Shupe v. Rocket Companies, Inc.

Opinion

UNITED STATES DISTRICT COURT EASTERN DISTRICT OF MICHIGAN NORTHERN DIVISION

CARL SHUPE, individually and on behalf of all others similarly situated,

Plaintiffs, Case No. 1:21-cv-11528

v. Honorable Thomas L. Ludington United States District Judge ROCKET COMPANIES, INC., JAY FARNER, JULIE BOOTH, ROBERT WALTERS, DANIEL GILBERT, and ROCKET HOLDINGS, INC.,

Defendants. _______________________________________/

OPINION AND ORDER GRANTING IN PART PLAINTIFFS’ MOTION TO STRIKE AND GRANTING IN PART DEFENDANTS’ MOTION TO DISMISS

Plaintiffs have brought a class action against Rocket Companies and some of its officers and directors. Plaintiffs allege that, between February 25 and May 5, 2021, Defendants artificially inflated the price of Rocket Class A common stock by publicly misrepresenting numerous adverse facts, violating the Securities Exchange Act. Plaintiffs also allege that the Chief Executive Officer and controlling shareholder, Daniel Gilbert, traded securities using insider information. Defendants filed a motion to dismiss, but their reply brief incorporated an analyst report that Plaintiffs did not submit or discuss. So Plaintiffs filed a motion either to strike the report or to convert the motion to dismiss to a summary-judgment motion. The questions presented are whether Defendants’ motion to dismiss must be converted to a summary-judgment motion, whether Plaintiffs have adequately alleged that Defendants made material misrepresentations in violation of the Securities Exchange Act, and whether Plaintiffs have adequately alleged that Daniel Gilbert committed insider trading. I. BACKGROUND In this securities-fraud class action, numerous shareholders of Rocket stock (NYSE:RKT) have sued Daniel Gilbert and Rocket Holdings (RHI),1 Rocket Companies (RCI), and some of RCI’s senior officers and directors: Jay Farner,2 Julie Booth,3 and Robert Walters.4 ECF No. 42. A.

In June 2021, Zoya Qaiyum brought this case under 15 U.S.C. § 78j(b), t(a) and 17 C.F.R. § 240.10b-5. ECF No. 1. The next month, District Judge Paul D. Borman recused himself from the case, which was then randomly reassigned to District Judge Judith E. Levy. ECF No. 4. In August 2021, six plaintiffs from later-filed Case No. 5:21-CV-11618 filed five motions to consolidate that case with this case, as well as to be appointed as the lead or colead plaintiff under 15 U.S.C. § 78u–4(a)(3)(B)(i), as amended by the Private Securities Litigation Reform Act of 1995 (PSLRA). See ECF Nos. 10; 13; 14; 15; 16. In April 2022, Judge Levy consolidated the cases under Federal Rule of Civil Procedure 42(a)(2) but did not determine the lead plaintiff under the PSLRA. See ECF Nos. 30; 31. Two days

later, the case was randomly reassigned to the undersigned so that Judge Levy could “effectively manage the ongoing Flint water cases.” E.D. Mich. Admin. Order 22-AO-024 (effective Apr. 19, 2022); see also In re Flint Water Cases, No. 5:16-CV-10444 (E.D. Mich. Feb. 23, 2023).

1 At all relevant times, “Daniel Gilbert and RHI [we]re Rocket’s majority stockholders with direct or indirect ownership of 99.9% of Rocket’s outstanding Class D Common Stock and 93.1% of Rocket’s Class A Common Stock on a fully exchanged and converted basis.” ECF No. 44 at PageID.1209. “Mr. Gilbert founded Rocket Mortgage in 1985 and is Chairman of the Board of Rocket and RHI.” Id. 2 At all relevant times, “Jay Farner [wa]s CEO of Rocket and RHI and Vice Chairman of Rocket’s Board.” ECF No. 44 at PageID.1209. 3 At all relevant times, “Julie Booth [wa]s Rocket’s CFO and Treasurer.” ECF No. 44 at PageID.1209. 4 At all relevant times, “Robert Dean Walters [wa]s Rocket’s President and COO.” ECF No. 44 at PageID.1209. In May 2022, Qaiyum voluntarily dismissed her complaint and Carl Shupe was appointed as Lead Plaintiff of the Rocket Class. Shupe v. Rocket Cos., 601 F. Supp. 3d 214, 217 n.1, 221 (E.D. Mich. 2022). A month later, Plaintiffs filed the Second Amended Complaint, alleging that:  Gilbert and RHI violated 15 U.S.C. § 78j(b) and 17 C.F.R. § 240.10b-5 (Count I);  Gilbert and RHI violated 15 U.S.C. § 78t-1 (Count II);  RCI, Farner, Booth, Walters, and Gilbert violated 15 U.S.C. § 78j(b) and 17 C.F.R. § 240.10b-5 (Count III); and  RHI, Farner, Booth, Walters, and Gilbert violated 15 U.S.C. § 78t (Count IV).

ECF No. 42 at PageID.1165–74. Seventeen days later, Defendants filed a motion to dismiss, ECF No. 44, which has been fully briefed, ECF Nos. 49; 50. In response to the Motion to Dismiss, Plaintiffs filed a motion to strike Defendants’ reply or, alternatively, to deny the Motion to Dismiss as a premature motion for summary judgment because there has been no discovery in this case. ECF No. 51. As required in the review of a motion to dismiss, Plaintiffs’ factual allegations are assumed true and all reasonable inferences drawn from those factual allegations are in Plaintiffs’ favor. Waskul v. Washtenaw Cnty. Cmty. Mental Health, 979 F.3d 426, 440 (6th Cir. 2020). B. Rocket Mortgage is the largest mortgage lender in the United States. Rocket lends funds, secured by mortgages, to its customers. ECF No. 42 at PageID.1031. Its most profitable business is “repackaging and selling” the customers’ loans to government-sponsored enterprises (GSE), which sell the loans as mortgage-backed securities on a secondary market. Id. at PageID.1032. Rocket capitalizes on what the mortgage industry calls the “primary-secondary spread”: the difference between mortgage rates for borrowers (the primary rate) and the GSEs’ yields on mortgage-backed securities (the secondary rate). Andreas Fuster et al., The Rising Gap Between Primary and Secondary Mortgage Rates, 19 ECON. POL’Y REV. 27 (2013). According to Plaintiffs, this spread substantially reflects the premium that the GSEs will pay for Rocket’s loans in the future, after adjusting for other components like the guarantee fee5 and the servicing fee.6 ECF No. 42 at PageID.1101. Rocket uses this formula, Plaintiffs add, to approximate the profit margins of future loans. Id. Plaintiffs allege that, during the Class Period (February 25–May 5, 2021), Defendants

violated the Securities Exchange Act by making seven “false and misleading statements to the market about Rocket’s” key performance metrics. See id. at PageID.1031–32. Statements 1 and 2. On February 25, 2021, Rocket held an Earnings Call.7 Morgan Stanley Analyst James Faucette asked Defendant Farner to “help shed some light on what we should expect in terms of new applications and growing—continuing to grow the business—particularly in the rising rate environment. . . . And where should we be particularly sensitive?” Id. at PageID.1123– 24. Farner responded that Rocket is “seeing strong consumer demand, especially in the housing

5 A guarantee fee is an upfront fee, usually a percentage of the loan amount, that the lender charges to guarantee a portion of the loan in case the borrower defaults—often required for government-backed loans, like loans from the FHA (Federal Housing Administration) loans, VA (Department of Veterans Affairs) loans, and USDA (United States Department of Agriculture). David Reiss, Underwriting Sustainable Homeownership: The Federal Housing Administration and the Low Down Payment Loan, 50 GA. L. REV. 1019, 1042–43 (2016). 6 A servicing fee is a monthly fee, usually a percentage of the outstanding loan balance, that the mortgage servicer charges for collecting payments from the borrower, sending statements, managing the escrow account, and other administrative tasks related to servicing the loan. John R. Brooks & Adam J. Levitin, Redesigning Education Finance: How Student Loans Outgrew the “Debt” Paradigm, 109 GEO. L.J. 5, 52–53 (2020). 7 An earnings call is a quasi-public conference call held by a company’s management after releasing quarterly or annual financial results to discuss the financial performance and to answer questions from participants. Edward B. Rock & Daniel L. Rubinfeld, Common Ownership and Coordinated Effects, 83 ANTITRUST L.J. 201, 235–36 (2020) (“[C]onsumers are unlikely to pay attention to earnings calls, while competitors will.”); see also William H. Page, Tacit Agreement Under Section 1 of the Sherman Act, 81 ANTITRUST L.J. 593, 636 (2017) (“[E]arnings calls . . . provide an opportunity to make far more detailed statements about competitive strategy than a bare announcement of a future price increase.”). market. . . . the strongest that it’s been here in the last decade,” and that “overall, [Rocket was] able to grow volume twice as fast as the industry in 2020.” Id. at PageID.1124 (emphasis omitted). In a follow-up to Defendant Booth’s “very specific guidance for the first quarter” of 2021, a Morgan Stanley analyst asked: “How should we think about, particularly the evolution of gain on sale through the course of the year, at least directionally? And any color . . . about like what the

current environment is for selling of loans and how you’re anticipating that, roughly, to evolve over the year?” Id. at PageID.1124. Booth replied that Rocket is “still in a very strong demand environment” and is “still seeing strong closed loan volume. . . . in Q4 of $107 billion and then looking ahead into Q1, between $98 million and $103 billion,” which “would be the second biggest quarter in [Rocket’s] history.” Id. at PageID.1124–25 (emphasis omitted). Statement 3. Later in the same Earnings Call, Goldman Sachs Analyst Ryan Matthew Nash asked Farner to discuss the “competitive dynamics [Rocket is] seeing given the changing [interest] rate backdrop,” particularly “in the context of the direct-to-consumer and the partner segments.” Id. at PageID.1126. Farner responded that the rising interest rates will give Rocket “an opportunity

. . . to lean in to spend more money,” and that in “the retail or Direct-to-Consumer space,” Rocket “[does not] see interest rates going up or down, really having an impact on [its] business one way or the other.” Id. at PageID.1126–27 (emphasis omitted). Statement 4. On March 3, 2021, Morgan Stanley held a virtual conference. “[A] Morgan Stanley analyst asked Farner about the growth rate for Rocket’s business channels.” Id. at PageID.1129. Farner responded that “you can probably sense from [his] passion, they’re all growing.” Id. at PageID.1129–30 (emphasis omitted). Statement 5. On March 11, 2021, a representative of Fox Business asked Farner “whether it was ‘good for Rocket Companies’ that the effects of the pandemic were winding down.” Id. at PageID.1131–32. Farner responded “You know, we’re going to see interest rates tick up a little bit here,” which Rocket takes “as an opportunity to grow. . . . So, really, interest rates moving around are a great benefit to [Rocket].” Id. at PageID.1132 (emphasis omitted). The same is true, he added, “when they drop back down.” Id. Statements 6 and 7. On March 17, 2021, Farner “boasted on Twitter that [Rocket] has

increased its market share and that volume from third-party originators had increased.” Id. at PageID.1134. Specifically, he tweeted, “Volume is up significantly at [Rocket]! Thank you partners. #BrokersCan[.]” Id. Defendant Gilbert retweeted this post two days later.8 Id. C. Plaintiffs contend that, as RHI’s CEO and Board Chairman, Gilbert committed insider trading by selling Rocket shares with the permission of Rocket’s Board, illustrated by ten details: First. On March 3, 2021, 26 days before Gilbert sold his shares, Rocket CEO Farner publicly stated that Rocket had yet to sell any shares since going public “because [they] believe strongly in the future of what [they were] building.” Id. at PageID.1149 (emphasis omitted).

Second. On March 23, 2021, Rocket’s Board of Directors gave Gilbert a material, nonpublic report projecting that, during 2021, Rocket would lose (1) 0.31% of its profitability, (2) “almost a billion dollars of revenue,” (3) 80% of its refinancing volume, and (4) profit margin. See id. at PageID.1033–34 (emphasis omitted), 1113–14, 1152. Third. Gilbert “submitted an exchange notice” to make the trade one week after “the insider trading window has already closed.” Id. at PageID.1149–50.

8 The Securities and Exchange Commission has charged securities fraud based in part on a retweet. E.g., Compl., SEC v. Musk, No. 1:18-CV-8865 (S.D.N.Y. Sept. 27, 2018), ECF No. 1 at PageID.12; see also In re Tesla, Inc. Sec. Litig., No. 3:18-CV-04865, 2022 WL 1497559, at *14– 19 (N.D. Cal. Apr. 1, 2022) (discussing tweets that demonstrated material misstatements). Fourth. Although Rocket’s Insider Trading Policy prohibited the trade, it permitted the “closed trading window” to be reopened “at any time, as deemed appropriate by the General Counsel or other senior members of management.” Id. at PageID.1150. Fifth. Gilbert and RHI own “99.9% of Rocket’s outstanding Class D Common Stock and 93.1% of Rocket’s Class A Common Stock on a fully exchanged and converted basis.” ECF No.

44 at PageID.1209 (citations omitted). Sixth. Thus, Gilbert “had effective control over Rocket’s board, management, and policies,” which Rocket acknowledges in its Form 10-K filings with the SEC because Gilbert is “the controlling shareholder of Rocket.” ECF No. 42 at PageID.1150–51 (“We are controlled by . . . Dan Gilbert, whose interests may conflict with our interests and the interests of other stockholders.”). Seventh. Gilbert leveraged his control as controlling shareholder to change Rocket’s company policy “for the limited purpose of conducting this sale,” without the advice or consent of Rocket’s General Counsel. Id.

