Judge Manish S. Shah

MEMORANDUM OPINION AND ORDER

Plaintiffs are former students of a for-profit university who enrolled because of marketing representations boasting about the high rate at which former graduates of the school found employment in their chosen fields. Plaintiffs claim that those representations were false and now sue the entities that operated the school, alleging violations of various states' consumer protection laws and unjust enrichment. Defendants move to dismiss the amended complaint. For the reasons explained below, the motion is granted.

I. Legal Standards

To survive a motion to dismiss, a complaint must contain factual allegations that plausibly suggest a right to relief. Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009). In considering a motion to dismiss, I accept the facts alleged in the complaint as true and draw reasonable inferences from those facts in plaintiffs' favor, but I do not accept as true the complaint's legal conclusions. Id. at 678-79. I consider the complaint, exhibits attached to the complaint, and, if they are central to the claims, documents referenced by the complaint. Tobey v. Chibucos, 890 F.3d 634, 648 (7th Cir. 2018).

II. Facts

Plaintiffs are former students of DeVry University, a for-profit university operated by defendants Adtalem Global Education, Inc. (formerly known as DeVry Education Group, Inc.), DeVry University, Inc., and DeVry/New York, Inc. [55] ¶¶ 2, 7, 52-54. DeVry used an expansive marketing campaign to recruit students, spending many millions of dollars each year on television, print, radio, and internet advertisements in addition to in-person pitches. [55] ¶¶ 61-62. For years, the employment rate of DeVry graduates was a core component of the recruitment campaign. [55] ¶¶ 64, 68. Generally, DeVry boasted that 90% of its graduates who were actively seeking employment obtained new jobs in their chosen field of study within six months of graduation. [55] ¶ 64.

The 90% representations appeared in advertisements in different variations. Often it was accompanied by explanations, footnotes, or definitions, and sometimes the employment rate slightly differed by a few percentage points. [55] ¶¶ 66-67. For example, one website advertisement said: "In 2012, 90% of DeVry University grads actively seeking employment had careers in their field within six months of graduation." [55] ¶ 69. A footnote clarified that the figure was based on self-reported data of those who were employed at graduation or actively seeking employment and that it did not include graduates who DeVry determined were not actively seeking employment or those who did not report their employment status. [55] ¶ 69. An asterisk to a similar 90% representation appearing on a brochure on the DeVry website noted that "[a]ctive job market includes those already employed prior to graduation." [55] ¶ 71. The fine print on another similar representation explained that the 90% figure did not include master's degree graduates. [55] ¶ 74. Many other 90% representations came with no caveats or explanations. [55] ¶¶ 70, 72, 75. The main text of the representations varied. One website said that "[o]f graduates in the active job market, 90.1% were employed in career-related positions within six months of graduation." [55] ¶ 79. The representations sometimes noted that the employment rate was based on data from specific time frames, and those time frames differed. See, e.g., [55] ¶ 79. Variations of the 90% representation appeared throughout DeVry's marketing, from television and YouTube to emails and phone calls. [55] ¶¶ 68-79.

Plaintiffs, all former DeVry students, saw or heard 90% representations before enrollment, and they all relied on the representations when deciding to enroll. [55] ¶¶ 6-7. If plaintiffs had known that the representations were false, they would not have enrolled at DeVry or they would have paid less for tuition. [55] ¶ 8.

Plaintiffs believe that the 90% representations were not true. [55] ¶ 81. Plaintiffs' belief is based on allegations from lawsuits or investigations conducted by the Federal Trade Commission, the Department of Education, the attorneys general of New York and Massachusetts, and DeVry shareholders. [55] ¶ 86. The allegations are that though DeVry relies on student files maintained by its career services office for the 90% figure, the files do not substantiate it. [55] ¶¶ 82-83. DeVry padded the employment rate by counting as employed graduates who worked at jobs they had before attending DeVry and graduates whose jobs are not reasonably considered to be in their chosen field (like food service workers and unpaid volunteers). [55] ¶ 83. And DeVry shrunk the figure's denominator by excluding certain students who were in fact actively seeking employment. [55] ¶ 84. These unreasonable methods distorted the employment rate, which was "markedly" below 90%. [55] ¶ 85.

As a result of a settlement with the FTC, DeVry had to stop using the 90% representations in its advertising campaign in December 2016. [55] ¶ 124. In April 2017, DeVry lowered tuition for its undergraduate Tech Path associate and bachelor's degree programs from $609 per credit hour to $487. [55] ¶ 126.

