000.

Schools That Could Be Affected by Closing the 90/10 Loophole Were the Subject of Lawsuits, Settlements, or Accreditor Sanctions

The Project’s report fails to acknowledge that several of the for-profit schools that would fail a revised 90/10 rule had active lawsuits or had reached settlements with federal or state law enforcement agencies, including American Military University, Ashford University, and the University of Phoenix.[30] The actions of these schools and their targeting of veterans and servicemembers highlight the need to close the 90/10 loophole, which has created a perverse incentive for for-profit schools to engage in such behavior. In addition, Spartan College of Aeronautics & Technology and Trident University International had been sanctioned by their accreditors. 

For-Profit Schools with Lawsuits or Settlements

American Public University

In August 2018, the Massachusetts Attorney General announced a $270,000 settlement with American Public University System, which owns American Military University (AMU), over allegations that the school violated the state’s for-profit and occupational school regulations aimed at protecting Massachusetts students from the deceptive and unfair practices of for-profit schools.

The investigation into AMU’s recruitment practices revealed that the school violated state regulations by failing to make important employment disclosures to prospective students and failing to provide certain other disclosures in advance of enrollment. The Attorney General’s regulations require for-profit schools offering Massachusetts programs to provide certain disclosures to prospective students 72 hours in advance of enrollment, including the cost of a program, the program’s graduation rate, and the percentage of students who are not paying their loans. Many for-profit schools are also required to list the percentage of students who obtained full-time permanent employment within their field of study.

The Attorney General’s Office also alleged that AMU engaged in high pressure enrollment tactics, including excessive telephone calling, which is a well-documented tactic that is specifically barred by state law.

Under the settlement terms, in addition to the paying $270,000 in the form of relief to eligible students, AMU agreed to change its disclosures to prospective students.

Ashford University

Ashford University faces an ongoing lawsuit by the California Attorney General. In addition, the Iowa Attorney General and the U.S. Consumer Financial Protection Bureau (CFPB) also reached settlements with Ashford totaling $39 million concerning the school’s alleged use of misleading recruiting practices.

California Attorney General. In November 2017, the California Attorney General filed a lawsuit against Ashford alleging that its admissions counselors “preyed” on veterans, making “a wide variety of false and misleading statements to prospective students to meet their enrollment growth targets, including how much financial aid students would get, how many prior academic credits would transfer into the school, and the school’s ability to prepare students for careers in fields like social work, nursing, medical billing, and teaching.” The lawsuit cited Ashford’s own internal audit findings that tens of thousands of misleading and deceptive statements each year and no system in place to stop it.

Iowa Attorney General. In 2014, the Iowa AG alleged that Ashford used “unconscionable sales practices through which Ashford telemarketers, under significant pressure to enroll students,” engaged in Iowa Consumer Fraud Act violations, including:

  • Making false or misleading statements to prospective students in order to convince them to enroll.
  • Utilizing unfair and high-pressure sales tactics, including emotionally-charged appeals to persuade prospective students to make uninformed decisions to enroll.
  • Failing to disclose material facts to prospective students.
  • Misrepresenting to prospective students who wished to become teachers that an online Ashford education degree would allow them to become classroom teachers when, in fact, many Ashford graduates are subject to additional requirements that may take more time, coursework, or money.
  • Charging a significant upfront, non-refundable “technology fee,” which Ashford did not refund even when a student dropped out shortly after making the payment. Ashford formerly assessed the fee in a student’s seventh week of attendance which, at times, ranged from $900 to $1,290. Ashford now charges a $50 technology fee, per course, for its online programs.

The settlement required Ashford University to pay $7.25 million to the state. Approximately $7 million of the payment was intended to help reimburse former and current Ashford online students from Iowa, and approximately $250,000 went toward administering the reimbursement program.

CFPB. In 2016, the CFPB concluded that Bridgepoint, the owner of Ashford University, was deceiving students into taking out private student loans that cost more than advertised. (Private loans count toward the 10 percent of “non-federal” funds, helping schools maintain compliance with the 90/10 rule.)

Starting in 2009, Bridgepoint offered private student loans to its students to help cover the cost of tuition. The Bureau found that the school deceived its students about the total cost of the loans by telling them the wrong monthly repayment amount. As a result, students at Bridgepoint were deceived into taking out loans without knowing the true cost and were obligated to make payments greater than what they were promised. Specifically, the CFPB found that Bridgepoint told students that borrowers normally paid off loans made by the school with monthly payments of as little as $25, an unrealistic amount.