Eighth. Knowing that the value of Rocket stock would decline when Rocket’s decreasing profitability and revenue went public, Gilbert sold personal Rocket shares for the first time ever: 20.2 million shares for $500 million. Id. at PageID.1034, 1115, 1120–21. Ninth. Even with the extended trading window, however, Rocket company policy still prohibited Gilbert’s sale because he did it “while in possession of material nonpublic information.” Id. at PageID.1151, 1157–58 (“The prohibition against trading while aware of, or tipping of, material non-public information applies even during an open trading window.”). Tenth. Gilbert sold his shares for 33% more than the closing price from the week after the information about Rocket stock’s declining business was disclosed to the public. Id. at PageID.1149. D. In Rocket’s Q1 2021 disclosures, released on May 5, 2021, Defendants “shocked the

market” by revealing Rocket was projecting reduced closed-loan volume and its “lowest quarterly Gain on Sale margin in two years” for Q2 2021. Id. at PageID.1034, 1137–39, 1152–54. In an earnings call later that day, Defendant Booth publicly attributed the “shocking” news to “three previously undisclosed [economic] factors,” after which “the price of Rocket Class A common stock dropped” from $22.80–$16.48 in six days. Id. at PageID.1034–35. During those six days, the value of Rocket stock plummeted by 28%, causing “significant damages” to shareholders. Id. at PageID.1035, 1141. In sum, Rocket executives told the market that internal financial metrics were improving, then it changed company policy without the insight of general counsel, allowing the Chairman of

the Board to sell more than twenty million shares a mere six days after he learned information that later caused the market for Rocket shares to decline to a record low. Id. at PageID.1034, 1119. II. PLAINTIFFS’ MOTION TO STRIKE EXHIBIT OR TO CONVERT MOTION First, Plaintiffs’ Motion to Strike will be granted in part. A. Courts resolving motions to dismiss may consider “the allegations of the complaint” and any “document[s] referred to or attached to the pleadings” that are “integral to plaintiff’s claims.” Burns v. United States, 542 F. App’x 461, 466 (6th Cir. 2013) (per curiam) (unpublished) (citing Com. Money Ctr. v. Ill. Union Ins., 508 F.3d 327, 336 (6th Cir. 2007)). In securities-fraud cases, such integral documents may include “the full text of SEC filings, prospectus, analysts’ reports, and statements.” In re Credit Acceptance Corp. Sec. Litig., 50 F. Supp. 2d 662, 669 (E.D. Mich. 1999) (quoting In re Royal Appliance Sec. Litig., 64 F.3d 663 (6th Cir. 1995) (per curiam) (unpublished table decision)).9 And courts may consider “public records, matters of which a court may take judicial notice, and letter decisions of governmental agencies.” Jackson v. City of

Columbus, 194 F.3d 737, 745 (6th Cir. 1999) (citations omitted), abrogated on other grounds by Swierkiewicz v. Sorema N.A., 534 U.S. 506 (2002). If a motion to dismiss “support[s] its position on the merits” by asking the court to consider the substance of anything else, then the court must convert the motion to dismiss to a summary-judgment motion. Burns, 542 F. App’x at 466–67; FED. R. CIV. P. 12(d) (“If . . . matters outside the pleadings are presented to and not excluded by the court, the motion must be treated as one for summary judgment under Rule 56.”); accord Carter v. Stanton, 405 U.S. 669, 671 (1972) (per curiam); Rondigo, LLC v. Twp. of Richmond, 641 F.3d 673, 680–81 (6th Cir. 2011). This situation can arise, for example, when the defendant invites dismissal of the case based on a

factual question demonstrated by an exhibit that is not attached to the complaint.

9 Technically, “the Sixth Circuit has instructed that district courts may consider the full text of SEC filings, prospectus, analysts’ reports, and statements integral to the complaint, even if not attached to the complaint[,] without converting a motion to dismiss []to a motion for summary judgment.” In re Credit Acceptance Corp. Sec. Litig., 50 F. Supp. 2d 662, 669 (E.D. Mich. 1999) (emphasis added) (citation omitted). Defendants suggest that text permits consideration of analyst reports that are not integral to the complaint. But the phrase “integral to the complaint” applies to the entire list because it “is a straightforward, parallel construction that involves all nouns or verbs in a series.” ANTONIN SCALIA & BRYAN A. GARNER, READING LAW: THE INTERPRETATION OF LEGAL TEXTS § 19, at 147–51 (2012) (emphasis omitted) (explaining the series-qualifier canon); see also id. § 20, at 152–53 (2012) (“When the syntax involves something other than a parallel series of nouns or verbs, a prepositive or postpositive modifier normally applies only to the nearest reasonable referent.” (emphasis omitted)); Allen v. United States, 572 F. Supp. 3d 411, 422 (E.D. Mich. 2021) (quoting Facebook, Inc. v. Duguid, 141 S. Ct. 1163, 1170 (2021)). And Defendants have not provided any authority to the contrary: permitting courts to consider nonintegral SEC filings, nonintegral prospectus, nonintegral analysts’ reports, or nonintegral statements. In sum, in a securities-fraud case, unless converting the motion to dismiss to a summary-judgment motion, the court may only rely on (1) the complaint’s allegations and (2) evidence that is (a) integral to the complaint, (b) a public record, (c) judicially noticeable, or (d) a governmental agency’s letter decision. B.

Plaintiffs contend Defendants’ Reply relies on a Wedbush analyst report that is neither integral to the Second Amended Complaint nor judicially noticeable. See ECF No. 51 at PageID.1814–25. Plaintiffs therefore request that this Court either convert Defendants’ Motion to Dismiss to a summary-judgment motion or strike the Wedbush Report from the record. Defendants’ only response is that the Wedbush Report is “easily” judicially noticeable. See ECF No. 52 at PageID.1884–89. On its face, the decision to include the Wedbush Report appears to require conversion of the Motion to Dismiss. If a motion to dismiss presents “matters outside the pleadings” that are “not excluded by the court,” then “the motion must be treated as one for summary judgment under Rule

56.” FED. R. CIV. P. 12(d). 1. However, the Wedbush Report is not integral; the Second Amended Complaint neither attaches nor relies on the Wedbush Report. See generally ECF No. 42. Defendants suggest the contrary by noting Plaintiffs’ reliance on reports from two other “reputable analysts”: Barclays and RBC Capital. See ECF No. 52 at PageID.1886 (quoting ECF No. 49 at PageID.1450). Thus, Defendants assure, the Wedbush Report merely “filled in the picture.” ECF No. 52 at PageID.1886. But that view encourages the court to color outside the lines. True, “a defendant may attach” documents to a motion to dismiss if “a plaintiff references or quotes” them. See In re Omnicare, Inc. Sec. Litig., 769 F.3d 455, 466 (6th Cir. 2014). But Plaintiffs did neither.10 So the Wedbush Report merely creates questions of fact. See ECF No. 50 at PageID.1718 (“[W]hen all analysts are considered—not just the two Plaintiffs cherry-pick—analysts actually predicted a range of 3.15% to 3.92%.” (citations omitted)). As any competent counsel in a jury trial surely knows, “[t]he judge decides questions of law; the jury[ decides] questions of fact.” Sparf v. United States, 156 U.S. 51,

89 (1895) (citation omitted); see also ECF No. 42 at PageID.1174 (“Plaintiffs hereby demand a trial by jury.”). And Defendants have included numerous factual assertions regarding the Report.11 For example, Defendants acknowledge they attached the Wedbush Report “solely” to refute the merits of “Plaintiffs’ insider trading claims.” ECF No. 52 at PageID.1882–83; see also id. at PageID.1887 (arguing the Wedbush Report “contains [a] figure . . . lower than the projections of Plaintiffs’ cherry-picked analysts and Rocket’s internal projection”). But evidence that does not support Plaintiffs’ claims is not integral to the Second Amended Complaint. See Burns v. United States, 542 F. App’x 461, 466 (6th Cir. 2013) (unpublished) (“Second, the exhibit was submitted solely

10 The closes that Plaintiffs come to mentioning the Wedbush Report is their honest disclosure of Wedbush as one of “at least thirteen financial analysts from different institutions” that “was” following Rocket’s stock. See ECF No. 42 at PageID.1158 & n.81. 11 Defendants’ Motion to Dismiss contends there are no questions of fact on numerous issues. See, e.g., ECF No. 44 at PageID.1216 (“No reasonable investor would construe the word “strong” in those snippets as promising that Rocket’s closed loan volume would stay the same or increase.”); id. at PageID.1217 (“[N]o reasonable investor would have interpreted Plaintiffs’ snippets as stating that closed loan volume would match or grow from its 2020 historic highs, when, in fact, Rocket disclosed the opposite.”); id. at PageID.1218 (“Those express warnings refute Plaintiffs’ core theory that reasonable investors had the implausible misimpression that closed loan volume would not decline, particularly amid extreme volatility during a global pandemic.”); id. at PageID.1220 (“Mr. Farner did no such thing, and no reasonable investor could have been under that misimpression because Rocket disclosed the exact opposite . . . .”); id. at PageID.1225 (“Nor would a reasonable investor construe Mr. Farner’s words as Plaintiffs would like . . . .”). to evidence that the Federal Bureau of Prisons took final administrative action; the substance of the exhibit is not integral to, but is in fact the Defendant’s opposition to, Plaintiff’s Complaint.”). In other words, Defendants ask this Court to grant their motion to dismiss based on the substance of a relevant report that is not integral to Plaintiffs’ Second Amended Complaint. Defendants then argue that there is a genuine question of fact as to Plaintiffs’ claims based on that

report. See ECF No. 52 at PageID.1883 (“If Plaintiffs think that analyst perceptions are relevant to determining the materiality of Rocket’s internal projection, they cannot object when Defendants provide the Court with the full picture.” (citation omitted)). Therefore, the motion to dismiss seemingly must be converted to a summary-judgment motion unless the Wedbush Report is a public record or subject to judicial notice. Burns, 542 F. App’x at 466–67. 2. The Wedbush Report is not a public record or subject to judicial notice. “[C]ourts may . . . refuse to take judicial notice of [documents] if they are ‘subject to reasonable dispute,’ are not ‘generally known’ within the court’s jurisdiction, or have source material that can ‘reasonably be

questioned.’” Matthew N. Preston II, The Tweet Test: Attributing Presidential Intent to Agency Action, 10 BELMONT L. REV. 1, 13 (2022) (quoting FED. R. EVID. 201(b)(2))). The Wedbush Report is reasonably disputable. While purporting to be “based on sources that [the analysts] consider reliable,” the Report “reflect[s] [the analysts’] personal opinions,” and “its accuracy is not guaranteed.” ECF No. 50-4 at PageID.1741–42. The Report even concedes the possibility of bias. Id. at PageID.1742 (“Thus, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report.”). The possibly “biased” nature of the Wedbush Report “leads the Court to conclude that the facts set forth therein may be subject to reasonable dispute.” Beaver Cnty. Ret. Bd. v. LCA-Vision Inc., No. 1:07-CV-00750, 2009 WL 806714, at *7 (S.D. Ohio Mar. 25, 2009). And, as explained, Defendants acknowledge that the Report demonstrates a genuine question of fact when contrasted with the reports on which Plaintiffs’ Second Amended Complaint relies. See ECF No. 52 at PageID.1882–83; discussion supra Section II.B.1. District courts “commit[] reversible error by taking judicial notice of . . . . facts which can be and are questioned and challenged by Plaintiff.” Burns, 542 F. App’x at 467;

see also id. at 466 (holding that if the plaintiff “disputes the validity of the statements in the document relied upon by Defendant,” then it “similarly disqualifies the exhibit from consideration on a motion to dismiss” (citing Passa v. City of Columbus, 123 F. App’x 694, 698 (6th Cir. 2005) (unpublished))). And the Report was not “publicly available to reasonable investors at the time the Defendant made the allegedly fraudulent statements or omissions.” Id. True, the Report is dated before some of the allegedly fraudulent statements. Compare ECF No. 50-4 at PageID.1736 (dated February 26, 2021), with discussion supra Section I.B (discussing statements made from February 25–March 17, 2021). But the Report is not publicly available on the internet—including

Wedbush’s website. See, e.g., News & Views, WEDBUSH, https://www.wedbush.com/news- views/page/13/ [https://perma.cc/C9JY-CGJ9] (filtering from January 13–February 26, 2021); https://www.wedbush.com/category/research-insights/page/9/ [https://perma.cc/KRJ2-9JGP] (filtering from February 1–March 19, 2021); https://www.wedbush.com/category/press- releases/page/2/ [https://perma.cc/3X2T-PYQF] (filtering from October 28, 2020–October 5, 2021). Although the Report claims not to be “based on any privileged information,” ECF No. 50- 4 at PageID.1742, it is not public, as Wedbush reports must be purchased, see, e.g., Rocket Companies (RKT), ZACKS INV. RSCH., https://www.zacks.com/stock/research/RKT/brokerage- reports [https://perma.cc/NBB3-SEMB] (selling Wedbush reports for $25.00). Because the Wedbush Report is neither public nor subject to judicial notice, Defendants’ reliance on it would ordinarily require “conversion of the motion to one for summary judgment.” See Burns, 542 F. App’x at 465–66. But not in a securities-fraud case. 3. Here, the Motion to Dismiss should not be converted. Instead—like a swollen appendix—

the nonintegral attachment should be severed. As District Judge Nancy G. Edmunds reasoned, “in the context of securities litigation,” converting a motion to dismiss to a summary-judgment motion “would undermine the provisions of Congress’s Private Securities Litigation Reform Act.” In re Credit Acceptance Corp. Sec. Litig., 50 F. Supp. 2d 662, 670 (E.D. Mich. 1999). Under the PLSRA, “motions to dismiss serve the special function of weeding out insufficient complaints before the discovery process commences” by requiring “an automatic stay of discovery during the pendency of a motion to dismiss.” Id. (citing 15 U.S.C. § 78u-4(b)(3)(B)). Indeed, once a motion to dismiss is filed, the PSLRA requires not only “the motion of any party” to open discovery, but also the court’s finding “that particularized discovery is necessary to

preserve evidence or to prevent undue prejudice to that party.” 15 U.S.C. § 78-4(b)(3)(B) (2022). But the Parties have not even raised an eyebrow in that direction. For those reasons, the Wedbush Report will be stricken, and the merits of Defendants’ Motion to Dismiss will be addressed. III. DEFENDANTS’ MOTION TO DISMISS Plaintiffs allegedly suffered losses when the share price of Rocket stock (NYSE:RKT) fell from $22.80 to $16.48 between May 5 and May 11, 2021. ECF No. 42 at PageID.1156. Plaintiffs allege that RCI, RHI, Farner, Booth, Walters, and Gilbert committed securities fraud by publicly telling investors one thing about Rocket’s gain-on-sale margin and closed-loan volume—the key internal metrics for assessing profitability, revenue, and success—while having knowledge or acting inconsistent with those public statements. Plaintiffs base their claims on allegations of material misrepresentations or omissions in an earnings call, a virtual conference, an interview, an internal company report, a 10-K filing, and two tweets. See discussion supra Part I. Specifically, Plaintiffs allege that:

 Gilbert and RHI violated § 10(b) and Rule 10b-5 of the Exchange Act (Count I);  Gilbert and RHI violated § 20(a) of the Exchange Act (Count II);  RCI, Farner, Booth, Walters, and Gilbert violated § 10(b) and Rule 10b-5 of the Exchange Act (Count III); and  RHI and Farner, Booth, Walters, and Gilbert violated § 20(a) of the Exchange Act (Count IV).

