Opinion
2:07-cv-461
05-06-2014
UNITED STATES OF AMERICA et al, Plaintiffs, v. EDUCATION MANAGEMENT LLC et al, Defendants.
MEMORANDUM OPINION AND ORDER
Pending before the Court is DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (ECF No. 369) filed by EDMC, with brief, Concise Statement of Material Facts, and exhibits in support. Instead of a response in opposition to the motion, Plaintiffs have filed a MOTION PURSUANT TO FED. R. CIV. P. 56(d) TO DENY DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (ECF No. 384), with brief and appendix in support. The motions are ripe for disposition without further input from the parties.
DEFENDANTS' MOTION FOR ORAL ARGUMENT ON DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (ECF No. 376) is DENIED. The issues may be resolved on the written submissions.
Prompt resolution of the motions is important so that the parties will have appropriate guidance in advance of their meeting with the Special Master on May 23, 2014 regarding discovery matters.
The Court appreciates EDMC's efforts to seek an end to this massive, complex and expensive litigation. The Court shares EDMC's concerns regarding the time and expense of discovery in this case, and it is certainly possible that EDMC has properly compensated its Assistant Directors of Admissions ("ADAs") in compliance with all government requirements. Unfortunately, the instant summary judgment motion -- prior to the close of discovery and based on EDMC's proffered "statistical evidence" -- cannot provide that avenue for finality.
Factual and Procedural Background
The design and structure of the federal student loan program, pursuant to Title IV of the Higher Education Act of 1965 ("HEA"), 20 U.S.C. § 1070 et seq., is fundamentally flawed. The entities that actually receive the billions of dollars of federal money (the educational institutions) have no obligation to pay the money back if their services are ineffective. To borrow a phrase from Warren Buffett, schools have no "skin in the game." The risk of defaults on student loans is borne not by the educational institutions but by the students and taxpayers, who absorb the cost of any defaults. Association of Accredited Cosmetology Schools v. Alexander, 979 F.2d 859 (D.C. Cir. 1992).
The total amount of student loan debt is staggering. As of December 31, 2013, as reported by the Federal Reserve Bank of New York, outstanding student loan balances reported on credit reports totaled $1.08 trillion ($1,080,000,000,000.00). See http://www.newyorkfed.org/householdcredit/2013-Q4/HHDC 2013Q4.pdf. The Federal Reserve reported that 11.5% of student loan balances are 90+ days delinquent or in default. Id.
Because the schools receive payment in full, there is little economic incentive for them to limit student enrollments. Id. This has led to perceived abuses of government funding by some schools. See generally "Abuses in Federal Student Aid Programs," Sen. Rep. No. 102-58 (May 17, 1991) ("Senate Report") (unscrupulous elements have exploited "both the ready availability of billions of dollars of guaranteed student loans and the weak and inattentive system responsible for them, leaving hundreds of thousands of students with little or no training, no jobs, and significant debts that they cannot possibly repay. While those responsible have reaped huge profits, the American taxpayer has been left to pick up the tab for the billions of dollars in attendant losses.").
In this case, Plaintiffs allege that EDMC pursued such an enrollment-maximization strategy, with a goal to dramatically increase student enrollment from 4,500 in 2006 to 50,000 in 2011. Intervenor Complaint ¶ 128(a)). Plaintiffs aver that EDMC accepted all potential students who completed an application, regardless of their high school grades or the quality of their written essay. Intervenor Complaint ¶ 106. As pled, the annual federal student aid funds received by EDMC increased rapidly, from $656 million in 2003-2004 to $2.578 billion in 2010-2011. Intervenor Complaint ¶ 73, 80. Nevertheless, the issue in this case is not whether such alleged opportunistic exploitation of a poorly-designed government program was unsavory. Rather, Plaintiffs must prove that EDMC's conduct was illegal.