CFPB ordered Bridgepoint to discharge all outstanding private loans the institution made to its students and to refund loan payments already made by borrowers. Loan forgiveness and refunds totaled $23.5 million in automatic consumer relief. Bridgepoint was also required to pay an $8 million civil penalty to the Bureau.

University of Phoenix

From 2004 through 2019, the University of Phoenix paid a total of $268 million to settle allegations of recruiting violations and misleading advertising to recruit students.

U.S. Department of Education. In 2004, the University of Phoenix agreed to pay the Department $9.8 million to settle allegations that it had intentionally paid recruiters based on the number of students they enrolled, a violation of the ban on “incentive compensation.” The rationale behind this ban is that schools cannot pay employees based on how many students are recruited or receive federal student aid because it “incentivizes” recruiters to be less than honest in order to increase enrollments and the recruiter’s pay. Incentive compensation harms both the students who incur student loan debt that they may not be able to repay and taxpayers who funded the student loans.

U.S. Department of Justice. In 2009, the Justice Department reached a $67.5 million settlement with the University of Phoenix over whistleblower allegations that the school had continued incentive compensation payments to recruiters after its 2004 settlement with the Department of Education in 2004.

Federal Trade Commission. In 2019, the University of Phoenix reached a $191 million settlement with the Federal Trade Commission alleging that the school had used deceptive advertisements that falsely touted its relationships and job opportunities with companies such as AT&T, Yahoo!, Microsoft, Twitter, and The American Red Cross. The University of Phoenix agreed to pay $50 million in cash as well as cancel $141 million in debts owed to the school by students who were harmed by the deceptive ads. The Commission’s press release stated that

“This is the largest settlement the Commission has obtained in a case against a for-profit school.”

The Commission’s lawsuit alleged that Phoenix relied heavily on advertising to attract students, including specific ads that specifically targeted military and Hispanic consumers. The University’s ads gave the false impression that it worked with those companies to create job opportunities for its students and tailor its curriculum for such jobs. In reality, these companies did not partner with the University. Instead, the University selected these companies for their advertisements as part of a marketing strategy to drive prospective student interest. Moreover, deceptive advertising and marketing materials by the University of Phoenix targeted active duty servicemembers, veterans, and military spouses. The University of Phoenix has been the largest recipient of Post-9/11 GI Bill benefits since the program’s inception.

For-Profit Schools with Accreditor Sanctions

Spartan College of Aeronautics & Technology

In September 2016, the Council on Occupational Accreditation (COE) conducted an unannounced focused review and determined that Spartan was not in compliance with COE accreditation standards. The school responded to COE’s findings on November 28, 2016. Based on an unsatisfactory school response, COE notified Spartan in January 2017 that it was being placed on probation for failure to adequately correct findings of noncompliance identified during the site visit. Among the 24 areas of noncompliance cited were (1) failure to give the accreditation team access to student complaints; (2) the need to replace very old equipment to ensure relevant practices were being taught; (3) failure to provide copies of all lawsuits filed against the school; (4) inability to document that instructors were qualified to teach, such as having appropriate certification or licensure in their area of expertise (see text box). COE lifted Spartan’s probation in November 2017.

Excerpt from COE Report

“On the date of the visit, the classroom sizes were 10-18 students. There were not enough equipment models for practice. Students reported they had to wait their turn. One student reported that much of the equipment was missing as well as many of the tools were broken. The equipment was old. One example is an aviation engine repair pump from 1945 (See picture). The majority of the manuals for teaching was [sic] outdated. There needed to be more teaching manuals that followed the current FAA regulations.”

  • “When the team considered the faculty credentials, there were 23 instructors listed as teaching the Airframe and Powerplant programs; however, 22 of the 23 had no documentation of an Associate degree in the field in which they were teaching.”

  • “The team was not allowed to review the complaint file. When pressed, the Director of Academics gestured that her lips were sealed. The team leader asked if there was a reason the President would not let us see them. She just smiled apologetically and gestured that her lips were sealed.”

Trident University International

Trident University has been regionally accredited by WSCUC—Western Association of Schools and Colleges, Senior Colleges and University Commission—since 2005.[31] WSCUC’s accreditation actions and any correspondence with schools are posted on its website.

