Post on 25-Jul-2020
transcript
No. 18-16375
IN THE UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT
MARTIN CALVILLO MANRIQUEZ, JAMAL CORNELIUS, RTHWAN DOBASHI and JENNIFER CRAIG, on behalf of
themselves and all others similarly situated
Plaintiffs-Appellees,
v.
ELISABETH DEVOS, in her official capacity as Secretary of the United States Department of Education and THE UNITED STATES
DEPARTMENT OF EDUCATION,
Defendants-Appellants.
On Appeal from the United States District Court for the Northern District of California
No. 17-cv-07210 Hon. Sallie Kim
PLAINTIFFS-APPELLEES’ SUPPLEMENTAL BRIEF
Attorneys for Plaintiffs-Appellees
Noah Zinner Megumi Tsutsui HOUSING & ECONOMIC RIGHTS ADVOCATES 1814 Franklin Street, Suite 1040 Oakland, CA 94612 (510) 271-8443
Eileen M. Connor Toby R. Merrill Joshua D. Rovenger LEGAL SERVICES CENTER OF HARVARD LAW SCHOOL 122 Boylston Street Jamaica Plain, MA 02130 (617) 522-0715
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 1 of 27
TABLE OF CONTENTS INTRODUCTION ..................................................................................................... 1
ARGUMENT ............................................................................................................. 3
I. Even assuming that the Department could use “value” as a means to calculate borrower defense relief, the Department failed to meet its goal of measuring the “value” received by any given Corinthian Student. .................................. 3
A. The Average Earnings Rule sets its comparison, and thus the borrower defense remedy, on the basis of an irrelevant denominator. ................. 4
B. The Average Earnings Rule sets its comparison, and thus the borrower defense remedy, on the basis of an unrepresentative numerator. ......... 9
C. The Average Earnings Rule, in its totality, is a conceptually flawed and illogical apples-to-oranges calculation. ............................................... 12
II. A borrower defense remedy, under longstanding department practice, regulation, and Students’ Master Promissory Notes, is tied to the state law cause of action and thus the average earnings rule is impermissible. ........... 14
CONCLUSION ........................................................................................................ 20
CERTIFICATE OF COMPLIANCE
CERTIFICATE OF SERVICE
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 2 of 27
TABLE OF AUTHORITIES
Cases
Am. Wild Horse Pres. Campaign v. Perdue,
873 F.3d 914 (D.C. Cir. 2014) ....................................................................... 20
Brower v. Evans,
257 F.3d 1058 (9th Cir. 2001) ......................................................................... 8
Encino Motorcars, LLC v. Navarro,
136 S.Ct. 2117 (2016) .................................................................................... 20
FCC v. Fox Television Stations, Inc.,
556 U.S. 502 (2009)....................................................................................... 15
Gonzales v. Cal. Dep’t of Corr.,
739 F.3d 1226 (9th Cir. 2014) ....................................................................... 19
Graham v. Bank of Am., N.A.,
226 Cal. App. 4th 594 (2014) ........................................................................ 19
Howe Investments Ltd. v. Quarles & Brady, LLP,
No. 17-55483, 2019 WL 581370 (9th Cir. Feb. 13, 2019) ............................ 19
Jicarilla Apache Nation v. U.S. Dep’t of the Interior,
613 F.3d 1112 (D.C. Cir. 2010) ..................................................................... 15
Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto Ins. Co.,
463 U.S. 29 (1983) ......................................................................................... 11
Regents of U. Cal. v. U.S. Dep’t of Homeland Security,
908 F.3d 476 (9th Cir. 2018) ........................................................................... 4
Statutes
5 U.S.C. § 552a ........................................................................................................ 14
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 3 of 27
Other Authorities
12 Harry D. Miller & Marvin B. Starr, Cal. Real Estate § 40:17 (4th ed. 2017). ... 19
79 Fed. Reg. 16426 (March 25, 2014) ....................................................................... 6
79 Fed. Reg. 64890 (Oct. 31, 2014). ................................................................. 5, 6, 7
Bureau of Labor Statistics, Occupational Employment Statistics, (May 2014) ...... 10
CJ Libassi & Ben Miller, How Gainful Employment Reduces the Government’s Loan Forgiveness Costs, Center for Am. Progress (June 8, 2017) ................. 7
Regulations
34 C.F.R. § 668.405 ................................................................................................. 13
34 C.F.R. § 668.406 ................................................................................................. 13
34 C.F.R. § 685.206 ................................................................................................. 18
34 C.F.R. § 685.222 ................................................................................................. 18
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 4 of 27
1
INTRODUCTION
Named Plaintiffs and members of the student class (“Corinthian Students” or
“Students”) attended for-profit programs operated by Corinthian Colleges. After a
lengthy investigation, the Department of Education (the “Department”) determined
that these programs were so rife with fraud that Students were entitled to a full loan
discharge upon the submission of a simple form (the “Corinthian Rule”). In 2017,
without reference to its prior findings and legal conclusions, the Department
abandoned the Corinthian Rule in favor of one that ties a Student’s borrower
defense remedy to the “value” she received from her Corinthian program (the
“Average Earnings Rule” or “AER”). This Court has requested supplemental
briefing on whether the Department acted arbitrarily and capriciously in
developing this Average Earnings Rule. It did.