ECF No. 42 at PageID.1165–74. Plaintiffs bring the case under Federal Rule of Civil Procedure 23, suggesting a putative class of “all persons and entities that purchased or otherwise acquired publicly traded Rocket Class A common stock during the period from February 25, 2021 through May 5, 2021.” Id. at PageID.1162. They seek compensatory damages. Id. at PageID.1174. Defendants have filed a motion to dismiss under Rules 12(b)(6) and 9(b). ECF No. 44. A. 1. Under Rule 12(b)(6), a pleading fails to state a claim if it does not contain allegations that support recovery under any recognizable legal theory. Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009). In considering a Rule 12(b)(6) motion, the court construes the complaint in the plaintiff’s favor and accepts its factual allegations as true. See Lambert v. Hartman, 517 F.3d 433, 439 (6th Cir. 2008). The plaintiff need not provide “detailed factual allegations” to survive dismissal, but the “obligation to provide the ‘grounds’ of his ‘entitle[ment] to relief’ requires more than labels and conclusions.” Bell Atl. Corp. v. Twombly, 550 U.S. 544, 555 (2007) (“[A] formulaic recitation of the elements of a cause of action will not do.”). Although the complaint “must contain sufficient factual matter, accepted as true, to state a claim to relief that is plausible on its face,” the court need not accept legal conclusions as true. Iqbal, 556 U.S. at 678–79 (quotations and citation omitted). Facial plausibility is established if

“the plaintiff pleads factual content that allows the court to draw the reasonable inference that the defendant is liable for the misconduct alleged.” Id. at 678; see also 16630 Southfield Ltd. v. Flagstar Bank, F.S.B., 727 F.3d 502, 503 (6th Cir. 2013) (“The plausibility of an inference depends on a host of considerations, including common sense and the strength of competing explanations for the defendant’s conduct.”). 2. Section 10(b) of the Securities Exchange Act of 1934 makes it unlawful for anyone to “use or [to] employ, in connection with the purchase or sale of any security . . . any manipulative or deceptive device or contrivance in contravention of such rules and regulations as the Commission

may prescribe as necessary or appropriate in the public interest for the protection of investors.” 15 U.S.C. § 78j(b). To enforce this statute, the SEC promulgated Rule 10b-5, making it “unlawful” to misstate or to omit any “material fact.” Matrixx Initiatives, Inc. v. Siracusano, 563 U.S. 27, 37 (2011) (quoting 17 C.F.R. § 240.10b-5(b)). “The scope of Rule 10b-5 is coextensive with the coverage of § 10(b).” SEC v. Zandford, 535 U.S. 813, 816 n.1 (2002) (first citing United States v. O’Hagan, 521 U.S. 642, 651 (1997); and then citing Ernst & Ernst v. Hochfelder, 425 U.S. 185, 214 (1976)). Based on the “text and purpose of § 10(b),” Supreme Court precedent permits an implied “private cause of action.” See Matrixx, 563 U.S. at 37. But § 10(b) imposes no “affirmative duty to disclose any and all material information.” Id. at 44; see also Basic Inc. v. Levinson, 485 U.S. 224, 239 n.17 (1988) (“Silence, absent a duty to disclose, is not misleading under Rule 10b-5.”). To plead a securities-fraud claim under § 10(b), the plaintiff must plausibly allege: (1) a material misrepresentation or omission by the defendant; (2) scienter; (3) a connection between the misrepresentation or omission and the purchase or sale of a security; (4) reliance upon the misrepresentation or omission; (5) economic loss; and (6) loss causation.

Matrixx, 563 U.S. at 37, 37–38 (quotation and citation omitted); see In re Omnicare, Inc. Sec. Litig., 769 F.3d 455, 469 (6th Cir. 2014). Misrepresentations, omissions, and scienter “have heightened pleading requirements.” City of Taylor Gen. Emps. Ret. Sys. v. Astec Indus., 29 F.4th 802, 810 (6th Cir. 2022). 3. “Federal Rule of Civil Procedure 9(b) and the PSLRA apply to the misstatement-or-omission element.” Id. (citing La. Sch. Emps.’ Ret. Sys. v. Ernst & Young, LLP, 622 F.3d 471, 478 (6th Cir. 2010)). Rule 9(b) requires a plaintiff to “state with particularity the circumstances constituting fraud or mistake.” FED. R. CIV. P. 9(b). To satisfy Rule 9(b), the Second Amended Complaint must: (1) specify the statements that the plaintiff contends were fraudulent, (2) identify the speaker, (3) state where and when the statements were made, and (4) explain why the statements were fraudulent.

Ernst & Young, 622 F.3d at 478 (quoting Frank v. Dana Corp., 547 F.3d 564, 569 (6th Cir. 2008)). And the PSLRA requires plaintiffs to “specify each statement alleged to have been misleading, the reason or reasons why the statement is misleading, and, if an allegation regarding the statement or omission is made on information and belief, the complaint shall state with particularity all facts on which that belief is formed.” 15 U.S.C. § 78u-4(b)(1). In sum, “when examining allegedly fraudulent statements, we ask: Who said the statement? What is the statement? Where did they say it? When did they say it? Why is it misleading?” City of Taylor, 29 F.4th at 810 (citations omitted). “If the complaint does not answer all of these

questions, it fails to sufficiently plead securities fraud.” Id. “A misrepresentation is an affirmative statement that is misleading or false.” Omnicare, 769 F.3d at 470. Misrepresentations can contain hard or soft information. Hard information— “historical information or other factual information that is objectively verifiable”—is actionable if “a plaintiff pleads facts showing that the statement concerned a material fact and that it was objectively false or misleading.” Id. By contrast, soft information “includes predictions and matters of opinion,” and “additionally” requires the plaintiff to “plead facts showing that the statement was ‘made with knowledge of its falsity.’” Id. First, the court only determines whether statements are actionable and, if so, whether they were material. Id. Then, during the scienter analysis, the court

determines whether the defendants made the statements with knowledge of their falsity. Id. at 471. Omissions are actionable in two circumstances: (1) if a company has an affirmative duty to disclose (e.g., when an insider makes a trade or a statute requires disclosure), or (2) if a company makes “an inaccurate, incomplete, or misleading prior disclosure” that it must then correct. Id. at 471 (citing City of Monroe Emps. Ret. Sys. v. Bridgestone Corp., 399 F.3d 651, 669 (6th Cir. 2005)). In the latter circumstance, the company has a duty to disclose hard information that “renders a prior disclosure objectively inaccurate, incomplete, or misleading.” Id. (citing Zaluski v. United Am. Healthcare Corp., 527 F.3d 564, 576 (6th Cir. 2008)). If “new information is soft, then a person or corporation has a duty to disclose it only if it is virtually as certain as hard facts and contradicts the prior statement.” Id. (cleaned up). At bottom, federal securities laws “require an actor to ‘provide complete and non- misleading information with respect to the subjects on which he undertakes to speak.’” Helwig v. Vencor, Inc., 251 F.3d 540, 561 (6th Cir. 2001) (en banc) (quoting Rubin v. Schottenstein, 143

F.3d 263, 268 (6th Cir. 1998)). Yet a misstatement or omission is only actionable if a reasonable investor would have viewed the information as “alter[ing] the total mix of information available” to the market. Basic Inc. v. Levinson, 485 U.S. 224, 232 (1988) (citation omitted). Thus, “vague, soft, puffing statements or obvious hyperbole” of “corporate optimism,” which can be either “forward looking” or generalized to the point that they are “not capable of objective verification,” may not form the basis of a § 10(b) claim. In re Ford Motor Co. Sec. Litig., 381 F.3d 563, 570 (6th Cir. 2004). [F]ederal courts everywhere have demonstrated a willingness to find immaterial as a matter of law certain kinds of rosy affirmation heard from corporate managers and numbingly familiar to the marketplace—loosely optimistic statements that are so vague, [and] so lacking in specificity, . . . that no reasonable investor could find them important to the total mix of information available.

Id. at 570–71 (quotations omitted). Similarly, if sufficient forward-looking cautionary language accompanies affirmative statements, then the “bespeaks caution” doctrine renders predictions of business results immaterial as a matter of law. See In re Donald J. Trump Casino Sec. Litig., 7 F.3d 357, 371–73 (3d Cir. 1993). That is, “forward-looking statements will not form the basis for a securities fraud claim if those statements did not affect the ‘total mix’ of information the document provided investors.” Id. at 371. This doctrine, “as an analytical matter, equally appli[es]” to both “affirmative misrepresentations and omissions concerning soft information.” Id. Applying the doctrine depends on the specific communication at issue and requires a case-by-case analysis. Id. B. Here, Plaintiffs allege four categories of misstatements or omissions. 1.

Consumer-Demand Statements. At Rocket’s 4Q 2020 Earnings Call on February 25, 2021, Jay Farner said: We’re seeing strong consumer demand, especially in the housing market. I mean, it’s the strongest that it’s been here in the last decade . . . . [W]e were able to grow volume twice as fast as the industry in 2020.

ECF No. 42 at PageID.1124. Later, Julie Booth said:

[W]e’re still in a very strong demand environment with the Fed buying 95% or so of all conforming mortgage production today. . . . The demand environment’s very strong. . . . [W]e’re still seeing strong closed loan volume. Our closed loan volume in Q4 of $107 billion and then looking ahead into Q1, between $98 million and $103 billion: with those numbers, that would be the second biggest quarter in our company’s history.

Id. at PageID.1124–25. Plaintiffs allege these soft statements “gave the false and misleading impression[s] that closed loan volume was going to stay the same or increase in 2021, and that consumers [we]re applying for loans at the same rate” while “all relevant metrics were on the decline.” Id. at PageID.1123–26. That is, Plaintiffs argue Defendants misled the market by saying closed-loan volume was strong instead of saying that closed-loan volume was not as strong as before. Defendants provide four arguments for why “[n]o reasonable investor would construe the word ‘strong’ in those snippets as promising that Rocket’s closed loan volume would stay the same or increase”: (1) Defendants “issued guidance that same day projecting that closed loan volume would decline in Q1 2021.” ECF No. 44 at PageID.1216–17. (2) “[C]haracterization of ‘the statistical facts already disclosed’ is not securities fraud.” ECF No. 44 at PageID.1217 (quoting Pittman v. Unum Grp., 861 F. App’x 51, 55 (6th Cir. 2021)). (3) “Rocket expressly referred investors to the risk warnings in its SEC filings at the beginning of the earnings call.” ECF No. 44 at PageID.1218 (citing ECF No. 44-5 at PageID.1296). (4) The “vague mention” of “strong performance” is a “rosy affirmation” that is “immaterial as a matter of law.” ECF No. 44 at PageID.1219 (quoting Pension Fund Grp. v. Tempur-Pedic Int’l, 614 F. App’x 237, 245, 247 (6th Cir. 2015).

ECF No. 44 at PageID.1216–19. For one thing, the consumer-demand statements are not puffery. To be puffery, a statement must be “so vague, so lacking in specificity, or so clearly constituting the opinions of the speaker, that no reasonable investor could find them important to the total mix of information available.” In re Ford Motor Co. Sec. Litig., Class Action, 381 F.3d 563, 571 (6th Cir. 2004) (citations omitted). True, using the word “strong” in reference to company “growth” and “competition” would likely be “numbingly familiar” puffery. E.g., Tempur-Pedic, 614 F. App’x at 245–46. The difference here is that Farner and Booth both used the word “strong” in reference to consumer demand and closed-loan volume—Rocket’s key internal metrics for profitability. Nor were the consumer-demand statements false. Farner truthfully said consumer demand was the strongest that it had been in the last decade and that Rocket grew volume twice as fast as the industry in 2020. See ECF Nos. 44 at PageID.1216–17; 49 at PageID.1450. Then Booth truthfully explained that closed-loan volume, though strong, was projected to decline from $107 billion to roughly $100 billion. Absent some material omission, truthful disclosures like these cannot establish claims for securities fraud as a matter of law—not even when paired with the word “strong.” See In re Envision Healthcare Corp. Sec. Litig., No. 3:17-CV-01112, 2019 WL 6168254, at *9 (M.D. Tenn. Nov. 19, 2019) (“[S]tatements of opinion are actionable if it was subjectively false (i.e. not honestly held) or omits material facts about the basis or reliability of the opinion.” (citing Omnicare, Inc. v. Laborers Dist. Council Constr. Indus. Pension Fund, 135 S. Ct. 1318, 1329 (2015))). For these reasons, the consumer-demand statements are not actionable.12 2.