Trident was placed on probation in 2011 when WSCUC received information that the school had awarded bachelor’s degrees to students who had not fulfilled the University’s General Education requirements.[32] Moreover, Trident had failed to notify WSCUC when this issue was brought to the school’s attention by an organization that works with the U.S. Navy to assure that the colleges attended by Naval personnel are appropriately accredited and military-friendly. This breach of standards, which WSCUC called “serious,” resulted in the imposition of an “Order to Show Cause” as to why Trident’s accreditation should not be terminated as of March 30, 2012.[33]

In a February 2013 letter to Trident, WSCUC lifted the probation, informing the school that the next WSCUC offsite visit would occur in Fall 2018, to be followed by an onsite visit a year later. However, WSCUC asked Trident to submit an interim report by March 1, 2015, on the progress of, among other things, (1) implementing multiyear faculty contracts, (2) improving retention and graduation rates, and (3) establishing enhanced student services and assessing their effectiveness. This interim report was not available on the WSCUC website.


The Project’s report argues against closing the 90/10 loophole because the requirement is based on inputs (revenue) rather than outcomes (graduation, default, and repayment rates, and earnings). The report fails, however, to acknowledge the utility of input measures (including the market viability test of the 90/10 rule) and cherry-picks outcome metrics to demonstrate that for-profit schools’ outcomes are on a par with or superior to non-selective public institutions. Moreover, it assumes more certainty and precision in the outcome metrics than is warranted given gaps in the reporting.

Through the College Scorecard, the Department of Education has made it possible to compare institutions using a myriad of outcome metrics. Some metrics are more useful than others and the Project’s focus on just four metrics without providing important context is misleading. For example, the graduation rate of first-time, full-time students tells only part of the story. A more complete picture could be obtained by including the completion rate of part-time students who had prior college credits. Nor can completion be meaningfully reported without examining the proportion of students who withdraw without a credential. Similarly, the other College Scorecard metrics touted by the Project’s report painted an incomplete picture of the student outcomes at both for-profit schools and public institutions.

As demonstrated in this critique of the Project’s report, input measures can be an important “value” metric that help policymakers evaluate the return-on-investment at for-profit schools that participate in Title IV. The loophole in the 90/10 rule has long incentivized the predatory behavior of for-profit schools, focusing a spotlight on their owners’ decisions to assign a higher priority to expenditures on marketing, profits, and executive salaries, while giving short shrift to investment on instruction. Other important input metrics include the presence of lawsuits and settlements, which highlight the for-profit sector’s dependence on misleading advertising and recruiting, and accreditation findings, which raise questions about the quality-of-education provided by some institutions.

The Project’s report did perform one public service: highlighting that the Education Department does not regularly report on for-profit schools’ dependence on all federal revenue, including tuition and fee payments on behalf of military-connected students. At the behest of Congress, the Department last documented that dependence in 2016 using academic year 2013-14 data. Given the bipartisan support for closing the loophole, the Department should report annually on the impact of closing the loophole. And, Congress should act quickly to close the 90/10 loophole to (1) protect veterans and servicemembers from the recruiting practices that some schools are incentivized to employ, and (2) ensure that the 90/10 rule’s market viability test works as intended.


[1]The cap was included in the Higher Education Amendment of 1992, P.L. 102-325, §481(b). From 1992 through 2000, 1,204 for-profit schools closed as a result of the 90/10 rule. A 2011 memo by the Congressional Research Service (CRS) provides context for the enactment of the 90/10 rule (see pg. 7 of this link).

[2]P.L. 105-244.

[3]In Cleland v. National College of Business, the U.S. Supreme Court upheld a precursor to the 90/10 rule, a requirement enacted in 1952 to address low quality for-profit schools that preyed on veterans’ educational benefits. Congress required that at least 15 percent of students in a degree program not be reliant on GI Bill benefits. The Supreme Court said that the 85/15 rule was a valid exercise of Congressional power “based upon the rational assumption that if ‘the free market mechanism [were allowed] to operate,’ it would ‘weed out those institutions [which] could survive only by the heavy influx of Federal payments.’” 435 U.S. 213, 220 (1978).

[4]State student aid and private student loans also count on the 15 percent side of the 85/15 equation.

[5]The College Scorecard suppresses outcome data for schools with small enrollment because of privacy concerns.

[6]The report estimates that this number would increase to 333 schools if the ratio were changed to 85/15.

[7]For this calculation, we converted the 55,192 part-time DOD students into full-time equivalents (18,378) in order to make an apples-to-apples comparison with total institutional enrollment, which is also presented as full-time equivalents. Our estimate of 44 percent military-connected students may be somewhat overstated because we have no way of determining the proportion of GI Bill students who may also be part-time.