First, even assuming that the Department could use a generic measure of
“value” imparted by a failed institution to reduce the amount of the borrower
defense remedy available to a given (or representative) Corinthian Student, the
Average Earnings Rule does not measure whether a Student obtained any benefit
from her Corinthian education. Indeed, the Department settled on an illogical
denominator and comparator for its assessment: the 2014 average earnings of
students from schools that passed a debt-to-earnings standard under the Gainful
Employment “GE” regulation. GE does not claim to measure “value,” and the
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 5 of 27
2
Department has not shown why the earnings of students attending passing GE
programs should set the standard for borrower defense relief or why this data
stands for “the value of the education…that would have met students’ reasonable
expectations of preparation for gainful employment.” (ER 111).
The Department’s selection of the numerator was no better. The Department
created 79 artificial groups that comprised students who attended different
Corinthian programs, at discrete times, in distinct locations; the data about these 79
groups says nothing about the value a Student received from a specific Corinthian
program at a given location and time. The numerator also included earnings from
students not covered by the Corinthian Rule and not included in GE metrics. It
then excluded earnings from Students who have yet to apply for borrower defense,
or have applied but, like Plaintiff Jennifer Craig, were attending school in 2014,
the very year the Department drew data from.
Taken as a whole, the comparison is conceptually flawed. It rests on the
unsupported premise that Students’ “reasonable expectations” were confined to a
choice between Corinthian—which induced them to enroll on the basis of
fraudulent misrepresentations—and the comparator schools. It simply does not
speak to the difference between what Corinthian promised, and what Corinthian
delivered. It also ignores the Department’s own findings that Corinthian falsely
took credit for placing Students in jobs that they obtained wholly apart from their
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 6 of 27
3
Corinthian program. In short, the Department has created an arbitrary rule that
compares apples to oranges.
Second, “value” as a function of earnings is not, in fact, a permissible way
for the Department to determine a borrower defense remedy in this instance. Both
the Students’ Master Promissory Notes, and the Department’s own regulations,
require the Department to use applicable state law to determine the amount of a
claimant’s loan discharge. Here, California law requires restitution. Until January
20, 2017, the Department agreed with this interpretation and used state law to
decide the remedy for each and every borrower defense application it decided,
under the Corinthian Rule and otherwise. The Department has not explained its
changed interpretation.
Ultimately, the Department’s arbitrary and capricious conduct is further
harming approximately 110,000 defrauded Students. This Court should protect
these Students by affirming the District Court’s entry of a preliminary injunction.
ARGUMENT
I. EVEN ASSUMING THAT THE DEPARTMENT COULD USE “VALUE” AS A MEANS TO CALCULATE BORROWER DEFENSE RELIEF, THE DEPARTMENT FAILED TO MEET ITS GOAL OF MEASURING THE “VALUE” RECEIVED BY ANY GIVEN CORINTHIAN STUDENT. The Department’s development of the Average Earnings Rule “must be
upheld, if at all, on the basis articulated by the agency itself.” Regents of U. Cal. v.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 7 of 27
4
U.S. Dep’t of Homeland Security, 908 F.3d 476, 505 (9th Cir. 2018) (internal
quotation and citations omitted). In its memorandum explaining the Average
Earnings Rule, the Department stated: “This methodology was developed to
provide borrowers relief consistent with and appropriate to the harm they incurred
from the misrepresentation by CCI, thereby making them whole. It is rooted in a
determination of the value of the claimant’s CCI education.” (ER 124); see also
(ER 115) (explaining that “harm to certain CCI students may be determined by
assessing the value students received from the education provided by CCI, as
compared to that provided by other schools to students in similar programs”). The
AER fails to achieve this goal.1
A. The Average Earnings Rule sets its comparison, and thus the borrower defense remedy, on the basis of an irrelevant denominator.