Interest-Rate Statements. At Rocket’s 4Q 2020 Earnings Call on February 25, 2021, Goldman Sachs Analyst Ryan Matthew Nash asked Jay Farner to discuss the “competitive dynamics [Rocket is] seeing given the changing [interest] rate backdrop,” particularly “in the context of the direct-to-consumer and the partner segments.” ECF No. 42 at PageID.1126. Defendant Farner responded: And so, usually, as we see these interest rates tick up a bit, what we’re going to see is an opportunity for us to lean in to spend more money and now to talk about the retail or Direct-to-Consumer space, same situation here. . . . And as others tend to step away or back away from the space, this is where we can lean in, we can grab market share. . . . So, I guess you can tell we’re pretty excited about it and don’t see interest rates going up or down, really having an impact on our business one way or the other.

ECF No. 42 at PageID.1126–27. Similarly, during an interview on March 11, 2021, a representative of Fox Business asked Defendant Farner “whether it was ‘good for Rocket Companies’ that the effects of the pandemic were winding down.” Id. at PageID.1131–32. Defendant Farner responded “You know, we’re going to see interest rates tick up a little bit here,” which Rocket takes “as an opportunity to grow. . . . So, really, interest rates moving

12 Because the truthfulness of the statements is dispositive, this Court need not reach the Parties’ other arguments regarding the consumer-demand statements. E.g., In re SmarTalk Teleservices, Inc. Sec. Litig., 124 F. Supp. 2d 527, 549 n.11 (S.D. Ohio 2000). around are a great benefit to [Rocket].” Id. at PageID.1132 (emphasis omitted). The same is true, he added, “when they drop back down.” Id. Plaintiffs allege these statements “misleadingly conveyed that Rocket was positioned to ‘weather the storm’ of rising rates when in reality Rocket was particularly sensitive to rising rates due to the composition, and downward trend, of its loan volume.” ECF No. 49 at 1456. They add

that “Rocket’s primary and secondary spread, Gain on Sale margins, volume, and overall profitability were significantly impacted in a negative manner by increased interest rates.” ECF No. 42 at PageID.1127–28. “These statements were false,” Plaintiff conclude, because “(i) refinancing loans accounted for 90% of Rocket’s loan originations; (ii) rising rates were causing and would cause a decline in loan volume; and (iii) Rocket’s closed loan volume was in decline at the time of his statements.” ECF No. 49 at PageID.1456 (internal citations omitted). That is, Plaintiffs allege Defendants misstated the impact of increased interest rates on Rocket’s business. Defendants provide two arguments in response. First, Defendants respond that investors would have to be “nitwits” to think that the interest-rate statements were misleading because

“Rocket disclosed the exact opposite” in a prospectus six months earlier. See ECF Nos. 44 at PageID.1220 (first quoting ECF No. 44-2 at PageID.1247; and then quoting Basic Inc. v. Levinson, 485 U.S. 224, 234 (1988)); 49 at PageID.1456–57. Defendants’ truth-on-the-market defense is curious because it acknowledges the interest-rate statements were false. And the Sixth Circuit has “not applied the truth-on-the-market defense.” Ley v. Visteon Corp., 543 F.3d 801, 809 (6th Cir. 2008). Even if it had, truth on the market is an issue for the trier of fact. Wilkof v. Caraco Pharm. Lab’ys, No. 2:09-CV-12830, 2010 WL 4184465, at *4 (E.D. Mich. Oct. 21, 2010) (“[A] ‘truth on the market defense’ . . . ‘is intensely fact-specific and is rarely an appropriate basis for dismissing a § 10(b) complaint for failure to plead materiality.’” (quoting Ganino v. Citizens Utils., 228 F.3d 154, 167 (2d Cir. 2000))); accord In re CBL & Assocs. Props. Sec. Litig., No. 1:19-CV-00181, 2022 WL 1405415, at *15 (E.D. Tenn. May 3, 2022) (“[T]he extent to which the market for [Rocket]’s stock was, or was not, aware of the true risks associated with the [prospectus] is a fact-intensive issue and improper for consideration here at the pleading stage.”) (collecting cases).

In support of their truth-on-the-market defense, Defendants merely rely on contextomy from the Second Circuit to argue that “Plaintiffs cannot plead fraud based on a misimpression that is ‘contradicted by plain and prominently displayed language in the prospectus[].’” ECF No. 50 at PageID.1709 (quoting Olkey v. Hyperion 1999 Term Tr., 98 F.3d 2, 9 (2d Cir. 1996)). What Defendants leave out, however, is that the Olkey court said the prospectus contradicted the plaintiffs’ allegations about the prospectus—not that it contradicted a different statement made by a different speaker, which is the case here. That is, the Olkey complaint relied on the prospectus. Thus, Olkey is inapposite. Indeed, it would be odd to say that the Securities Act creates a cause of action for misleading statements unless the truth came first. For lies usurp truth. That is why “the

corrective information must be conveyed to the public ‘with a degree of intensity and credibility sufficient to counter-balance effectively any misleading information created by’ the alleged misstatements.” Ganino, 228 F.3d at 167 (quoting In re Apple Comput. Sec. Litig., 886 F.2d 1109, 1116 (9th Cir. 1989)). But Defendants have not provided any evidence that could cure the effect of the false interest-rate statements. In sum, a reasonable investor would have been misled by Farner’s interest-rate statements because they altered the total mix of information available then. Second, Defendants assert “Farner’s statement would be protected by the PSLRA’s safe harbor for forward-looking statements.” ECF No. 44 at PageID.1221. The PSLRA protects the interest-rate statements, Defendants say, because “Rocket disclosed that the ‘call includes forward-looking statements’ and ‘referred investors to the earnings release that [it] issued [that ]day as well as risks described in the filings with the SEC.’” ECF No. 44 at PageID.1221 (quoting ECF No. 44-5 at PageID.1296) (cleaned up)). True, some forward-looking statements are not actionable under the PSLRA’s safe harbor. See 15 U.S.C. § 78u-5. But “the safe harbor does not extend to ‘a statement of present or historical fact.’”

Dougherty v. Esperion Therapeutics, Inc., 905 F.3d 971, 983 (6th Cir. 2018) (quoting Miller v. Champion Enters., 346 F.3d 660, 678 (6th Cir. 2003)). And the interest-rate statement is not forward looking. “Rather, it is a backward-looking statement concerning a future event,” because “[w]hen [Farner] spoke to [Nash], it was objectively discernable whether [Farner] had [known that changing interest rates would impact Rocket’s business]—either []he had, or []he had not.” Id. Even if the interest-rate statement is forward looking, “[n]o degree of cautionary language will protect material misrepresentations or omissions where defendants knew their statements were false when made.” In re Nat’l Century Fin. Enters., Inv. Litig., 541 F. Supp. 2d 986, 1005 (S.D. Ohio 2007) (quoting Milman v. Box Hill Sys. Corp., 72 F. Supp. 2d 220, 231 (S.D.N.Y. 1999));

Miller v. Champion Enters., 346 F.3d 660, 677 (6th Cir. 2003) (“[T]o be protected by the safe harbor provisions of the PSLRA, these statements must also have been accompanied by meaningful cautionary language.”). Here, not only have Defendants acknowledged that the interest-rate statement is false, ECF No. 44 at PageID.1220, but also Plaintiffs have alleged Farner knew it was false when he said it, ECF No. 49 at PageID.1459. For these reasons, Plaintiff Farner’s interest-rate statement is actionable. 3. Volume-by-Channel Statements. At a Morgan Stanley virtual conference on March 3, 2021, “a Morgan Stanley analyst asked Farner about the growth rate for Rocket’s business channels”: Can you talk a little bit about what portion each of those types of [partner] channels or processes are of the overall volume and business today? What their relative growth rates are? And how—what you think that looks like in the very long run?

ECF No. 42 at PageID.1129. Defendant Farner responded: So I know we don’t break down the percentages, and I’m probably not going to go any further than what we’ve already laid out in our earnings call, but as you can probably sense from my passion, they’re all growing. And with what about less than 10% market share, wherever we are, it’s hard to say today. If you think about all those different channels that can grow and give us reach, that’s why we get excited about what this company looks like in the years to come.

Id. at PageID.1129–30 (emphasis omitted). Plaintiffs allege these “statements were false and misleading when made because, contrary to Defendant Farner’s answer that Rocket’s various channels are ‘all growing,’ loan volume in each of Rocket’s channels was not growing but was actually in decline from its peak in November 2020.” Id. at PageID.1130. Plaintiffs add that the volume-by-channel statements “gave investors the false impression that growth would continue to occur in the near future, when all relevant metrics pointed to a decline in” refinancing-loan volume, closed-loan volume, and gain-on-sale margin. Id. at PageID.1131. That is, Plaintiffs allege Farner deceptively misstated that all Rocket’s business channels were growing while some key metrics were declining. Defendants first argue “no reasonable investor would have construed Mr. Farner’s comments about growth ‘in the very long run’ as remarking on growth of closed loan volume in the ‘near future.’” ECF No. 44 at PageID.1223. But that argument conflates two distinct points from the volume-by-channel statements. The analyst first asked Farner about the “overall volume and business today” of “each of” Rocket’s channels, followed by “[w]hat their relative growth rates are”—a present-tense question. Farner responded with a present-tense answer: “as you can probably sense from my passion, they[ a]re all growing.” Then the analyst separately asked “what [Farner] thinks that looks like in the very long run,” to which Farner responded “we get excited about what this company looks like in the years to come”—both future tenses. As indicated, the distinct future-tense question came after the present-tense question. So too did Farner’s responses. As explained above, “the safe harbor does not extend to ‘a statement of present or historical fact.’” Dougherty v. Esperion Therapeutics, Inc., 905 F.3d 971, 983 (6th Cir. 2018) (quoting Miller v. Champion Enters., 346 F.3d 660, 678 (6th

Cir. 2003)). Then Defendants mix in another truth-on-the-market defense that “Rocket disclosed that it projected closed loan volume to decrease in Q1 2021.” ECF No. 50 at PageID.1712. But there was no “contemporaneous” release of information that “informed the market” that closed-loan volume was declining. See Raab v. Gen. Physics Corp., 4 F.3d 286, 289 (4th Cir. 1993). As explained above, that information was disclosed six months earlier. See discussion supra Section III.C.2. So a reasonable investor would have been misled by Farner’s volume-by-channel statements because they altered the total mix of information available then. Defendants also suggest the volume-by-channel statements are unactionably “vague.” ECF

No. 44 at PageID.1223. Not so. Those statements specifically responded to an inquiry about “each of” Rocket’s partner channels, which includes closed-loan volume. And, as Plaintiffs allege, the present-tense volume-by-channel statement was objectively false; closed-loan volume was declining—not growing—when Farner said it was growing. ECF No. 42 at PageID.1130–31. One might ask, “How is each channel growing if they are collectively declining?” This inconsistency is precisely why Farner’s volume-by-channel statement is actionable. For it altered the total mix of information that a reasonable investor would have used when deciding to invest in Rocket stock. In re Omnicare, Inc. Sec. Litig., 769 F.3d 455, 470 (6th Cir. 2014) (holding that “‘typically historical information or other factual information that is objectively verifiable’ . . . is actionable if a plaintiff pleads facts showing that the statement concerned a material fact and that it was objectively false or misleading” (citations omitted)). Defendants also assert Farner did not have to disclose the truth about declining closed-loan volume because he “did not know what question was coming, had to answer off the cuff, and did

not have an opportunity to . . . edit his answer.” ECF No. 44 at PageID.1224 (quoting Plumbers Loc. Union 719 v. Zimmer Holdings, Inc., 679 F.3d 952, 956 (7th Cir. 2012)). But Farner’s volume-by-channel statement was not merely “evasive,” like the statement in Plumbers. See id. Here, as Plaintiffs allege, Farner’s statements were objectively false. Indeed, Farner could have refused to answer, which might have been “short of fraudulent.” Id. (citing Bronston v. United States, 409 U.S. 352 (1973)). But Farner went the extra mile, explaining, “[W]e don’t break down the percentages [of each partner channel], . . . but . . . they’re all growing.” ECF No. 42 at PageID.1129–30. “Thus, rather than resolve the factual dispute over whether plaintiff should have understood the word[s] ‘[they’re all growing]’ to carry a special meaning in this

context, th[is] Court must assume, as the complaint asserts, that [they] carr[y] [their] ordinary meaning.” City of Warren Police & Fire Ret. Sys. v. World Wrestling Ent., 477 F. Supp. 3d 123, 130 (S.D.N.Y. 2020). In sum, “a reasonable investor could have interpreted defendants’ discussions of [each partner channel] to indicate [that closed-loan volume was growing].” Id. For these reasons, Plaintiffs Farner’s volume-by-channel statements are actionable. 4. Broker Tweets. On March 4, 2021, United Wholesale Mortgage (UWM), Rocket’s primary competitor, threatened huge penalties against brokers who continued to work with Rocket, thus, forcing brokers to choose between UWM or Rocket. JC Reindl, UWM CEO Issues Ultimatum, Escalates War Against Rocket Companies, DET. FREE PRESS (Mar. 4, 2021, 3:12 PM), https://www.freep.com/story/money/business/2021/03/04/united-wholesale-mortgage- ultimatumrocket-companies/4578357001/ [https://perma.cc/HX9N-V6LE]. On March 17, 2021, Defendant Farner tweeted: Great News: Rocket Pro TPO has increased its market share in the weeks since UWM’s ultimatum to brokers. More than 8,000 independent brokers have rejected becoming a controlled UWM branch. The results? Volume is up significantly at @rocketprotpo! Thank you partners. #BrokersCan

Jay Farner (@JDFarner) TWITTER (Mar. 17, 2021, 12:22 PM), https://twitter.com/JDFarner/status/1372221904694697991 [https://perma.cc/24FR-9KTQ].13 Two days later, Gilbert retweeted Farner’s tweet, making Gilbert a “speaker,” ECF No. 42 at PageID.1134, which Defendants concede, see ECF No. 44 at PageID.1233 (acknowledging Gilbert made “only one statement”). As Plaintiffs put it, these tweets were misleading because “Gilbert and Farner boasted on Twitter that [Rocket] had increased its market share and that volume from third-party originators had increased, despite knowing that the pipeline was slowing.” ECF No. 42 at PageID.1134. Defendants provide two arguments in response. First, they assert that “Plaintiffs do not plead that Mr. Farner’s March 17, 2021 tweet was false when posted.” ECF No. 44 at PageID.1226 (emphasis omitted). Then Defendants add that the tweets refer to “a specific, narrower period: March 4 to 17, 2021.” ECF No. 50 at PageID.1712. But the tweets were objectively false and misleading when made. As Plaintiffs allege, Rocket’s broker market share and broker loan volume decreased in both February and March of