The Higher Education Act does not require educational institutions to repay student loan proceeds if the students are unable to do so. Instead, Congress chose a weak and indirect method to offset the substantial economic incentive for schools to over-expand enrollments - namely, by restricting the ways in which schools compensate their employees who recruit prospective students. United States ex rel. Main v. Oakland City Univ., 426 F.3d 914, 916 (7th Cir. 2005); United States ex rel. Lopez v. Strayer Educ., Inc., 698 F. Supp.2d 633, 635 (E.D. Va. 2010). The "Incentive Compensation Ban" was originally enacted by Congress in 1992, shortly after the Senate Report, and remains in place. Specifically, 20 U.S.C. § 1094(a)(20) states: "The institution will not provide any commission, bonus, or other incentive payment based directly or indirectly on success in securing enrollments or financial aid to any persons or entities engaged in any student recruiting or admission activities or in making decisions regarding the award of student financial assistance. . . ." This statutory prohibition was furthered weakened by the implementing regulation adopted by the DOE, 34 C.F.R. § 668.14(b)(22), which created a "Safe Harbor." Of particular relevance to this case, the Safe Harbor provided that an institution would not violate the Incentive Compensation Ban by: "The
The Safe Harbor regulation remained in effect for the entire time period at issue in this case. The regulation was amended and the Safe Harbor was eliminated as of July 1, 2011. EDMC has submitted an email in which a DOE staffer described the Safe Harbor as a loophole visible from outer space. CSMF ¶ 6. The Court need not address Plaintiffs' argument regarding the alleged conflict between the statute and regulation.
payment of fixed compensation, such as a fixed annual salary or a fixed hourly wage, as long as that compensation is not adjusted up or down more than twice during any twelve month period, and any adjustment is not based solely on the number of students recruited, admitted, enrolled, or awarded financial aid."
In its Memorandum Opinion and Order dated May 11, 2012, the Court recognized that Plaintiffs were pursuing two distinct theories of fraud under the False Claims Act. The Court granted EDMC's motion to dismiss Plaintiff's theory that the ADA compensation plan violates the Incentive Compensation Ban "as written" but denied the motion to dismiss regarding Plaintiffs' challenge to the Plan "as implemented," which it summarized as follows: "EDMC has used the written compensation plan as a sham or cover-up for its actual implementation of a compensation system based solely on quantitative factors, thereby violating the Incentive Compensation Ban." Plaintiffs allege a knowing decision by EDMC executives to perpetuate a company-wide sham Plan to cover up for prohibited compensation practices. To put it starkly, Plaintiffs allege a coordinated, multi-billion dollar corporate-wide fraud. Plaintiffs concede that EDMC's written compensation policy contains a "matrix" which combines student enrollment numbers with a variety of "quality factors." However, they argue that the "quality factors" have no real impact on the manner in which EDMC's compensation system is actually implemented.
In other words, Plaintiffs' theory of the case is that even if the paperwork looked correct on its face, such paperwork was only a pretext or cover-up and did not reflect EDMC's real compensation practices. The Court recognized that Plaintiffs are attempting to "prove a negative"—i.e., that the quality factors were not actually used, even if the paperwork showed otherwise. Direct evidence of this alleged fraud would be possessed by a select number of top EDMC executives, but Plaintiffs are also entitled to an opportunity to prove their case through circumstantial evidence.
EDMC has attempted several times to end this litigation by pointing to its internal documents regarding ADA compensation (i.e., its Salary Worksheets). In support of its motion to dismiss, EDMC pointed to Lynntoya Washington's own Performance Review, which appears to show that the quality factors were taken into account. The Court explained that this document could not dispose of the "as implemented" claim, and pointed to the allegations in the Complaint which indicated that the quality factors, in practice, had no effect on her compensation.
In support of its pending Motion to Dismiss Relators' False Claims Act Claims with Prejudice (ECF No. 378), which will be ruled on separately, EDMC has submitted additional documents which reflect that Washington successfully challenged her rating on quality factors on at least one occasion. The Court is not aware of any evidence which reflects that the increased rating resulted in any increase in her actual compensation. See Intervenor Complaint ¶ 138 (alleging Washington was told that because she met her enrollment numbers, her quality factor rating "would have no effect whatsoever on her compensation").
In the initial discussions regarding case management and discovery, EDMC sought to carve out a preliminary phase for: (1) production of its ADA Salary Worksheets database; (2) a tutorial hearing at which EDMC would provide expert statistical evidence to demonstrate why the salary adjustment forms should be dispositive of this case; (3) an opportunity for Plaintiffs to "get behind the numbers" and engage their own experts regarding the Salary Factor Database; and (4) dispositive motions at the close of this phase of discovery. In its November 8, 2012 Memorandum Order, the Court rejected this proposal; authorized conventional, unlimited discovery as proposed by Plaintiffs; and stated that EDMC's focus on statistical and documentary evidence is difficult to reconcile with Plaintiffs' "as implemented" theory of the case.