The denominator of the AER metric is “2014 earnings of students
completing programs with the same CIP code and credential [as the relevant “CCI
Program Credential/Group” of applicants] that earned passing scores as part of the
1 The record fully details the Department’s development of the Average Earnings Rule. See (ER 106) (Declaration of James Manning); (ER 113) (Declaration of Phillip Juengst); (ER 124) (Borrower Defense Relief Methodology for CCI Claimants Memorandum); (Supp ER 24) (Declaration of Frank Curran); (Supp ER 29) (Crosswalk of CIP Codes); (Supp ER 39 & 63) (findings lists with relief amounts).
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 8 of 27
5
programs in the Department’s GE program[.]” (ER 126). The choice of this
comparator is the product of pretext and faulty reasoning.
The Department claims that GE metrics released in January 2017 showed
that “programs at CCI schools performed relatively well, overall” (ER 110),
causing the Department to “reevaluate the assumption that CCI students … had not
received any value from their education.” (ER 110-11).2 But data were available
for only 106 Corinthian programs (ER 116), at 17 of 37 locations (ER 116),
representing 24 CIP codes (id.). There were no data for any of the approximately
800 Heald programs, nor for 56 of 61 WyoTech programs. (Dist. Ct. Dkt. No. 42-
2 at 62-63) (Corinthian GE results). Yet, the Department concluded that a new
rule was needed for all CCI Programs, which represented exactly twice the number
of CIP codes as for which there was GE data. (ER 130-31) (Crosswalk of 48 CIP
codes).3
2 The Department also suggests that its conduct was warranted by the significant number of pending borrower defense claims. (ER 111). However, this is no rational basis for change because the Department knew the precise number of Students eligible for relief when it crafted the Corinthian Rule and actively worked with and encouraged states to find additional applicants. See Br. of Amicus Curiae Cal., et. al. at 9-12. Further, the Department highlights the limited Student testimony on harm in their attestation forms, (ER 109), but ignores the fact that it designed the forms to eliminate a Student’s need to show that she was “affected by the school’s fraud.” 80 Fed. Reg. 32944, 45 (June 10, 2015).
3 In promulgating GE, the Department projected that the majority (74 percent) of programs would pass the test, and only a small fraction (8 percent) would fail. 79 Fed. Reg. 64890, 64921 (Oct. 31, 2014). Nothing about the small set of Corinthian’s results was noteworthy.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 9 of 27
6
More fundamentally, GE does not measure “value.” A school’s program
passes GE if a cohort of students completing the program in a given year earn
enough, on average, two years later to “manage their debt” under one of two
metrics. 79 Fed. Reg. 64890, 64913 (Oct. 31, 2014); accord (ER 110) (noting that
the goal is to “see if those students have the ability to repay the educational debt
incurred for attending such programs”). One GE metric compares a typical
program graduate’s annual earnings against the typical debt load of program
graduates from that cohort, and the other compares the discretionary earnings of a
typical program graduate against the typical debt load.4 Both are designed to
measure “the relationship between the total student loan debt and earnings,” 79
Fed. Reg. 16426, 16651 (March 25, 2014), and “not . . . actual benefits,” 79 Fed.
Reg. at 64916; see also 79 Fed. Reg. at 64914 (“As a responsible lender, one
important role for the Department is to hold all GE programs to a minimum
standard that ensures students are able to service their debt without undue hardship,
regardless of whether students experience earnings gains upon completion”).5 Put
4 The AER allows Corinthian Students’ earnings to be compared to programs that passed GE solely on the basis of this discretionary income metric. A program that fails the annual earnings metric still passes GE if it yields sufficient discretionary income because it “ha[s] graduates who have higher earnings even though they have large amounts of debt.” 79 Fed. Reg. at 64921.