13 See Matthew N. Preston II, The Tweet Test: Attributing Presidential Intent to Agency Action, 10 BELMONT L. REV. 1, 13 (2022) (explaining that courts may take judicial notice of verified tweets) (citing FED. R. EVID. 201(b)(2)). 2021—when the tweets were posted—and continued to decrease in April 2021. ECF No. 42 at PageID.1084–85, 1097–98. And Defendants acknowledge that the tweets are “statements” under the Securities Exchange Act. See ECF No. 44 at PageID.1233 (“Plaintiffs attribute only one statement each to Mr. Gilbert and Ms. Booth.” (citations omitted)); see also Blessing v. Chandrasekhar, 988 F.3d 889, 902 n.12 (6th Cir. 2021) (accepting concession of plaintiffs’

counsel “that tweets should not be treated any differently than letters”); cf. James Madison Project v. DOJ, 302 F. Supp. 3d 12, 24 (D.D.C. 2018) (“‘[A] [tweet] by any other name would smell as sweet’ as any other official statement, at least for purposes of the official acknowledgement doctrine.” (quoting WILLIAM SHAKESPEARE, ROMEO AND JULIET act 2, sc. 2 (second alteration in original))). Because the tweets were false on their face, they are actionable. In re Omnicare, Inc. Sec. Litig., 769 F.3d 455, 469–70 (6th Cir. 2014) (statements of “historical information . . . that [are] objectively verifiable” are actionable). And the issue of whether the values declined for three months but not two weeks in March is an issue for the summary-judgment stage—notably because

Defendants have not included any evidence to corroborate that claim. In sum, Plaintiffs’ pleadings are “sufficient to show that [Rocket’s market share and volume] were adversely affected within that short window.” Williams v. Globus Med., Inc., 869 F.3d 235, 242 (3d Cir. 2017). For these reasons, the broker tweets are actionable. 5. In sum, the following statements are actionable: Farner’s interest-rate statements, Farner’s volume—by-channel statements, and Farner’s and Gilbert’s broker tweets. Notably, no statements were made by Defendants Robert Walters. And Julie Booth’s only statement is not actionable. See discussion supra Section III.C.1. Because Defendants Walters and Booth “did not ‘make’ any of the [challenged] statements,” they will be dismissed from the case with prejudice. Janus Cap. Grp. v. First Derivative Traders, 564 U.S. 135, 142, 146–47 (2011) (“[T]he maker of a statement is the person or entity with ultimate authority over the statement, including its content and whether and how to communicate it.”). Consequently, the remaining elements will be analyzed with respect to only Defendants

Jay Farner (imputing RCI) and Daniel Gilbert (imputing RHI). C. Next, Defendants argue that Plaintiffs have not plausibly alleged loss causation. ECF No. 44 at PageID.1232–33. 1. The PSLRA requires Plaintiffs to plead “loss causation”: that an alleged misstatement “caused the loss for which the[y] seek[] to recover damages.” 15 U.S.C. § 78u-4(b)(4). “Loss causation requires ‘a causal connection between the material misrepresentation and the loss.’” Brown v. Earthboard Sports USA, Inc., 481 F.3d 901, 920 (6th Cir. 2007) (quoting Dura

Pharms. v. Broudo, 544 U.S. 336, 342 (2005)). “As pleading requirements go, this one is ‘not meant to impose a great burden upon a plaintiff.’” Norfolk Cnty. Ret. Sys. v. Cmty. Health Sys., Inc., 877 F.3d 687, 695 (6th Cir. 2017) (quoting Dura, 544 U.S. at 347). “Rather it is meant to prevent disappointed shareholders from filing suit merely because their shares have lost value and then using discovery to determine whether the loss was due to fraud.” Id. (citing Dura, 544 U.S. at 347–48). Loss causation “has been likened to proximate cause in tort law.” Earthboard Sports, 481 F.3d at 920 (citing AUSA Life Ins. v. Ernst & Young, 206 F.3d 202, 213 (2d Cir. 2000)). At the pleading stage, a plaintiff need only to “provide a defendant with some indication of the loss and the causal connection that the plaintiff has in mind.” Norfolk Cnty. 877 F.3d at 347. 2. Plaintiffs’ loss causation can be demonstrated with a plug-and-play excerpt of a similar theory that the Sixth Circuit found sufficient:

[Plaintiffs]’ claims can be summarized as follows: [Defendants] made materially false statements and omissions regarding [loan volume, gain-on-sale margin, and market share]. When these risks were realized, the market price of [Rocket] stock dropped dramatically. Had [Plaintiffs] known the truth, [they] would not have invested in [Rocket] common stock, or would have done so at a much lower purchase price. [Plaintiffs] conclude[] that the 2[8]% loss in stock price that occurred on [May 5, 2021], when [Rocket] disclosed [declining loan volume, gain-on-sale margin, and market share], is “directly attributable to the market’s reaction to revelations of the nature, extent, and impact of the fraud at [Rocket].”

Ohio Pub. Emps. Ret. Sys. v. Fed. Home Loan Mortg. Corp., 830 F.3d 376, 388 (6th Cir. 2016) (internal citations omitted); accord Norfolk Cnty., 877 F.3d at 695 (“Plainly the loss that [Plaintiffs] had in mind is the value that their [Rocket] shares lost when the market realized that [Rocket]’s [loan volume and gain-on-sale margin] were padded with improper [data].”). Simply put, Plaintiffs allege that they would have suffered fewer to no losses if Jay Farner and Daniel Gilbert did not make material misrepresentations—or better yet had told the truth— about Rocket’s key performance indicators (KPI): declining loan volume, gain-on-sale margin, and market share. See ECF Nos. 42 at PageID. 1137–39, 1170; 1153–56; 49 at PageID.1465, 1473– 74. As they correctly note, “Nothing more is needed from Plaintiffs at the pleading stage.” ECF No. 49 at PageID.1474 (citing Ohio Pub. Emps., 830 F.3d at 388). Yet Defendants assert the alleged misstatements do not “relate to” the Q2 2021 forecasts that caused the stock price to decline, so Rocket’s Q2 2021 forecasts did not cause Plaintiffs’ losses. See ECF Nos. 44 at PageID.1232–33; 50 at PageID.1716. But that argument has no merit. If Defendants are correct in assuming that the Q2 2021 forecasts ignited the declining stock price, then their earlier misrepresentations were the kindling and Plaintiffs’ Rocket stock were logs that burned to ash. That is, after Defendants publicly stated KPIs were increasing despite knowing otherwise, they truthfully disclosed that the same KPIs were declining in Q2 2021 forecasts, which caused the stock price to fall. Under this version of

the events, it would be reasonable to infer that Defendants’ misstatements slowed or stopped the declining stock price between the misstatements and the Q2 2021 forecasts, which was when Daniel Gilbert sold $500 million of Rocket shares. Thus, even viewing the facts as Defendants frame them, the misrepresentations and the Q2 2021 forecasts were contributing causes of Plaintiffs’ losses. Chamberlain v. Reddy Ice Holdings, Inc., 757 F. Supp. 2d 683, 718 (E.D. Mich. 2010) (“‘[A]s long as the misrepresentation is one substantial cause of the investment’s decline in value, other contributing forces will not bar recovery under the loss causation requirement’ but will play a role ‘in determining recoverable damages.’” (quoting In re Daou Sys., Inc., 411 F.3d 1006, 1025 (9th Cir. 2005) (alteration in

original)). Yet they maintain that their misrepresentations did not cause Plaintiffs’ losses. Defendants’ proposed regime would absolve liability for misrepresentations that are not followed by a correct disclosure that drops the market value of the stock. But that reading of the Securities Exchange Act would create a perverse incentive for corporations not to correct their public misrepresentations. Of course, a corrective disclosure would likely cause stock prices to fall. But that is not the only way that investors could learn of material misrepresentations. See, e.g., Ohio Pub. Emps., 830 F.3d at 385 (permitting plaintiffs to plead loss causation based on “materialization of the risk” because “defendants accused of securities fraud should not escape liability by simply avoiding a corrective disclosure” (citing Mass. Ret. Sys. v. CVS Caremark Corp., 716 F.3d 229, 240 (1st Cir. 2013))); Robert N. Rapp, Plausible Cause: Exploring the Limits of Loss Causation in Pleading and Proving Market Fraud Claims Under Securities Exchange Act §10(b) and SEC Rule 10b-5, 41 OHIO N.U. L. REV. 389, 428 (2015) (“Revelation of the truth for purposes of loss causation in market fraud cases need not happen by way of a single event or disclosure.” (citing Pub. Emps. Ret. Sys. of Miss. v. Amedisys, Inc., 769 F.3d 313, 320–22 (5th Cir.

2014))). At this stage, Plaintiffs need not allege “the most likely” theory, only a “plausible” one. Id. at 388 (citing Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007)). Thus, Plaintiffs would succeed on loss causation by alleging “that the misstatement or omission concealed something from the market that, when disclosed, negatively affected the value of the security.” Stratte- McClure v. Morgan Stanley, 598 F. App’x 25, 29 (2d Cir. 2015) (unpublished) (citing Lentell v. Merrill Lynch & Co., 396 F.3d 161, 173 (2d Cir. 2005)). As explained above, Plaintiffs allege Defendants misstated or omitted Rocket’s declining loan volume, gain-on-sale margin, and market share. See also ECF No. 49 at PageID.1474 (“On May 5, 2021, the market was informed that

Rocket’s business would be performing much worse than previously expected, causing a 17% decrease in share price in a single day.” (citing ECF No. 42 at PageID.1152–56)). For these reasons, Plaintiffs have adequately pled loss causation. D. The final issue concerning the § 10b-5 claims is whether Plaintiffs have pled a strong inference of scienter. 1. To survive a motion to dismiss, Plaintiffs must “state with particularity facts giving rise to a strong inference that the defendant acted with the required state of mind” when making each alleged misrepresentation or omission. 15 U.S.C. § 78u-4(b)(2)(A), (3)(A); see also Tellabs, Inc. v. Makor Issues & Rts., Ltd., 551 U.S. 308, 312 (2007); Doshi v. Gen. Cable Corp., 823 F.3d 1032, 1039 (6th Cir. 2016). “In securities-fraud cases, scienter [is] a ‘knowing and deliberate intent to manipulate, deceive, or defraud’ or ‘recklessness.’” City of Taylor Gen. Emps. Ret. Sys. v. Astec Indus., 29

F.4th 802, 812 (6th Cir. 2022) (quoting Doshi, 823 F.3d at 1039). Recklessness is “akin to conscious disregard,” and the presence of “multiple, obvious red flags” demonstrating an “egregious refusal to see the obvious or to investigate the doubtful” is sufficient. Id. (quoting Doshi, 823 F.3d at 1039). Because Congress enacted the PSLRA to place an “elephant-sized boulder” before private securities-fraud cases, In re Omnicare, Inc. Sec. Litig., 769 F.3d 455, 461 (6th Cir. 2014), deciding whether a plaintiff adequately pleaded a strong inference of scienter requires a three-part test. Dougherty v. Esperion Therapeutics, Inc., 905 F.3d 971, 979 (6th Cir. 2018) (citing Tellabs, 551 U.S. at 322–23). “First, [this Court] must accept all factual allegations in the complaint as true.”

Tellabs, 551 U.S. at 322. Second, this Court must review the allegations holistically “to determine ‘whether all of the facts alleged, taken collectively, give rise to a strong inference of scienter.” Dougherty, 905 F.3d at 979 (quoting Tellabs, 551 U.S. at 322–23). And then this Court “‘must take into account plausible opposing inferences’ and decide whether ‘a reasonable person would deem the inference of scienter cogent and at least as compelling as any opposing inference one could draw from the facts alleged.’” Id. (quoting Tellabs, 551 U.S. at 323–24). As part of that analysis, this Court must consult a nonexhaustive list of considerations known as the Helwig factors.14 Doshi, 823 F.3d at 1039. Here, the relevant Helwig factors include

14 The nine Helwig factors are: “insider trading at a suspicious time or in an unusual amount,” “divergence between internal reports and external statements on the same subject,” “closeness in time of an allegedly fraudulent statement or omission and the later disclosure of inconsistent information,” and “disregard of the most current factual information before making statements,” “the personal interest of certain directors in not informing disinterested directors of an impending sale of stock,” and “the

self-interested motivation of defendants in the form of saving their salaries or jobs.” Doshi, 823 F.3d at 1039–40 (citing Helwig v. Vencor, Inc., 251 F.3d 540, 552 (6th Cir. 2001) (en banc), abrogated on other grounds by Tellabs, 551 U.S. at 314)). 2. a. First, during the Class Period, Defendants acknowledged that they “monitored a number of key performance indicators [KPI] to evaluate the performance of [Rocket’s] business operations”: identifying (1) gain-on-sale margin, (2) closed-loan volume, and (3) market share. ECF No. 42 at