Plaintiffs brought this putative class action lawsuit against DeVry in 2016. [1]. DeVry moved to dismiss the complaint, [23], and I granted the motion, dismissing the complaint without prejudice. [46]. The dismissal was primarily due to plaintiffs' failure to meet the heightened pleading standard for fraud. Plaintiffs did not adequately plead what specific representations each plaintiff saw or when or where they saw them, nor did they sufficiently plead the basis for their "on information and belief" allegations that the 90% representations were false. I also found plaintiffs' allegations of damages to be too speculative. Plaintiffs amended their complaint. [55]. DeVry now moves to dismiss the amended complaint, arguing that the amendment did not cure the deficiencies I identified in the first complaint. DeVry also moves to strike the class allegations.

III. Analysis

Plaintiffs allege that DeVry's misrepresentation of its graduates' employment rate violates the consumer protection statutes of various states and constitutes unjust enrichment. DeVry's primary arguments are that plaintiffs fail to sufficiently allege damages caused by the 90% representations and do not meet Rule 9(b)'s particularity requirement for allegations of fraud. Though each state's consumer protection statutes and accompanying case law may vary, these two arguments are common to all the statutory claims (plaintiffs have not argued otherwise).

A. Damages

Plaintiffs rely on a benefit-of-the-bargain theory of damages, a viable theory for consumer fraud claims. See Kim v. Carter's Inc., 598 F.3d 362, 365 (7th Cir. 2010). Plaintiffs allege that DeVry's misrepresentation of its graduates' employment success caused them to pay more for their education than they otherwise would have if they had known the truth. DeVry criticizes plaintiffs' damages theory, noting that "the Complaint does not preclude the possibility that each Plaintiff landed a dream job in a career-related field within days of graduating from [DeVry]." [60] at 14. But plaintiffs' damages theory does not depend on their actual employment outcomes. Even if a plaintiff went on to a lucrative dream job after graduation, she would still have been damaged if an allegedly false 90% representation caused her to pay more for her education than it was worth. Put another way, plaintiffs' theory is based on an assumption that the value of a DeVry education "is a function of the probabilistic career-advancing potential of the education . . . irrespective of whether [an] individual student ultimately realizes the potential." Harnish v. Widener Univ. Sch. of Law, 833 F.3d 298, 308 (3d Cir. 2016). One can imagine being willing to pay more to attend a school with an 80% employment rate than one with a 20% employment rate, even understanding that the historical rate is not a promise or a guarantee of your own outcome, because the historic employment rate suggests that the education program you are paying for is of a certain level of quality that produces those results. It is not a promise or guarantee of any individual's outcome but rather it can be reasonably understood to be a statement about the quality of the education.

But is it plausible that the 90% representation caused plaintiffs to pay more in tuition? Plaintiffs allege that several months after an injunction issued prohibiting DeVry from making 90% representations, DeVry lowered its tuition for certain of its undergraduate programs by 20%. [55] ¶ 127. It is reasonable to infer from the tuition decrease that DeVry's allegedly inflated employment-rate representations caused tuition price to be higher than it otherwise would have been, meaning it is plausible that plaintiffs suffered actual damage in some amount. DeVry is right that the tuition decrease could have been caused by other factors, like a change in business model or cost-cutting, but at this stage of the litigation, I draw inferences in plaintiffs' favor, not DeVry's.

That said, I do not agree with plaintiffs' position that the 2017 post-decrease tuition amount reflects the "true value" of plaintiffs' educations. That number would require further factual development to address things like DeVry's points about economic factors and the passage of time and probably expert testimony too. Anyway, plaintiffs cannot rely on a price-inflation theory to prove actual damages in consumer fraud claims. See Harnish, 833 F.3d at 312; New Jersey Citizen Action v. Schering-Plough Corp., 367 N.J.Super. 8, 15 (App. Div. 2003). Instead, they must prove that the misrepresentations caused each individual plaintiff to pay more. But at this stage, the tuition decrease lends plausibility to plaintiffs' individual allegations that they were damaged. And though damages must be calculable, see Morris v. Harvey Cycle & Camper, Inc., 392 Ill.App.3d 399, 402 (1st Dist. 2009), it is too early to say that plaintiffs would not be able to calculate the true value of the education they paid for.

B. Compliance with Rule 9(b)

Even though plaintiffs have cleared the damages hurdle, they have not adequately alleged fraud. Plaintiffs' pleading burden is heightened for allegations of fraud, which must be stated with particularity. Fed. R. Civ. P. 9(b). Plaintiffs must describe "the who, what, when, where, and how" of the fraud, though precisely how much information is required for each of those descriptions varies depending on the facts of a case. Webb v. Frawley, 906 F.3d 569, 576 (7th Cir. 2018). Plaintiffs are required to conduct their own careful pretrial investigations to "inject[ ] precision and some measure of substantiation into their allegations of fraud." United States ex rel. Berkowitz v. Automation Aids, Inc., 896 F.3d 834, 840 (7th Cir. 2018).