5 In adopting the GE regulation, the Department expressly rejected proposals that would better measure “value” because GE, it clarified, was solely focused on “discouraging institutions from saddling students with unmanageable amounts of debt.” 79 Fed. Reg. at 64913; see also id. at 64915.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 10 of 27
7
another way, the thresholds are not set to identify “good” programs offering
“value,” but to create an eligibility criteria designed to weed out “bad” programs
that present an untenable financial risk to students and taxpayers. See 79 Fed. Reg.
at 64913.
In any event, even if the Department now believes that a GE passing
program is, by definition, one of high “value,” the AER’s basis of comparison is
arbitrary.6 The AER includes programs in the denominator that passed a debt-to-
earnings ratio, but measures “value” solely on the basis of earnings. This is ironic
because it is debt, not earnings, that drives whether a program passes or fails GE.
See Carmello (CJ) Libassi & Ben Miller, How Gainful Employment Reduces the
Government’s Loan Forgiveness Costs, Center for Am. Progress (June 8, 2017),
available at: http://perma.cc/NN7DHKNG. In fact, programs can pass GE in spite
of the low earnings of completers. By Plaintiffs’ count, there are 443 programs
included in the AER denominator where Students earned less than the 2014
national minimum wage (approximately $15,080). (Dist. Ct. Dkt. No. 42-2 at 65-
192) (GE data chart). The typical graduate of each of the 443 programs had zero
6 The converse is that any program that failed GE provided low or no value. But the AER does not follow this logic. The six programs reviewed that failed GE, (Dist. Ct. Dkt. No. 42-2 at 62-63) (Corinthian GE results), are slotted for between 20 and 50 percent loan cancellation under the AER, (Supp ER 38-61).
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 11 of 27
8
debt. The inclusion of these programs in the denominator skews the results of the
AER, rendering it arbitrary and capricious.
For example, graduates of the Pharmacy Tech program at Austin
Community College had zero debt and average yearly earnings of approximately
$25,000. (Dist. Ct. Dkt. No. 42-2 at 121) (GE data chart). In contrast, graduates of
the Pharmacy Tech program at Everest in Colorado earned a comparable salary,
but owed an average of $13,000 in student debt over a ten-year period. (Dist. Ct.
Dkt. No. 42-2 at 67) (GE data chart). Under the Department’s logic, this Everest
Pharmacy Tech student obtained virtually the same “value” from her program, and
should be forced to pay back the majority of her loans. The Department ignores
that the Everest Student would necessarily spend her earnings on loan payments
and that the debt has significant and longstanding financial consequences. See
Brower v. Evans, 257 F.3d 1058, 1065 (9th Cir. 2001) (explaining that an agency
cannot “entirely fail[] to consider an important aspect of the problem”). The
Department acted irrationally in “failing to consider” debt as part of its “value”
analysis.7
7 The Department notes that it made several decisions that were to the benefit of the borrower. Defs’ Reply Br. 24. However, a rule that utterly fails to achieve the Department’s articulated goal cannot be saved merely because the Department did not make the wrong choice at every juncture.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 12 of 27
9
B. The Average Earnings Rule sets its comparison, and thus the borrower defense remedy, on the basis of an unrepresentative numerator.
The borrower defense remedy that any Corinthian Student receives under the
AER is also driven by which one of 79 “CCI Program/Credential groups” (ER 126)
the Department places her in, which other Corinthian students the Department
placed in that group, and the average income that certain members of the group—
those who applied for borrower defense prior to July 31, 2017, subject to additional
exclusions—earned, according to the SSA, in 2014. The Department offers no
basis to conclude that the numerator is a reasonable estimate of a given Corinthian
student’s earnings, let alone the “value” that she received from Corinthian.