(1) insider trading at a suspicious time or in an unusual amount; (2) divergence between internal reports and external statements on the same subject; (3) closeness in time of an allegedly fraudulent statement or omission and the later disclosure of inconsistent information; (4) evidence of bribery by a top company official; (5) existence of an ancillary lawsuit charging fraud by a company and the company’s quick settlement of that suit; (6) disregard of the most current factual information before making statements; (7) disclosure of accounting information in such a way that its negative implications could only be understood by someone with a high degree of sophistication; (8) the personal interest of certain directors in not informing disinterested directors of an impending sale of stock; and (9) the self-interested motivation of defendants in the form of saving their salaries or jobs. Doshi v. Gen. Cable Corp., 823 F.3d 1032, 1039–40 (6th Cir. 2016) (quoting Helwig v. Vencor, Inc., 251 F.3d 540, 552 (6th Cir. 2001) (en banc)). PageID.1054–62, 1143. Defendants also stated that these three KPIs “enable [Rocket] to monitor [its] ability to generate gain on sale revenue,” demonstrating the significance. Id. at PageID.1062. Such self-admitted monitoring infers scienter. See Winslow v. BancorpSouth, Inc., No. 3:10-CV- 00463, 2011 WL 7090820, at *23 (M.D. Tenn. Apr. 26, 2011) (“[S]pecific admissions from top executives that they . . . monitored portions of the company’s database are factors in favor of

inferring scienter.” (quoting In re Daou Sys. Inc., 411 F.3d 1006, 1022 (9th Cir. 2005))). Plus, Confidential Witness 1 (CW-1) corroborates that Defendants monitored the KPIs, stating that Rocket management tracked gain-on-sale margin “daily” if not “at least four times a week,” that Rocket set “monthly” forecasts of gain-on-sale margin that would have been approved by Rocket’s executives, and that the loan pipeline was discussed three to four times per week in meetings with the Chief Revenue Officer. ECF No. 42 at PageID.1065, 1078–79. b. Second, the Second Amended Complaint demonstrates the “divergence between internal reports and external statements on the same subject.” City of Taylor, 29 F.4th at 813. “Securities

fraud claims typically have sufficed to state a claim based on recklessness when they have specifically alleged defendants’ knowledge or access to information contradicting their public statements.” Chamberlain v. Reddy Ice Holdings, Inc., 757 F. Supp. 2d 683, 711 (E.D. Mich. 2010). Here, Defendants had access to: (1) “dashboards” that demonstrated declining closed-loan volume, gain-on-sale margin, and other KPIs that tracked the “life cycle” of a loan all the way to “closed folder status”—which were each expounded by retail and broker-business segments, ECF No. 42 at PageID.1065–67 (indicating that the dashboard metrics pulled data from Rocket’s data lake through a program called “Sequel”); (2) the “BIG platform,” which was a compilation of all Rocket’s various dashboards, ECF No. 42 at PageID.1067 (statement of CW-2); (3) information on mortgage loans generated through Rocket’s “Core Digital Media” subsidiary and RocketMortgage.com website visits, ECF No. 42 at PageID.1069–71; and (4) machine-learning “models” and “predictive analytics” that allowed Rocket management to track—in “realtime”—“every step” of the loan pipeline, thus, permitting Rocket to know its total loan volume and to predict future loan volume, see ECF No. 42 at PageID.1071–80 (statements of CW-3 and CW-4).

And, given Rocket’s acknowledged 45-to-60-day lag between initiating and closing loan applications, see ECF No. 42 at PageID.1080, Defendants would also have insight into future loan volume based on current application volume, see ECF No. 42 at PageID.1080–83. Indeed, Defendants acknowledge that they had “real-time insights” into Rocket’s loan “application volume” based on Rocket’s “data lake” of information, which encompassed 85% of the United States’s adult population; that they “leverage[d] this information”; and that this information “consistently guide[d] [them] to the right decisions.” ECF No. 42 at PageID.1071–72. As Farner emphasized, “data science has driven more than $75 billion in application volume. The importance of data cannot be overstated. It is this information that allows [Rocket] to quickly innovate, anticipate the market and, in some cases, develop new companies all together.” Id. At the Earnings Call on February 25, 2021, Farner acknowledged that Defendants reviewed and used those data points to evaluate Rocket’s present state and to forecast future growth and demand through management-level loan-volume analysis. See ECF No. 42 at PageID.1081–82 (“I’m going to go back to the importance of the data from the top of the funnel, even pre-application to understand pull-through rates.”). And CW-3 corroborates these allegations by demonstrating that Rocket had the capability through “predictive analytics” to forecast closed-loan volume and gain-on-sale margin. ECF No. 42 at PageID.1075–78. Other CWs also corroborate Defendants’ access and review of the data. For example, CW- 4 stated that “every stage of the loan approval process” could be tracked—including the initial application, the registration, and the updated status dates as it moved through the approval process—and that the “top people” at Rocket could access the entire pipeline using Salesforce. Id. at PageID.1080. And CW-1 explained that the Senior VPs had access to the “dashboards” and believed that the dashboards were available to all Director-level managers and above, including the Defendants. See id. at PageID.1065.15 In response, Defendants invoke Konkol v. Diebold, Inc., 590 F.3d 390, 397 (6th Cir. 2009)

to argue that their generalized “access to . . . financial information” does not support a strong inference of scienter. ECF No. 44 at PageID.1229. But that argument misunderstands Konkol. There, the plaintiffs failed to meet their burden because—unlike here—the complaint lacked sufficient “facts regarding the financial reports, how they were used, and their connection to the Defendants.” Konkol, 590 F.3d at 397. Here, as explained above, the Second Amended Complaint details the “reports” alleged, “what the software reflected [i.e., the dashboards, BIG platform, Salesforce, predictive analytics, etc.,] and what would have been obvious to a reasonable person upon examining the software.” Id. at 398; see ECF No. 42 at PageID.1064–76.

15 Defendants assert that CW-1, CW-2, CW-3, and CW-4 were “several levels removed” from the executive team and, therefore, their allegations should be discredited. See ECF No. 44 at PageID.1229. On the contrary, “plaintiffs may rely on confidential witnesses if they plead facts with sufficient particularity to support the probability that a person in the confidential witness’s position would possess the information alleged.” Doshi v. Gen. Cable Corp., 823 F.3d 1032, 1037 n.2 (6th Cir. 2016); e.g., Plymouth Cnty. Ret. Sys. v. Evolent Health, Inc., No. 1:19-CV-01031, 2021 WL 1439680, at *35 (E.D. Va. Mar. 24, 2021) (finding “that the Plaintiffs have pleaded facts with particularity that demonstrate that the CWs would have had knowledge of the matters on which they spoke,” despite “not alleg[ing] any direct interaction with any Individual Defendant”). Here, all the CWs were described with sufficient particularity to support the basis of their allegations. See ECF No. 42 at PageID.1042–44. c. Third, Defendants argue that the allegations of CW-5, CW-6, CW-7, CW-8, and CW-9 do not “advance the ball on scienter,” because they are “all former entry-level Rocket employees.” ECF No. 44 at PageID.1230. But those CWs provide “details regarding the witnesses and the alleged information they

perceived,” which further “contribute[s] to the holistic assessment of scienter.” Karinski v. Stamps.com, No. 2:19-CV-01828, 2020 WL 281716, at *16 (C.D. Cal. Jan. 17, 2020) (disregarding low-level-employee argument under holistic framework). Specifically, CW-4, CW-5, and CW-9 all furnish facts that reflect decreased demand and volume that corroborate Plaintiffs’ proposed expert data. Compare ECF No. 42 at PageID.1091–92 (statement of CW-4) (demonstrating that, by December 2020, Rocket was “behind goal” and management was “freaking out”), and id. at PageID.1093 (statement of CW-5) (stating that, from 2020–2021, the number of loans that CW-5 reviewed decreased by about 25% to 50%), and id. at PageID.1092 (statement of CW-9) (stating that CW-9’s supervisor, the Mortgage Banking Director, had trouble closing sales from November

2020–May 2021), with id. at PageID.1084–87 (providing expert analysis demonstrating that volume in each of Rocket’s channels was decreasing, not growing). In addition, CW-7 said UWM’s ultimatum was discussed in “virtually every one of the weekly meetings” run by the EVP of Rocket’s broker business—who reported to CEO Defendant Farner. Id. at PageID.1095. And CW-4, CW-6, and CW-8 corroborate that the UWM ultimatum caused Rocket to “los[e] a good chunk of business” and a lot of partners since March of 2021, and that CW-4 dropped from 30 new loans per month to 5 per month within a month following the UWM ultimatum. Id. at PageID.1095–97. This information also corroborates that Rocket’s market share and broker loan volume were declining. See id. at PageID.1097. d. Fourth, the inference of Farner’s scienter is strengthened by the March 10, 2021 Audit Committee Meeting that he attended. See ECF No. 42 at PageID.1110–12, 1146. The primary purpose of the meeting was for management to review Rocket’s quarterly earnings releases and SEC filings and, according to the meeting minutes, to discuss certain “business matters.” Id. at

PageID.1110–11. Given the importance of market share, gain-on-sale margin, and rising interest rates to Rocket’s quarterly financials, see id. at PageID.1054–62, 1107, Farner likely would have discussed the impact of rising interest rates, as well as the declining broker-market share caused by the March 4, 2021 UWM ultimatum—which was announced a mere six days before the meeting. See id. at PageID.1095; No. 84 Emp.–Teamster Joint Council Pension Tr. Fund v. Am. W. Holding Corp., 320 F.3d 920, 943 (9th Cir. 2003) (crediting executives’ attendance at board meetings as demonstrating their scienter of company’s “maintenance and operational problems and the misstatements made”). And—most importantly—all this information was confirmed at the March 23, 2021 Board

Meeting that Gilbert attended, where Farner and Booth presented and discussed the revised forecasts for the year. ECF No. 42 at PageID.1112. e. The third and sixth Helwig factors also support scienter. Here, the six-to-nine weeks’ gap from the February and March statements to the May disclosures “comfortably” supports scienter. See Dougherty v. Esperion Therapeutics, Inc., 905 F.3d 971, 981 (6th Cir. 2018). Before making those statements, Farner and Gilbert disregarded the “most current factual information,” id., regarding loan volume, interest rates, market share, and gain-on-sale margin. Indeed, Defendants acknowledge that such information was available to them on at least a monthly basis. See ECF No. 42 at PageID.1154 (acknowledging Farner’s knowledge of purchase application volume for March and April of 2021); see also id. at PageID.1047–48, 1084–86, 1097–98. f. Chairman Gilbert’s insider trade of $500 million also supports a strong inference of scienter under the first Helwig factor. See Chamberlain v. Reddy Ice Holdings, Inc., 757 F. Supp. 2d 683,

710 (E.D. Mich. 2010) (“‘Insider trading at a suspicious time or in an unusual amount’ comprises one of the ‘fixed constellations of facts that courts have found probative of securities fraud.’” (quoting PR Diamonds, Inc. v. Chandler, 364 F.3d 671, 691 (6th Cir. 2004))). Much more than “suspicious,” id. at 710–11, Gilbert had never sold Rocket stock during Rocket’s public history, and his March 29, 2021 sale belied Defendant Farner’s public statements to investors—made a mere 26 days earlier—that “between myself and [Gilbert] . . . we have not sold shares . . . because we believe strongly in the future of what we’re building.” ECF No. 42 at PageID.1034, 1149. What changed during those three weeks? Defendants do not say. Yet Gilbert specifically requested the Audit Committee—who relied on the approval of Farner rather than Rocket’s

General Counsel—to facilitate the trade because Rocket’s internal trading window was closed, prohibiting Gilbert’s trade. Id. at PageID.1149–50. And “restrictions on an insider’s ability to trade are important in determining whether the trading pattern is suspicious.” Ronconi v. Larkin, 253 F.3d 423, 436 (9th Cir. 2001) (citing In re Silicon Graphics Inc. Sec. Litig., 183 F.3d 987–88 (9th Cir. 1999)). Thus, Farner’s vote strongly infers scienter for him and for Gilbert. If Gilbert was already planning to sell Rocket stock—rather than reacting to learning adverse facts about Rocket’s key performance indicators—then why didn’t he avoid the difficulty by selling during the open trading window mere days earlier? ECF No. 42 at PageID.1149–50.16 Defendants do not say. Rather, Defendants counter that Gilbert’s insider trading is not suspicious, because he only sold 1% of the Rocket stock that he owned. ECF No. 44 at PageID.1230–31. However, they provide no legal authority for the proposition that the Securities Exchange

Act provides a more flexible footing for more significant owners. Contra BILLY JOEL, Only the Good Die Young, on THE STRANGER (Columbia 1978). True, they cite a few cases to argue “the inference of scienter is weak where an officer sells only a small fraction of the shares owned.” In re Credit Acceptance Corp. Sec. Litig., 50 F. Supp. 2d 662, 677 (E.D. Mich. 1999); accord In re PEC Sols. Sec. Litig., 418 F.3d 379, 390 (4th Cir. 2005); Ronconi v. Larkin, 253 F.3d 423, 435– 36 (9th Cir. 2001); Acito v. IMCERA Grp., 47 F.3d 47, 54 (2d Cir. 1995). But those cases are not controlling precedent here. Pratt v. KSE Sportsman Media, Inc., 586 F. Supp. 3d 666, 674 (E.D. Mich. 2022) (first quoting Hall v. Eichenlaub, 559 F. Supp. 2d 777, 782 (E.D. Mich. 2008); and then quoting Malam v. Adducci, 469 F. Supp. 3d 767, 788 (E.D. Mich. 2020)). Even if they were

controlling, it would only mean those sales did not increase the inference of scienter—not that selling a low percentage of shares lowers the inference of scienter in all circumstances. See, e.g.,

16 Defendants assert that Gilbert sold stock in Rocket’s IPO, see ECF No. 44 at PageID.1231 (citing In re Silicon Graphics Inc. Sec. Litig., 183 F.3d 970 (9th Cir. 1999)), but they do not corroborate it and Plaintiffs do not plead it. Even if Gilbert sold stock in the IPO, his one-time sale of half a billion dollars would still be suspicious under the “highly context-specific” inquiry that “depends on the other allegations offered in the Complaint.” In re Cardinal Health Inc. Sec. Litig., 426 F. Supp. 2d 688, 728 (S.D. Ohio 2006). The overwhelming evidence here points to the trade being suspicious under the circumstances. See Tellabs, 551 U.S. at 324, 326. (“The inference that the defendant acted with scienter need not be irrefutable, i.e., of the ‘smoking-gun’ genre.”). And, contrary to Defendants’ insinuations, “[the Sixth Circuit] ha[s] never held that the absence of insider trading defeats an inference of scienter.” PR Diamonds, Inc. v. Chandler, 91 F. App’x 418, 436 (6th Cir. 2004) (unpublished). So the underlying rationale for their out-of-circuit cases is inapposite. Ronconi, 253 F.3d at 436 (“Also, DeBuono’s trading supports only a weak inference, not a strong one, in light of what the other, equally knowledgeable, insiders were doing.”). Yet, aside from the myriad of other factual differences that make those cases inapposite, precedent cuts the other way too. For example, the Second Circuit has rejected the more-money-less-problems argument that Defendants make here. See, e.g., San Leandro