EDMC continued to rely on Dr. Lazear's statistical evidence in objecting to the scope of discovery (see ECF No. 266-4). In R&R #2, the Special Master noted: "the Court has twice rejected Defendants' contention that statistical evidence alone can disprove Plaintiffs' allegations and the Special Master has previously ruled similarly." United States v. EDMC, 2013 WL 3854453 *21 (W.D. Pa. 2013). In denying EDMC's objections to Special Master R&R #2, the Court again confirmed that: "Plaintiffs are entitled to prove their allegations that the ADA Compensation Plan was a sham through circumstantial evidence that the non-enrollment factors on the Grid (i.e., the Quality Points) were not actually of significance. Such evidence could allow a jury to conclude that ADA compensation was based solely on enrollments." The Court remained unpersuaded by EDMC's citation to Dr. Lazear and adhered to its earlier statements regarding the limitations of statistical evidence at this stage of the case. ECF No. 291 at 7.
The pending motion for summary judgment represents another bite at the "statistical evidence" apple. Discovery is still in a relatively early stage. The parties have exchanged interrogatories and document requests; have made rolling productions of documents; and have taken the depositions of relators Washington and Mahoney. The Court recently authorized EDMC to take other early depositions (involving a DOE attorney and two Rule 30(b)(6) designees of DOE), while cautioning that resolution of the difficult attorney-client privilege and work product issues implicated by those depositions would not be a valid reason to postpone other discovery. As relevant to the pending motion, Plaintiffs have not yet had a full opportunity to take discovery to develop evidence in support of their "as implemented" theory.
Discussion
EDMC contends in the instant motion for summary judgment that the statistical analyses of its contemporary business records and the Declaration of Edward P. Lazear, Ph.D., prove that ADA compensation was not based solely on student enrollments. EDMC submits that its Salary Worksheets database reflects the values that were used to calculate the actual salary adjustments (including enrollments and quality points) for each ADA, and that Dr. Lazear has performed various rigorous analyses of that data. However, EDMC proceeds from a flawed fundamental premise -- namely, that the data in its Salary Worksheets is real. But the validity of the underlying data is disputed. Plaintiffs recognize that quality factor ratings were filled out as to each ADA, but argue that those ratings were based on no real standards and had no real impact on actual compensation. To repeat, Plaintiffs allege that EDMC knowingly schemed to evade the Safe Harbor. The fact that EDMC's paperwork and salary database appear to be compliant, on its face, is entirely consistent with Plaintiffs' theory of the case. A salary database with clear red flags would not provide much cover or be very effective camouflage. EDMC is accused of mendacity, not incompetence.
Even if the authenticity of the EDMC salary documents is established, the accuracy of the quality factor ratings contained in the Salary Worksheets cannot be established based on the existing record. Dr. Lazear made no independent evaluation of the data contained in the Salary Worksheets. Instead, as described in his Declaration, Dr. Lazear assumed that each version of the Plan used a number of factors on which recruiters' salary adjustments were to be based, including "quality factors." Lazear Declaration ¶ 21. The compensation factors used to evaluate each recruiter during each review period were simply obtained from the EDMC Salary Worksheets. Lazear Declaration ¶ 24. To put it another way, Dr. Lazear took the data from the face of the Salary Worksheets, but he did not conduct any analysis to verify whether that data reflected the honest and accurate evaluation of each supervisor regarding the performance of each ADA on each quality factor. Similarly, Lynn Walwender Kossick, Product Manager in Compensation for EDMC, who was responsible for aggregating the ADA performance evaluation data which comprised the database, stated that she "received data from the ADAs supervisor, corresponding to the qualitative evaluation of the ADA for the prior six months." Kossick Declaration ¶ 5.
An analysis based solely on EDMC's own data cannot be dispositive at this stage of the case. What EDMC portrays as "getting behind the numbers" may really be only manipulation of its numbers. The real issue is not whether quality factor evaluations were assigned to each ADA but whether those evaluations were real and legitimate. For example, Plaintiffs allege that the quality factor evaluations may have been filled in after ADA salary was already determined based solely on enrollments; were assigned arbitrarily; or were given a default value of "meets expectations." Intervenor Complaint ¶ 389-391. Similarly, in ¶ 29-30 of his Declaration, Dr. Lazear posits that the existence of a counter-example (e.g., an instance in which ADAs recruited the same number of students but received different compensation) conclusively proves that Plaintiffs' allegation of fraud is false. Not so. While the existence of a counter-example(s) may support the inference that there was no fraud, it cannot disprove the theory that such counter-example(s) was part of the camouflage. In sum, Dr. Lazear's descriptive analysis and statistical/regression analysis of the EDMC-created records cannot prove whether or not the quality factors were a proxy or smoke screen. See United States v. Corinthian Colleges, 655 F.3d 984, 994 (9th Cir. 2011) ("as implemented" theory cannot be resolved simply on the facial compensation numbers, without an understanding of what the employee must do to achieve those numbers). In short, as a predicate to Dr. Lazear's statistical analysis, EDMC must establish that the data contained in the Salary Worksheets reflects the actual evaluations of each ADAs performance on the quality factors, such that those quality factors had an actual effect on ADA compensation. Thus, the motion for summary judgment must be denied at this stage.