Initially, despite claiming that it grouped Corinthian data “on a program by
program basis,” Defs’ Reply Br. 23, the Department actually cobbled together
Students from different programs, who attended school at various locations, at
distinct times, into 79 “CIP Program/Credential Group[s].” For example, the
Department points to the “Pharmacy Technician (Diploma) group as an example of
a “program” where all students will receive a partial denial. Defs’ Br. at 2. But,
the grouping is unreasonable given the variety of Students crammed into it,
including: Students who attended the Pharmacy Technician Program at Everest in
Miami in July 2010, and Students who attended in Aurora, Colorado, in September
2012, and Students who attended in San Francisco in 2014. (Supp ER 29-38)
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 13 of 27
10
(showing the various and divergent programs that were lumped together into 79
programs). Although the Department claims that these 79 groups are “essentially
the equivalent of a GE Program for comparison purposes,” (ER 124), this is not
true. A GE “program” includes only completers of a specific program at a specific
school in a particular year. The AER is based off of 2014 earnings, and determines
borrower defense remedy for all Corinthian Students, regardless of whether they
completed and when they attended.8
The Department then erred by calculating the earnings for each grouping
based on a dataset that is not representative of those to whom it is applied. The
dataset is drawn from the subset of 61,717 former Corinthian borrowers who
submitted borrower defense claims by July 31, 2017 “regardless of whether a claim
had already been decided by the Department,” (ER 117), and irrespective of whether
the student was asserting a job placement rate claim, (ER 124) (referencing
Guaranteed Jobs and Transferability of Credits claims).9 The Department fails to
8 The Rule also ignores critical distinctions in earnings that are driven solely by a person’s location. For instance, in 2014, the median annual income for a medical assistant in San Francisco was nearly $40,000, whereas it was $24,630 Springfield, MO. Similarly, a massage therapist in the Boston area earned, on average, $43,130 a year, whereas a massage therapist in Merillville, IN earned $21,470. See Bureau of Labor Statistics, Occupational Employment Statistics, (May 2014) available at: https://www.bls.gov/oes/tables.htm.
9 It is telling that the Department misled a co-equal federal agency to engage in this data experiment and that it did so in a way that betrayed information to the SSA—that these were individuals who applied for borrower defense—beyond what would have been included in a proper GE disclosure, and in a manner that was blatantly inconsistent with the authority provided in that agreement, (ER 153).
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 14 of 27
11
explain how the average earnings of Students who applied for borrower defense
before this arbitrary cut-off reflect the earnings of the tens of thousands of Students
who have not yet even applied for borrower defense or the thousands of Students
whose data was discounted from the disclosure.
In addition to excluding Students who had not yet applied by July 31, 2017,
the data subset did not include students “for which the department lacked sufficient
data to map to CIP code and credential level . . . students who started CCI programs
in 2014 or after,” and “claimants in groups with fewer than 10 individuals.” (ER
118).10 The AER’s numerator expressly excludes—and cannot be representative
of—the “value,” lack of “value,” or harm suffered by Students like Plaintiff Jennifer
Craig, who were attending Corinthian in 2014 and 2015. The Department excluded
her and others because it “assumed CCI programs would have been unlikely to have
influenced 2014 earnings outcomes for such students.” (ER 118). There is nothing
to support this assumption, and reason to question whether this exclusion inured to
the benefit of Corinthian Students if, for example, students did not pursue full time
10 The Department’s reason for excluding the latter students from 2014 earnings dataset is that “groups with fewer than 10 individuals may not produce earnings data that would be reasonably representative of the value a CCI program.” (ER 118). The same concern infects the AER in a number of other ways that should have given the Department pause. It is the Department’s burden, not the Plaintiffs’, to establish that it has “examined the relevant data and articulated a satisfactory explanation for its action including a rational connection between the facts found and the choice made.” Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto Ins. Co., 463 U.S. 29, 43 (1983) (internal quotation and citation omitted).
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 15 of 27
12
employment while enrolled at Corinthian. In any event, applying the results of the
AER to Students who attended at precisely the time that the Department found
Corinthian’s behavior to be getting worse, see, e.g., (Dist. Ct. Dkt. No. 35-7 at 5-6)
(fine action letter), fails to address the unique harm suffered by such Students.11
C. The Average Earnings Rule, in its totality, is a conceptually flawed and illogical apples-to-oranges calculation.
The AER methodology is complex. But, complexity does not always
indicate rigor or rationality. Here, even if the Department had adopted a
reasonable formula, the rule is arbitrary and capricious because it is conceptually
flawed.
The Department baldly asserts that “the harm” to Students from Corinthian’s
conduct can be “measured as the difference between the value of the education that
the students received and an education that would have met students’ reasonable
expectations of preparation for gainful employment,” (ER 111) and further
determines, implicitly, that every Corinthian Student’s 2014 annual earnings as
reported by SSA is reasonable as a sole proxy for “the value of the education”
received. The AER is based on the faulty and unsupported assumption that, had
Corinthian not defrauded Students, and convinced them to enroll on the basis of
11 To be clear, even if “value” were a permissible metric for borrower defense relief, Plaintiffs do not believe that an individualized assessment of every Student’s experience would be required. However, the Department, in the name of efficiency, cannot rely on an irrational methodology that fails to capture what it purports to.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 16 of 27
13
false representations, the Students would have enrolled in some other for-profit
program subject to gainful employment. (ER 111). The Department provides no
explanation or evidence to justify this premise.