Emergency Med. Grp. Profit Sharing Plan v. Philip Morris Cos., 75 F.3d 801, 814–15 (2d Cir. 1996) (holding that, rather than “vitiate any insider trading liability,” defendant’s “retention of a large position in Philip Morris and his pre-announcement purchases . . . . give[] rise to an inference that he engaged in insider trading, which, if successful, would trigger a duty to disgorge profits”). Most notably, Gilbert traded half a billion dollars of Rocket stock in a single day. “[W]here, as here, stock sales result in a truly astronomical figure, less weight should be given to the fact that they may represent a small portion of the defendant’s holdings.” Nursing Home Pension Fund, Loc. 144 v. Oracle Corp., 380 F.3d 1226, 1232 (9th Cir. 2004) (finding suspicious a stock sale of 2.1% of defendant’s total holdings due to the large dollar amount of the trade). Thus, Gilbert’s

$500 million trade raises a strong inference of scienter. Chamberlain v. Reddy Ice Holdings, Inc., 757 F. Supp. 2d 683, 711 (E.D. Mich. 2010). g. Seventh, “[c]ourts may presume that high-level executives are aware of matters related to their business’ operation where the misrepresentations or omissions pertain to ‘central, day-to-day operational matters.’” In re Cardinal Health Inc. Sec. Litig., 426 F. Supp. 2d 688, 724 (S.D. Ohio 2006). During the Class Period, Jay Farner was the CEO of Rocket and RHI and the Vice Chairman of Rocket’s Board. ECF No. 44 at PageID.1209 (citing ECF No. 42 at PageID.1037). The question of Farner’s scienter is reduced to whether he knowingly or recklessly misrepresented (1) the impact of rising interest rates on Rocket’s key performance indicators at Rocket’s 4Q 2020 Earnings Call on February 25, 2021, (2) Rocket’s declining loan volume at a Morgan Stanley virtual conference on March 3, 2021, (3) Rocket’s declining market share and loan volume in a tweet on March 17, 2021. The latter allegation is the basis for Gilbert’s scienter, who was the

Founder of Rocket Mortgage, the Chairman of the Board of Rocket and RHI, and Rocket’s majority stockholder with direct or indirect ownership of 99.9% of Rocket’s outstanding Class D Common Stock and 93.1% of Rocket’s Class A Common Stock on a fully exchanged and converted basis. Id. at PageID.1209 (citing ECF Nos. 42 at PageID.1037–38, 1040, 1044; 44-8 at PageID.1324). Market share and closed-loan volume are not only Rocket’s internal KPIs and form the basis for the revenue that Rocket generated in its two primary channels (i.e., broker and retail), but also Rocket’s most fundamental metrics along with interest rates and gain-on-sale margin. See ECF No. 42 at PageID.1147–48.

h. As a “more compelling inference,” Defendants argue that Farner and Gilbert “reasonably believed that they ‘properly disclosed relevant information’ during an evolving economic situation.” ECF No. 44 at PageID.1231 (quoting Jones v. Perez, 550 F. App’x 24, 26 (2d Cir. 2013) (unpublished)). To them, their conduct is “more consistent with a company playing ‘a constant game of catch-up than with fraud.’” Id. at PageID.1232 (quoting Pittman v. Unum Grp., 861 F. App’x 51, 57 (6th Cir. 2021) (unpublished)). If Defendants were trying to “catch-up,” as they urge, then they eventually would have released a corrective disclosure for their alleged misrepresentations. But Defendants have not provided any corrective disclosures. So the more plausible inference is that Gilbert and Farner knew that Rocket’s key performance indicators were declining, released a truthful explanation in the Q1 2021 disclosures, and then misrepresented those declines in numerous public statements to quell the imminent drop in stock price that would follow the Q2 2021 disclosures, which revealed that the same key performance indicators were declining, thereby

allowing Gilbert to sell half a billion dollars of his Rocket shares for 33% more than he otherwise would have. For those reasons, Plaintiffs have adequately pled a strong inference of scienter against Jay Farner and Daniel Gilbert. Frank v. Dana Corp., 646 F.3d 954, 961–62 (6th Cir. 2011). E. The § 20(a) claims against RHI, Farner, and Gilbert rise and fall with the § 10(b) claims. To that end, Plaintiffs claim that RHI, Farner, and Gilbert acted as “controlling persons” in the § 10(b) violation, which would constitute a violation of § 20(a) of the Securities Exchange Act, which provides that:

Every person who, directly or indirectly, controls any person liable under any provision of this chapter or of any rule or regulation thereunder shall also be liable jointly and severally with and to the same extent as such controlled person to any person to whom such controlled person is liable (including to the Commission in any action brought under paragraph (1) or (3) of section 78u(d) of this title), unless the controlling person acted in good faith and did not directly or indirectly induce the act or acts constituting the violation or cause of action.

15 U.S.C. § 78t(a) (2011). Section 20(a) claims are predicated upon at least one underlying violation committed by a controlled party. Dana Corp., 646 F.3d at 962 (citations omitted). “Because Plaintiffs have adequately pleaded scienter as to [Farner] and [Gilbert]— [respectively RCI’s and RHI]’s chief executive officer and [RCI’s and RHI’s] controlling shareholder—they have also pleaded scienter as to [RCI and RHI].” Id. at 963 (collecting cases). F. Finally, Plaintiffs allege insider trading against Daniel Gilbert and RHI under § 20A of the Securities Exchange Act. 15 U.S.C. § 78t-1. 1. To plead a § 20A claim, a plaintiff must separately allege a predicate violation of the

securities law and show that the violator purchased or sold “the same class” of securities “contemporaneously with the purchase or sale of securities that [are] the subject of [the] violation.” Beach v. Healthways, Inc., No. 3:08-CV-00569, 2009 WL 650408, at * 5–6 (M.D. Tenn. Mar. 9, 2009) (citing In re Enron Corp. Sec., Derivative & ERISA Litig., 258 F. Supp. 2d 576, 599 (S.D. Tex. 2003)). An insider who trades company stock “has a duty to disclose any nonpublic, material information he knows about the company before trading.” J & R Mktg. v. Gen. Motors Corp., 549 F.3d 384, 398 (6th Cir. 2008) (citing Chiarella v. United States, 445 U.S. 222, 227–29 (1980)). Failure to do so violates § 10(b) of the 1934 Act and can be a predicate for a § 20A claim. E.g.,

Beach, 2009 WL 650408, at *6; see Tellabs, Inc. v. Makor Issues & Rts., Ltd., 551 U.S. 308, 318 (2007) (citing 17 CFR § 240.10b-5). Section 20A claims “sound in fraud” and must be pleaded with particularity under Rule 9(b). See Picard Chem. Inc. Profit Sharing Plan v. Perrigo Co., 940 F. Supp. 1101, 1130 (W.D. Mich. 1996). Here, Gilbert made the relevant trade through RHI, which he controlled. ECF No. 42 at PageID.1039–40. Defendants do not dispute and, therefore, concede that RHI’s insider-trading liability hinges on Gilbert’s liability. Defendants also do not dispute Gilbert’s status as an insider, that he made the trade as alleged, that he avoided losses through that trade, that he obtained specific projections at the Board Meeting shortly before that trade, or that he did not disclose the projections before trading. Indeed, “for purposes of this Motion to Dismiss, alleging ‘contemporaneous’ sales by the Defendants is sufficient.” Beach, 2009 WL 650408, at *6. Rather, Defendants challenge the insider-trading claims by arguing (1) the information was not material, (2) the information was public, and (3) Gilbert did not act with scienter.17 ECF No. 44 at PageID.1234–38. These arguments will be addressed in turn.

2. The first issue is whether the information that Gilbert knew when he traded (the “Projection”) was material. a. “‘Materiality depends on the significance the reasonable investor would place on the withheld or misrepresented information.’ . . . That is, would the information . . . have ‘significantly altered the total mix’ of information made available?’” City of Monroe Emps. Ret. Sys. v. Bridgestone Corp., 399 F.3d 651, 669 (6th Cir. 2005) (citations omitted). At the motion-to-dismiss stage, a supposed lack of materiality may not support dismissal unless the information is “so

obviously unimportant to a reasonable investor that reasonable minds could not differ on the question.” Helwig v. Vencor, Inc., 251 F.3d 540, 563 (6th Cir. 2001) (collecting cases), abrogated on other grounds by Tellabs, 551 U.S. 308. “[M]aterial information . . . includes any sort of information that might affect the value of the stock in question.” DONALD C. LANGEVOORT, 18 INSIDER TRADING REGULATION,

17 In their introduction but nowhere else, Defendants assert Gilbert spent the proceeds of his insider trading on charity. ECF No. 44 at PageID.1208. Even if true, charitable expenditures do not affect liability. See City of Warren Police & Fire Ret. Sys. v. World Wrestling Ent. Inc. (“WWE”), 477 F. Supp. 3d 123, 130 (S.D.N.Y. 2020) (disregarding use of proceeds to fund other venture); see also Salman v. United States, 580 U.S. 39 (2016) (per curiam) (holding that insider-trading liability includes “making a gift of confidential information to a trading relative”). ENFORCEMENT AND PREVENTION § 5:3 (May 2022). Notably, “[p]rojections and estimates regarding income and earnings can certainly be material,” even if they merely make investors’ “prior assumptions” less reliable. Id. (citing In re Atl. Fin. Mgmt., Inc. Sec. Litig., 718 F. Supp. 1012 (D. Mass. 1988).); see also Hedick v. Kraft Heinz Co., No. 1:19-CV-01339, 2021 WL 3566602 (N.D. Ill. 2021) (opining that projections might not be material yet finding sufficient

grounds for plaintiffs’ claims based on a controlling shareholder’s sale of more than a billion dollars of stock with knowledge that the company was in financial peril). “What is most important is that the information, by itself, be meaningful to the investing public in light of all other information available to it.” Id. (emphasis added). b. Here, the Projection includes slides that were presented to Mr. Gilbert during the Board Meeting on March 23, 2021, which were partially prepared by Defendant Booth. ECF No. 42 at PageID.1112–13. The Projection demonstrates that, throughout 2021, management forecasted a 9% decline in Rocket’s gain-on-sale margin: a $950 million decline in revenue. Id. The Projection

discussed with Mr. Gilbert also includes other KPIs, like an 80% decrease in refinancing volume and compression of Rocket’s primary-secondary spread. Id. at PageID.1034. The discussions of those material KPI heavily leans toward materiality. “Internal projections of profits and revenues are the type of confidential information that a corporate officer obtains by reason of his position. Such projections give rise to a duty to abstain or [to] disclose if they are material.” SEC v. Hoover, 903 F. Supp. 1135, 1144 (S.D. Tex. 1995) (citing SEC v. Fox, 855 F.2d 247, 252–53 (5th Cir. 1988)); see, e.g., United States v. Smith, 155 F.3d 1051, 1064–66 (9th Cir. 1998) (holding that “forecasts of future sales and revenue” may “constitute ‘material’ information within the meaning of Rule 10b–5”); Johnson v. Aljian, 490 F.3d 778, 779–80 (9th Cir. 2007) (holding that a “report [that] projected a ‘significant’ decline in free cash flows” was MNPI because the defendant “had unrestricted access to the report”); SEC v. Brethen, No. 3:90-CV-00071, 1992 WL 420867, at *18 (S.D. Ohio Oct. 15, 1992) (holding that “Monthly Management Report and its Preliminary and Rolling Sales Forecasts” were MNPI because they were distributed to top management, contained actual or projected sales, and “ke[pt]

top management informed as to how [the company’s] divisions were then performing”). The undisclosed Projection was particularly significant because “Gain on Sale of loans, net” was more than 85% of Rocket’s total revenue. ECF No. 42 at PageID.1045–46. Investors understood that gain-on-sale margin was the “most important measure of profitability” in Rocket’s industry, and it was the primary focus of analysts in earnings calls. Id. at PageID.1056–57. Indeed, Defendants referred to it as Rocket’s key “measure of profitability,” id. at PageID.1046, 1056, touted the figure in every quarterly report, id. at PageID.1062–63, and closely monitored it as Rocket’s main KPI, id. at PageID.1055–56. Rocket’s internal view of the Projection as “significant” also heavily lends to its

materiality. Johnson v. Aljian, 394 F. Supp. 2d 1184, 1202 (C.D. Cal. 2004); e.g., Brethen, 1992 WL 420867, at *2. And to the extent that the Projection was merely a draft, “a proposal need not be final or consummated in order to be relevant to a shareholder’s decision-making process.” SEC v. Fox, 855 F.2d 247, 253 (5th Cir. 1988) (collecting cases). Materiality is also bolstered by the market’s reaction to Rocket’s public disclosure of declining KPIs nearly a month after Gilbert’s sale, which projected a decline in gain-on-sale margin and other KPIs. For Rocket’s stock price fell 17% in one day and 28% in four trading days after learning of Rocket’s declining gain-on-sale margin. ECF No. 42 at PageID.1141. “[T]he materiality of disclosed information may be measured post hoc by looking to the movement, in the period immediately following disclosure, of the price of the firm’s stock.” Oran v. Stafford, 226 F.3d 275, 282 (3d Cir. 2000); accord Brethen, 1992 WL 420867, at *18 n.100. In sum, the information in the Projection was among the “most important” when it comes to materiality. LANGEVOORT, supra, § 5:3 (“What is most important is that the information, by itself, be meaningful to the investing public in light of all other information available to it.”).