[8]The glossary’s link for “graduation rate data” is to a generic College Scorecard “download the data” page, not to a description of the actual metric used.

[9]The Project’s report inaccurately states the authors used 2016-17 Title IV data because more current GI Bill data was unavailable. However, the Department of Veterans Affairs has been reporting fiscal year 2018 GI Bill tuition payments to participating schools since mid-2019.

[10]Schools that are now closed include Brightwood College campuses, Miller-Motte, Harris School of Business, and University of the Rockies.

[11]Ashford has a pending lawsuit filed by the California Attorney General alleging the use of misleading advertising and recruiting, preying on both veterans and individuals of limited means.

[12]A 2015 Brookings Institution report examined the validity and reliability of College Scorecard earning data.

[13]Although this paper focused on the comparison of outcomes at the 87 schools that could fail a revised 90/10 rule to the bottom 20 percent of less selective public institutions, the Project’s report also examined outcomes at “median less selective” public institutions. Median earnings ($27,100) and repayment rates (45.4 percent) exceeded those of the for-profits that passed or failed a revised 90/10 rule. Default rates were comparable (12 percent). Only graduation rates (36.4 percent) were lower than those at the for-profits that passed or failed a revised rule.

[14]This analysis is from our 2018 report on outcomes at 26 schools whose owners were members of the Career Education Colleges and Universities, the association that represents for-profit schools.

[15]Our analysis found that 14 of the 87 schools had no graduation rate data.

[16]Overall, 49 of the 87 for-profit institutions that the Project reports could fail a revised 90/10 rule had 6-year graduation rate data that included all students. Data for the 38 schools with missing data was not reported on College Scorecard because of privacy concerns considering the small number of students in the cohort.

[17]Only 75 of the 87 schools reported part-time enrollment. Our calculation is based on data schools report annually to the Education Department data.

[18]See Figure 7 in the Clearinghouse report.

[19]See table on pg. 4 of the link.

[20]Our withdrawal rate analysis relied on the College Scorecard metric that focuses on all student rather than just the smaller cohort of full-time, first-time students.

[21]Independent students are financially independent from their parents, older, and more likely to have dependents than 18-year olds who enroll after graduating from high school. Veterans are also considered independent students.

[22]According to recent data from the Department of Education, for-profit schools, on average, charge substantially more in tuition and fees compared to public sector institutions. See Table 2 at this link.

[23]See Figure 7 of our 2019 report on veteran student loan debt.

[24]Sixteen of the 87 schools had no cohort default rate.

[25]For context, the average for-profit school default rate (14.7 percent) in fiscal year 2017 was more than double that of nonprofit institutions (6.7 percent) and about one third higher than that of public institutions (9.3 percent).

[26]Schools with default rates of 30 percent or greater for 3 years in a row risk losing access to Title IV funds. Furthermore, any school with a default rate of 40 percent or higher is at risk of immediate loss of eligibility to participate in federal student aid.

[27]Although forbearance and deferment may help borrowers avert default, it comes at a price because their loan balances will continue to grow. During deferment, borrowers are responsible for the accrued interest on unsubsidized loans only. In contrast, borrowers in forbearance are responsible for accrued interest on both subsidized and unsubsidized loans.

[28]Earnings for former student 6 years after enrolling who are employed includes data for both graduates and those who withdrew without a credential.

[29]See Figure 7 at this link and Table 3 at this link.

[30]The Project’s report found that the University of Phoenix would pass a revised 90/10 rule that included military education benefits if the cap remained at 90 percent but would fail if the cap were reduced to 85 percent.

[31]Trident was purchased by Perdoceo (formerly Career Education Corporation) in 2020.

[32]According to the 2012 Senate HELP Committee report, the Show-Cause Order followed a March 2010 warning letter, expressing concern about Trident’s Capacity and Preparatory Review (CPR) report, a key report in WSCUC’s accreditation review process. WSCUC acknowledged that “considerable effort had been undertaken by a large number of people in support of the University’s CPR report.” Even with that effort, however, WSCUC also noted that its review team “found the report difficult to follow and lacking in reflection and supportive evidence beyond assertions.”

[33]The Show-Cause Order was rescinded in March 2012 and replaced with a probation after a Nov. 2011 inspection found that Trident had made significant progress in addressing deficiencies in 10 areas.