Furthermore, the rule ignores the Department’s own findings about the
specific way that Corinthian wronged its students. In April 2015, the Department
fined Corinthian $30 million for, among other things, including “placements that
were clearly out of the student’s field of study as in-field placements,” and failing
“to disclose that it counted as ‘placed’ those graduates whose employment began
prior to graduation, and in some cases even prior to the graduate’s attendance at
Heald.” (Supp ER 140-41). In other words, the Department recognized that many
Students found employment for reasons entirely independent of Corinthian. In
developing the Average Earnings Rule, the Department simply shrugged and
ignored these prior findings.
Nor does the AER provide Students with process to highlight these
distinctions and to contest the Department’s inapt “value” assessment. Unlike the
process provided to institutions in the GE regulations, see, e.g., 34 C.F.R. §§
668.405 & 406, here, Students have no opportunity to contest any step of the
Department’s construction of the AER, and its application to them. This is exactly
the kind of challenge that the Department should have provided pursuant to the
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 17 of 27
14
Privacy Act. See, e.g., 5 U.S.C. § 552a(p) (requiring the Department to provide
Students with an opportunity to contest the findings of the matching program).
Although the Department boasts that the AER cuts “the overall amount of
relief granted by around 60 percent” (ER 74), and saves the Department tens of
millions of dollars (Supp ER 27-28), the Department’s rule, taken piecemeal and
collectively, creates an apples-to-oranges comparison that fails to meet the
Department’s stated goal of determining the “value” a given Corinthian Student
received. It is arbitrary and capricious.12
II. A BORROWER DEFENSE REMEDY, UNDER LONGSTANDING DEPARTMENT PRACTICE, REGULATION, AND STUDENTS’ MASTER PROMISSORY NOTES, IS TIED TO THE STATE LAW CAUSE OF ACTION AND THUS THE AVERAGE EARNINGS RULE IS IMPERMISSIBLE. The Department independently acted in an arbitrary and capricious manner
in developing the Average Earnings Rule because it failed to “forthrightly
acknowledge[]” its past interpretation of the borrower defense provision, and failed
to provide a “reasoned explanation…for disregarding facts and circumstances that
underlay or were engendered by the prior policy” under which it adjudicated the
claims of Corinthian Students. FCC v. Fox Television Stations, Inc., 556 U.S. 502,
12 Even accepting “earnings” as a reasonable, stand-alone proxy for “value,” and accepting “value” as the appropriate inquiry, a true apples-to-apples comparison could entail assessment of a borrower’s earnings before attending a school and her earnings after leaving the school.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 18 of 27
15
515-17 (2009). An agency “that neglects to do so acts arbitrarily and
capriciously.” Jicarilla Apache Nation v. U.S. Dep’t of the Interior, 613 F.3d 1112,
1119 (D.C. Cir. 2010).
As discussed supra, Section I(A), the Department has proffered a pretextual
explanation for reworking its policy with respect to defrauded Corinthian Students.
Additionally, it mischaracterized the rule it replaced as “based on the assumption
that CCI borrowers received a worthless education and therefore that the discharge
of the total amount of borrowers’ loans and reimbursement of all payments was
appropriate for all CCI borrowers with valid claims.” (ER 109).