1. Who Saw What

DeVry argues that the amended complaint does not allege the misrepresentations plaintiffs relied on with the requisite specificity, a problem from the first complaint. Plaintiffs have increased the level of detail surrounding the misrepresentations each of them heard or saw, often providing the general time frame and location. For example, Polly saw DeVry television ads "in or around July 1999 through January 2000," that "made the specific representation that 90% of DeVry's graduates actively seeking employment had careers within their field of study within six months of graduation from DeVry," and she heard "this same representation" from a DeVry recruiter named Gary in September 1999. [55] ¶ 11. Most of the other plaintiffs allege the representations they heard with a similar level of specificity, with two exceptions that I discuss below. The representation each plaintiff heard is the same (with the exception of Anderson, Diaz, Jara, and Villa, who heard the same representation except with 85%, 90.3%, 95%, and 96% figures, respectively). The allegations about what plaintiffs heard do not include whether the representation included any asterisks, footnotes, or other explanation. In the context of this case and this complaint, those allegations are not enough because, as I explained in the first opinion, "the lack of uniformity among the representations alleged and the potential consequences of that variation" require plaintiffs to be more specific. [46] at 7.

If there were only one DeVry ad or all DeVry ads had exactly the same 90% representation, then perhaps what plaintiffs have alleged would be sufficient. DeVry would know the wording of the representation made to plaintiffs along with the asterisks and explanations that came with it. But the complaint shows that the representations varied. For example, some representations explain that the 90% figure includes students who were already employed at graduation, see, e.g., [55] ¶ 71, which, one could argue, reasonably discloses that the figure includes graduates with jobs they had before enrolling (a category of students plaintiffs allege it was misleading for DeVry to include). Others disclosed that it was based on self-reported employment status, see, e.g., [55] ¶ 74, and the reliability of self-reports could be material. And plaintiffs' revived argument that DeVry is in the better position to compile a list of the representations is unavailing now for the same reason that it was before—plaintiffs are the only ones who know what they saw.

Plaintiffs argue that since they have alleged a fraudulent scheme, they only need to plead representative samples and need not detail each individual representation. See Goldberg v. Rush Univ. Med. Ctr., 929 F.Supp.2d 807, 819 (N.D. Ill. 2013). As I explained in the earlier opinion, the representative examples must nonetheless be "tied to the plaintiffs—a necessary element of their claims." [46] at 6. Most of the representative samples are not tied to any plaintiff, so they do not provide the specific representations that plaintiffs allege caused their damages. There are two exceptions—the complaint alleges that plaintiffs Annette Pearson and Thomas Pearson both saw particular representations that are attached to the complaint. [55] ¶¶ 28, 31. Annette allegedly saw an email containing a 90% representation, but that email was sent to Pearson after she had already graduated from DeVry, [55-1] at 2, meaning it could not have caused her enrollment. Thomas saw a particular webpage with a 90.2% representation in August 2008, [55-10] at 2, which sufficiently demonstrates the content of the representation. But the complaint alleges that Thomas heard other representations and that he "relied upon Defendants' '90%' representation in choosing to enroll at DeVry," [55] ¶ 29, without stating that the 90.2% website representation was what caused him to enroll (as opposed to the generally alleged representations). Anyway, even if plaintiffs had alleged the representations that they saw with sufficient particularity, their claims would fail because they do not adequately allege falsity.

2. Falsity

Generally, "Rule 9(b) does not require a plaintiff to demonstrate that a representation was indeed false." Hefferman v. Bass, 467 F.3d 596, 601 (7th Cir. 2006). See also Camasta v. Jos. A. Bank Clothiers, Inc., 761 F.3d 732, 737 (7th Cir. 2014). That is to say, a plaintiff can just plead that something was false without pleading factual allegations demonstrating the falsity. But here, plaintiffs plead the falsity of the 90% representation "on information and belief," [55] ¶ 81, meaning they do not actually know whether DeVry's representations are false, but they have a good-faith basis for believing it is based on second-hand information. Cincinnati Life Ins. Co. v. Beyrer, 722 F.3d 939, 948 (7th Cir. 2013). Courts "frown on making allegations 'on information and belief' in the fraud context and generally find that such claims do not meet Rule 9(b)'s particularity requirement," but the practice is allowed under limited circumstances. Id.

Plaintiffs can only plead fraud "on information and belief" if "(1) the facts constituting the fraud are not accessible to the plaintiff and (2) the plaintiff provides 'the grounds for his suspicions.'" Pirelli Armstrong Tire Corp. Retiree Med. Benefits Tr. v. Walgreen Co., 631 F.3d 436, 442 (7th Cir. 2011) (citation omitted). "The grounds for the plaintiff's suspicions must make the allegations plausible." Id. at 443 (emphasis in original). The facts that would establish the falsity of DeVry's 90% representations are not accessible to the plaintiffs, so the question is whether the grounds for plaintiffs' claim that the representations are false make it plausible.