To be sure, Plaintiffs continue to face a difficult burden to succeed on their claims. They must develop evidence of a top-down, corporate-wide fraud - not merely isolated instances of inadequate evaluations by supervisors. Pressure on recruiters to increase enrollments, without more, will not suffice. At trial, Plaintiffs must "prove a negative" and show that the quality factors were not used, such that EDMC violated the Incentive Compensation Ban. The quality factors are inherently subjective, such that individual evaluations may be difficult to criticize (or defend), and may require credibility determinations by the jury. Some aspects of the Lazear Declaration have initial appeal, in particular the data which appear to reflect that ADAs who recruited similar numbers of students received very different compensation.
To clarify an apparent misunderstanding of Plaintiffs (see Brief at 4, 14), the mere fact that EDMC closely tracked ADA student enrollment numbers does not establish liability. Rather, a stark contrast in how EDMC tracked enrollment numbers as compared to quality factors might support an inference that the quality factors were not actually used.
The Court is aware that possible counter-arguments exist and are not yet ripe for presentation.
However, to succeed on summary judgment, the burden is on EDMC to demonstrate -- despite drawing all inferences in the light most favorable to Plaintiffs -- that no reasonable jury could conclude that ADA compensation was based solely on enrollments. For example, EDMC may be able to conclusively demonstrate that the quality factors were, in fact, rigorously applied in the ADA evaluations. One would expect to find robust internal controls to ensure compliance with the Incentive Compensation Ban because EDMC's corporate revenues were so heavily dependent upon government funding. For example, there may be executive-level directives regarding the importance of the quality factors in ADA compensation and ensuring compliance with the Incentive Compensation Ban may affect executive and management compensation. There should be corporate-wide standards and definitions for application of the quality factors; training materials and guidance memoranda for all levels of management tasked with implementing the Incentive Compensation Ban; training materials and guidance memoranda for the Directors of Admission ("DOAs") specifically tasked with evaluating the ADAs; questions and correspondence from the DOAs regarding the application of quality factors; instances in which DOAs were counseled or sanctioned by their supervisors for mis-application of the quality factors; documentation of individual ADA performance on the quality factors; and reports from internal and/or external auditors regarding the effectiveness of EDMC's implementation of the quality factor rankings and Incentive Compensation Ban. See generally Special Master R&R #2 (discussing relevance of Plaintiffs' discovery requests). In the Court's view, the implementation of the Incentive Compensation Ban through various levels of EDMC management, culminating in the evaluation of quality factors by the DOAs, is likely to be the heart of this case. Discovery must proceed. The parties may then seek summary judgment, if appropriate.
Some such documents appear to exist (see, e.g., Quality Factor Descriptions, ECF No. 145, Exh. 5), although it is uncertain at this stage of the case whether they were actually used.
--------
Conclusion
In accordance with the foregoing, PLAINTIFFS' MOTION PURSUANT TO FED. R. CIV. P. 56(d) TO DENY DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (ECF No. 384) will be GRANTED; DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (ECF No. 369) will be DENIED; and DEFENDANTS' MOTION FOR ORAL ARGUMENT ON DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (ECF No. 376) will be DENIED.
An appropriate Order follows.
McVerry, J.
UNITED STATES OF AMERICA et al, Plaintiffs,
v.
EDUCATION MANAGEMENT LLC et al, Defendants.
2:07-cv-461
ORDER
AND NOW, this 6th day of May, 2014, in accordance with the foregoing Memorandum Opinion, it is hereby ORDERED, ADJUDGED, and DECREED that PLAINTIFFS' MOTION PURSUANT TO FED. R. CIV. P. 56(d) TO DENY DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (ECF No. 384) is GRANTED; DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (ECF No. 369) is DENIED; and DEFENDANTS' MOTION FOR ORAL ARGUMENT ON DEFENDANTS' MOTION FOR SUMMARY JUDGMENT (ECF No. 376) is DENIED.
BY THE COURT:
Terrence F. McVerry
United States District Judge
cc: All counsel of record