In reality, the Department unfailingly applied the Corinthian Rule until
January 20, 2017, and the Rule rested on far more than an “assumption” about
value.13 A Department Inspector General Report from December 2017
underscores the factual findings and legal analysis that justified the Corinthian
Rule. It notes: “it was the OUS [Office of the Under Secretary], OGC [Office of
the General Counsel], and the Special Master who determined in 2015 that “full
relief” (defined as full discharge of loans associated with the program at issue with
a full refund of amounts paid) was appropriate for Heald, Everest, and WyoTech
13 The District Court disagreed with Plaintiffs and stated that “the parameters of the Corinthian Rule are not clearly defined.” (ER 38). For the reasons discussed in this section, the District Court erred in this conclusion.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 19 of 27
16
borrowers with approved Job Placement Rate (JPR) claims.” (Supp ER 130). It
further reflects that “BD [Borrower Defense] attorneys continued review of JPR
claims under the same review criteria, legal framework, and relief previously
established by OUS, OGC, and the Special Master. Enforcement agreed with
OGC’s legal conclusions that: 1) borrowers who met the criteria required in the
Heald and Everest/WyoTech attestation forms were eligible for borrower defense
discharges; and 2) that full relief was appropriate.” (Supp ER 131) (emphasis
added); see also (Supp ER 131) (“[T]here was no uncertainty regarding OGC’s
legal positions on review criteria, legal framework, or relief”).14
Similarly, the Special Master for Borrower Defense explained in his first
report that “the Department looked to California law and determined that Heald’s
misrepresentations of placement rates constituted prohibited unfair competition
under California’s Unfair Competition Law (UCL). Accordingly, students that
relied on such misleading placement rates when they enrolled at Heald would have
a cause of action under state law.” (Dist. Ct. Dkt. No. 35-7 at 5) (Special Master
14 This Report references a number of memorandum that further detail the legal and factual basis of the Corinthian Rule. (Supp ER 120). The Department has taken the position that some of these reports are privileged and will form no part of the administrative record. (Dist. Ct. Dkt. No. 64 at 3-5) (Defs’ Supplemental Br. in Response to Preliminary Injunction Order). The dispute about the record was partially briefed, (Dist. Ct. Dkt. No. 80) (Pls’ Motion for Declaration that Documents are Not Privileged), before the District Court stayed the litigation pending this appeal.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 20 of 27
17
First Report). The remedy for a UCL violation is restitution. See Br. of Amicus
Curiae Cal., et. al. at 13-18.15 In his Second Report, the Special Master further
stated that, “the Department’s determination, after consultation with the Office of
the California Attorney General, that students who relied upon false or misleading
placement rate disclosures in enrolling in Heald College programs would have
established a borrower defense claim as to which relief would be granted under
California law.” (ER 62).
And, another Department memorandum, concerning the borrower defense
remedy for students at a different for-profit school, provides additional evidence
that the Corinthian Rule was not based on a mere “assumption.” Indeed, it refers
to a rich factual record evidencing “[r]epeated misleading statements to students,
regulators, and accreditors; misrepresentations regarding completion rates; [and]
elaborate job placement fraud.” (Supp ER 6). It further highlighted student
testimony on how “their affiliation with the school was an impediment rather than
an asset as they sought employment.” Id.
This same memo also confirms the Department’s longstanding
interpretation of the borrower defense regulation and the Students’ Master
15 A California Court has already determined that California law demands full restitution for Corinthian students. Final Judgment, People v. Heald Coll., Case No CGC-13-534793, 2016 WL 1130744 (Cal. Sup. Ct., filed Mar. 23, 2016) (Karnow, J.) (“California Corinthian Judgment”).
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 21 of 27
18
Promissory Notes (“MPN”) as tethering the borrower defense remedy as a whole
to the state law cause of action. (Supp ER 5-6) (discussing damages and remedy as
a function of state law). This interpretation was well justified. See Pls’ Opening
Br. at 48-50. The governing regulation states:
(c) Borrower defenses.
(1) In any proceeding to collect on a Direct Loan, the borrower may assert as a defense against repayment, any act or omission of the school attended by the student that would give rise to a cause of action against the school under applicable State law. … (2) If the borrower’s defense against repayment is successful, the Secretary notifies the borrower that the borrower is relieved of the obligation to repay all or part of the loan and associated costs and fees that the borrower would otherwise be obligated to pay. The Secretary affords the borrower such further relief as the Secretary determines is appropriate under the circumstances. Further relief may include, but is not limited to…[.]
34 C.F.R. § 685.206. 16 It then goes on to specify certain discretionary remedial
steps within the Secretary’s authority under the Higher Education Act. Id. The
MPN similarly defines a borrower defense as “a legal cause of action against the
school under applicable state law.” (ER 140).