The grounds for plaintiffs' suspicions come entirely from others' lawsuits against DeVry—untested allegations made by the FTC, Department of Education, and attorneys general of Massachusetts and New York and confidential witness statements made in a DeVry securities class action. "Although the federal rules are flexible in this realm, allegations based on the mere filing of other lawsuits generally will not provide much in the way of plausible corroboration of a plaintiff's fraud." Id. The government allegations that plaintiffs rely on are just that—allegations. All four government actions resulted in settlement, and plaintiffs point to no factual findings. The complaint does allege that the FTC's allegations came after it received millions of documents from DeVry, [55] ¶ 88, suggesting that the allegations are based on that documentary evidence. Still, the FTC's allegations are unproven allegations that do little to move the needle of plausibility. DeVry also argues that plaintiffs cannot rely on confidential witnesses who they appear to never have even spoken to or met. Plaintiffs did not respond to the argument, and I agree with DeVry. Anyway, the confidential witnesses added little corroborative value to the falsity of DeVry's 90% representations even in the original lawsuit. See Pension Tr. Fund for Operating Engineers v. DeVry Educ. Grp., Inc., No. 16 C 5198, 2017 WL 6039926, at *8 (N.D. Ill. Dec. 6, 2017).

Plaintiffs point to Pension Trust Fund, where the district court found the plaintiff to have done "just enough" to plead material falsity for a security claim where they relied on essentially the same corroboration from government actions against DeVry. Id. at *9. There, the court expressed doubt that the government allegations and confidential witness testimony (the same testimony plaintiffs seek to piggyback on now) were convincing corroboration of falsity but ultimately rested its conclusion that the plaintiff had done "just enough" on the fact that it "[made] the most diligent pre-complaint inquiry within its power and [stated] enough facts to 'raise a reasonable expectation that discovery will reveal' additional evidence of falsity, or alternatively, to permit defendants to quickly marshal evidence disproving plaintiff's allegations, to survive defendants' motion to dismiss for failure to allege material falsity." Id. This analysis focuses primarily on the first Pirelli prong—whether the facts of the fraud are accessible to plaintiffs—and does not explicitly address the second—whether the grounds for plaintiffs' suspicions make it plausible that the 90% representations were false. To the extent the court found that the government allegations alone were enough to plausibly allege the falsity of DeVry's 90% representations, I disagree.

The fact that plaintiffs are unable to access the DeVry records lowers the Rule 9(b) bar for their fraud allegations, but it does not eliminate the requirement that they articulate plausible grounds for their belief that DeVry's statements were false. See Pirelli, 631 F.3d at 443 ("The grounds for the plaintiff's suspicions must make the allegations plausible, even as courts remain sensitive to information asymmetries that may prevent a plaintiff from offering more detail." (emphasis removed)). If not based on knowledge, fraud claims must at least be based on something more than "reading a daily law bulletin," id. at 444, and "on information and belief" must mean something more than "rumor has it." U.S. ex rel. Bogina v. Medline Indus., Inc., 809 F.3d 365, 370 (7th Cir. 2016).

Because plaintiffs have not met Rule 9(b)'s requirements for pleading fraud, their statutory claims are dismissed.

C. Unjust Enrichment

Allegations supporting unjust enrichment claims must comply with Rule 9(b)'s requirements when they allege fraud. See Pirelli, 631 F.3d at 447. Plaintiffs' unjust enrichment claims are premised on the same misrepresentations as the statutory claims, see [55] ¶ 381, so the particularity deficiencies that defeated the statutory claims defeat the unjust enrichment claim too.

D. Leave to Amend

Plaintiffs request leave to amend, which should be freely granted when required by justice. Fed. R. Civ. P. 15(a)(2). However, I retain broad discretion to deny leave to amend where there has been a repeated failure to cure deficiencies. Right Field Rooftops, LLC v. Chicago Cubs Baseball Club, LLC, 870 F.3d 682, 693 (7th Cir. 2017). This is plaintiffs' second failure to state a claim, and their amendment did not correct the Rule 9(b) particularity deficiencies I identified in the original complaint. For that reason, plaintiffs' request for leave to amend is denied, and the dismissal is with prejudice.

IV. Conclusion

DeVry's motion to dismiss [59] is granted, and the complaint is dismissed with prejudice. The motion to strike class allegations [62] is terminated as moot. Enter judgment and terminate civil case. ENTER:

/s/_________

Manish S. Shah

United States District Judge Date: February 13, 2019