The term “cause of action” is critical; that phrase implicates both the
substance of a claim and a claimant’s remedy. For example, to plead a cause of
16 A new borrower defense regulation went into effect in October 2018. See 34 C.F.R. § 685.222. That regulation incorporates the borrower defense provision in 34 C.F.R. § 685.206(c) for individuals who borrowed loans before the new regulation was effective. The change does not impact the governing regulation and MPN in this case.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 22 of 27
19
action for fraudulent misrepresentation under California law, a “plaintiff must
prove that the defendant caused actual harm.” Howe Investments Ltd. v. Quarles &
Brady, LLP, No. 17-55483, 2019 WL 581370 at *1 (9th Cir. Feb. 13, 2019); see
also Graham v. Bank of Am., N.A., 226 Cal. App. 4th 594, 605-06 (2014); but see
Gonzales v. Cal. Dep’t of Corr., 739 F.3d 1226, 1232-33 (9th Cir. 2014) (utilizing
“primary right” language). Similarly, a plaintiff who seeks specific performance
on a breach of contract claim must “plead facts that establish a right to recover for
breach as well as the right to receive the equitable remedy.” 12 Harry D. Miller &
Marvin B. Starr, Cal. Real Estate § 40:17 (4th ed. 2017). A plaintiff’s remedy is
inexorably tied to the cause of action she asserts.
Notably, the Department has agreed with this interpretation for over twenty
years. There is no dispute that the Department has, without fail, used state law to
decide every borrower defense claim (including remedy) before January 20, 2017.
See (Supp ER 4) (using Massachusetts law); (Supp ER 8) (instructions for
determining borrower defense relief); (Supp ER 84) (utilizing North Dakota law);
(Supp ER 131) (referencing the relevant state statute of limitation for the remedy for
Corinthian transferability of credit claims). Nor do the parties dispute that the
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 23 of 27
20
Average Earnings Rule is detached from state law.17 The Department’s shift is an
impermissible “unexplained inconsistency in agency policy” and a “reason for
holding [the] interpretation to be an arbitrary and capricious change.” Encino
Motorcars, LLC v. Navarro, 136 S.Ct. 2117, 2126 (2016); Am. Wild Horse Pres.
Campaign v. Perdue, 873 F.3d 914 (D.C. Cir. 2014) (finding arbitrary and capricious
an unexplained change of a non-binding policy).
CONCLUSION
Corinthian defrauded hundreds of thousands of students who were simply
seeking the promise of higher education. The Department of Education, after
failing to provide proper oversight of Corinthian on the front end, has now adopted
an irrational borrower defense policy that compounds the harm Corinthian caused.
The Court should affirm the District Court’s preliminary injunction because the
Department’s actions were arbitrary and capricious.
Date: March 5, 2019
/s Joshua D. Rovenger
Noah Zinner Megumi Tsutsui HOUSING & ECONOMIC RIGHTS ADVOCATES PO Box 29435
17 Under the Department’s new interpretation—one that severs substance and remedy into two separate questions—an individual could bizarrely establish a “successful borrower defense claim,” but be entitled to no relief.
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 24 of 27
21
Oakland, CA 94604 Tel.: (510) 271-8443 Fax: (510) 280-2448 Eileen M. Connor Toby R. Merrill Joshua D. Rovenger LEGAL SERVICES CENTER OF HARVARD LAW SCHOOL 122 Boylston Street Jamaica Plain, MA 02130 Tel.: (617) 390-3003 Fax: (617) 522-0715 Attorneys for Plaintiffs-Appellees
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 25 of 27
CERTIFICATE OF COMPLIANCE
I certify that this brief complies with the requirements of the Court’s Order
requesting supplemental briefing and the requirements of Federal Rules of
Appellate Procedure 32(a)(5) and 32(a)(6), because it has been prepared in a
proportionately spaced typeface using Microsoft Word in Times New Roman 14-
point font, and does not exceed 20 double-spaced pages.
Date: March 5, 2019
/s Joshua D. Rovenger___ Joshua D. Rovenger
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 26 of 27
CERTIFICATE OF SERVICE
I hereby certify that on March 5, 2019, I electronically filed the foregoing
with the Clerk of the Court for the United States Court of Appeals for the Ninth
Circuit by using the appellate CM/ECF system.
Participants in the case who are registered CM/ECF users will be served by
the appellate CM/ECF system.
Date: March 5, 2019
/s Joshua D. Rovenger Joshua D. Rovenger
Case: 18-16375, 03/05/2019, ID: 11216954, DktEntry: 60, Page 27 of 27