For these reasons, the information that Gilbert learned at the Rocket Board Meeting six days before he sold the Rocket stock (i.e., the Projection) was material nonpublic information (MNPI). c. In contest, Defendants first argue “Rocket’s ‘prediction[] . . . do[es] not constitute material inside information,’ because the ‘facts’ underlying the projection were ‘public[ly] availab[le].’” ECF No. 44 at PageID.1236 (quoting Freeman v. Decio, 584 F.2d 186, 199 (7th Cir. 1978) (alterations in original)). But Defendants acknowledge that the analysis in the Projection was not publicly available.

ECF No. 44 at PageID.1237 (“Rocket never provided any fiscal year 2021 guidance to the market.”). Defendants also acknowledge that “two of 13 analysts covering Rocket projected a gain-on-sale margin ‘well above’ Rocket’s internal projection.” ECF No. 50 at PageID.1817–18 (quoting ECF No. 49 at PageID.1450). As explained above, gain-on-sale margin was not “so obviously unimportant . . . that reasonable minds could not differ on the question.” City of Monroe Emps. Ret. Sys. v. Bridgestone Corp., 399 F.3d 651, 679 (6th Cir. 2005); see discussion supra Sections III.C.2, III.D.2.d, g. And the whole (of the Projection) is greater than the sum of its parts (the publicly available information)—especially in the context of expert analysis that investors have not the information to conduct. Indeed, what would be the purpose of a quarterly disclosure if investors are expected to know exactly what it will disclose? Defendants do not say. True, Defendants add that the outcome might be different “when all analysts are considered.” ECF No. 50 at PageID.1718. But that is an issue for summary judgment. Thus, this argument lacks merit. Second, Defendants add that Rocket’s public release of Q1 2021 guidance makes the Projection immaterial. See ECF No. 44 at PageID.1236–37. In sum, they argue the information is

unimportant because the Q1 2021 disclosures primed the market for the Projection. See id. But that argument is just another version of the first argument. Even so, Defendants’ own exhibit demonstrates that Rocket’s Q1 2021 disclosures publicly boasted about “Explosive Growth,” an “increase of 90–99%” in year-over-year guidance for closed-loan volume, and a substantial year-over-year “improvement” of 3.6–3.9% for gain-on-sale margin. ECF No. 44-4 at PageID.1281, 1284. Based on that information, within the next few days, Rocket’s stock soared from $19.90–$28.01 per share. ECF No. 42 at PageID.1129. But, soon after, Gilbert learned of the Projection: that those numbers would actually decline. It is a matter of common sense that the Projection’s bad news would not have increased the value of Rocket’s stock. Plus, some analysts

expected Rocket’s 2021 gain-on-sale margin to be equal to or greater than the Q1 2021 guidance. Id. at PageID.1128–29. Thus, despite what the market knew about gain-on-sale margin, it did not know that Rocket would lose nearly one billion dollars of revenue in 2021, as Rocket’s Q1 2021 guidance did not include revenue. See ECF No. 44-4 at PageID.1284. “Bad news travels at the speed of light, good news travels like molasses.” Tracy Morgan: ‘You Got to Watch What You Say,’ Interview, N.Y. TIMES (Apr. 15, 2014), https://www.nytimes.com/2014/04/20/magazine/tracy-morgan-you-got-to-watch-what-you- say.html [https://perma.cc/T9CH-4W6Z]. But the market did not negatively react until learning of the Q2 2021 guidance regarding reduced gain-on-sale margin. ECF No. 42 at PageID.1141–42, 1156 (demonstrating a nearly 17% decline in one day and a 28% decline over four days). Again, this reaction is among the “most important” evidence demonstrating that the market revered the overall decline. See LANGEVOORT, supra, § 5:3. So this argument lacks merit. Third, Defendants next argue the Projection is too “remote” to be material because its predictions would “not come into fruition” until the end of 2021. ECF No. 44 at PageID.1236.

However, approximately one month after the Board gave the Projection to Mr. Gilbert, Rocket publicly reported a decline in market share, closed-loan volume, and gain-on-sale margin for Q1 2021, ECF No. 42 at PageID.1063–64, and much lower guidance regarding gain-on-sale margin for Q2 2021, id. at PageID.1112–14, 1137–38. As explained above, acting on the Projection directly belies Defendants’ argument that the information in it was not material. And Defendants reported that the forecasted decline would occur throughout 2021. Id. at PageID.1113– 12. Although a more distant forecast could be more speculative than a shorter one, more time does not necessarily render the Prediction “obviously unimportant.” See LANGEVOORT, supra, § 5:3 (“Projections and estimates regarding income and earnings can certainly be material.”) (collecting

cases). Indeed, investors often rely on annual guidance issued at or before the beginning of a given year. See, e.g., Flowserve Corp., Exchange Act Release No. 19154, 85 SEC Docket 146, 2005 WL 677813, at *1 (Mar. 24, 2005) (holding that “[c]ompany’s earnings guidance for the year” was “material nonpublic information”). So Defendants’ remoteness argument lacks merit under the circumstances. Yet numerous cases have found insiders liable for trading based on forecasts similar to the Projection. See Johnson, 394 F. Supp. 2d at 1187, 1201–02 (holding that “Operative Planning 1999–2001” report that projected declining cash flows for 1999 but was revealed to insider in March of 1999 was MNPI); United States v. Nacchio, 519 F.3d 1140, 1145 (10th Cir. 2008) (same for April 2001 disclosure that year-end forecast for 2001 might be lower than annual guidance due to market conditions), rev’d in part on other grounds, 555 F.3d 1234 (10th Cir. 2009) (upholding conviction); see also Hedick v. Kraft Heinz Co., No. 1:19-CV-01339, 2021 WL 3566602 (N.D. Ill. 2021). So Defendants’ second argument lacks merit too. Fourth, Defendants argue that undisclosed information cannot be material without a corresponding prior public statement. See ECF No. 44 at PageID.1237 (“Rocket never provided

any fiscal year 2021 guidance to the market.”). But the absence of 2021 guidance does not make the Projection immaterial. At least one case has firmly rejected the idea that information is material only if it directly contradicts prior statements. E.g., United States v. O’Hagan, 521 U.S. 642, 647–48 (1997) (insider liable for trading on news of forthcoming merger despite no previous public comments on the prospect of that merger). In sum, the materiality analysis is holistic and based on the totality of the circumstances, which supports the conclusion that the information that Mr. Gilbert learned days before he traded half a billion dollars of Rocket stock was material.

3. The next issue is whether the information in the Projection was public. Information is nonpublic until it is disclosed “to achieve a broad dissemination to the investing public.” Dirks v. SEC, 463 U.S. 646, 653 n.12 (1983). “As with the concept of materiality, it is rarely at issue in the typical insider trading case.” LANGEVOORT, supra, § 5:4. There are two dominant theories for testing whether the information was public: the efficient-market hypothesis, and the public-access test. Id. Defendants’ arguments fail under both. Defendants do not, and could not, argue that the Projection was public under either theory. Indeed, they have conceded that the Projection was not disclosed to the public when they argued it was immaterial because the information used to create it was public. See discussion supra Section III.F.2.e. This ends the inquiry. But Defendants’ argument would fail on the merits too. Under the efficient-market hypothesis, the court asks whether the market has yet “fully” internalized the information. See United States v. Libera, 989 F.2d 596, 601 (2d Cir. 1993). Generally, this means that the insider must wait for some time after the information has been

communicated to a critical mass of marketplace traders (who are not themselves acting unlawfully), SEC v. Tex. Gulf Sulphur Co., 401 F.2d 833, 848 (2d Cir. 1968) (en banc), which means one full trading day at a minimum, LANGEVOORT, supra, § 5:4, or that it was already in the hands of a significant number of investment analysts, Tex. Gulf, 401 F.2d at 848–49. But neither was the case here, as Gilbert traded on the Projection six days after he learned of it—before it was disclosed to anyone. The public-access test is no more favorable to Defendants. Mostly used by the SEC, the stricter public-access test “essentially require[es] some form of dissemination of the information via nationwide media before insiders may trade.” LANGEVOORT, supra, § 5:4. As explained above,

that did not occur here. Yet Defendants argue that the Projection is not MNPI because it reflected “internal analysis of public data.” ECF No. 44 at PageID.1235–36. But internal projections reflecting management’s judgment are MNPI, even if based on some public information. See United States v. Smith, 155 F.3d 1051, 1064 n.20 (9th Cir. 1998) (“Surely the average investor’s interest would be piqued by a company’s internal projections of economic downturn.”); United States v. Rajaratnam, 802 F. Supp. 2d 491, 514 (S.D.N.Y. 2011) (“Whatever the market knew about [Rocket’s] problems, the jury [is] well within reason to conclude that the market did not know that those problems would force [Rocket] to guide down.”).18 Even if the Projection incorporated some public data, there would still be an issue of fact regarding whether all the data was publicly disseminated in a widespread manner such that it was available to every investor. See Dirks, 463 U.S. at 653 n.12; In re CBL & Assocs. Props. Sec. Litig.,

No. 1:19-CV-00181, 2022 WL 1405415, at *15 (E.D. Tenn. May 3, 2022). And—most importantly—the Projection was not a mere “internal analysis of public data,” as Defendants posit. See ECF No. 44 at PageID.1235–36. Rather, the Projection is a product of management’s private judgments based on nonpublic facts. Defendants insist that Rocket’s projections were based on only the yield on 10-year Treasury bonds, the primary-secondary spread, and the Mortgage Bankers Association’s forecast. See ECF No. 44 at PageID.1234. But Rocket’s forecasted gain-on-sale margin would have been calculated based on numerous nonpublic factors that only Rocket insiders could know, including Rocket’s future net gain for selling loans, the unpaid principal balance of all Rocket loans, and the “pull-through factor” estimate on such loans.

See ECF No. 42 at PageID.1057–59. In this way, only Rocket could know what analysis and determinations it made when arriving at its full-year 2021 internal projections.19

18 The cases that Defendants cited are inapposite because they involve publicly available information, not internal projections based on publicly available information. E.g., ECF No. 44 at PageID.1236 (citing Ponsa-Rabell v. Santander Secs., 35 F.4th 26 (1st Cir. 2022)). Further, though every corporate forecast incorporates some publicly available information, that does not immunize insiders who trade on such projections. See discussion supra Section III.F.2 (discussing insider-trading cases involving projections). 19 For this reason, Defendants’ citation to Whirpool Fin. Corp. v. GN Holdings, Inc., 67 F.3d 605, 610 (7th Cir. 1995) is inapposite. See ECF No. 44 at PageID.1235. Further, Freeman v. Decio, 584 F.2d 186, 198–99 (7th Cir. 1978), supports Plaintiffs’ position. There, the plaintiff alleged that, because the defendants were aware of price increases in lumber and the existence of price controls (both public information), the defendants “must have known” that the reported earnings for the quarter would decline. Freeman, 584 F.2d at 198–99. Yet the plaintiff presented “no evidence that the defendants actually made any such pessimistic predictions,” id.—a far cry from the facts here. 4. The final issue is Mr. Gilbert’s scienter. There is not much to add beyond the earlier analysis of Gilbert’s scienter. See discussion supra Section III.D.2. Yet Defendants’ arguments will be addressed. Scienter is usually met if the complaint adequately alleges the insider had “actual

knowledge of undisclosed material information; knew it was undisclosed; and knew it was material.” SEC v. MacDonald, 699 F.2d 47, 50 (1st Cir. 1983); e.g., In re Envision Healthcare Corp. Sec. Litig., No. 3:17-CV-01112, 2019 WL 6168254, at *25 (M.D. Tenn. Nov. 19, 2019) (“Plaintiffs have sufficiently alleged that the [defendants] knew of EmCare’s out-of-network billing strategy.”); Picard Chem. Inc. Profit Sharing Plan v. Perrigo Co., 940 F. Supp. 1101, 1130– 31 (W.D. Mich. 1996) (upholding §10(b) insider trading claim against entity because entity “may be assumed to have had the same knowledge as the [individual defendant]”). Here, the Second Amended Complaint exhibits the actual slides from the March 23, 2021 Board Meeting, which Rocket’s Board presented to Mr. Gilbert. ECF No. 42 at PageID.1112–14.

It also places Mr. Gilbert in the Board Meeting, where the Board gave him the Projection (i.e., the MNPI)—less than a week before the Board voted to change Rocket’s policy to permit him to sell his Rocket shares. Id. These facts are sufficient to show that Plaintiff has plausibly alleged Mr. Gilbert’s knowledge of the Projection. See LANGEVOORT, supra, § 5:3 (“[I]t is not even necessary that the tippee ever learn of the specific facts underlying the tip.” (collecting cases)).

Indeed, Plaintiffs’ theory here is not that Gilbert and RHI failed to disclose the mortgage interest rate; rather, they alleged that Gilbert and RHI failed to disclose the Projection, which demonstrated Rocket management’s original analysis and determination of a 9% decline in profit margin and a $950 million decline in revenue. Plus, Defendants consider gain-on-sale margin and revenue to be material. ECF No. 42 at PageID.1046. As Chairman of the Board, Mr. Gilbert would have known that the projected decline in revenue and gain-on-sale margin were nonpublic, as the public was not invited to the Board Meeting. See id. at PageID.1110–12. For those reasons, Plaintiffs have adequately pled:

 Violations of § 10(b) of the Exchange Act and Rule 10b-5 against Defendants Daniel Gilbert and RHI (Count I);  Violations of § 20A of the Exchange Act against Defendants Daniel Gilbert and RHI (Count II);  Violations of § 10(b) of the Exchange Act and Rule 10b-5 against Defendants RCI, Jay Farner, and Daniel Gilbert (Count III); and  Violations of § 20(a) of the Exchange Act against Defendants RHI, Jay Farner, and Daniel Gilbert (Count IV).

Consequently, Plaintiffs’ Motion to Strike will be granted in part, Defendants’ Motion to Dismiss will be granted in part, Defendants Julie Booth and Robert Walters will be dismissed with prejudice, and a scheduling order will be entered. IV. CONCLUSIONS Accordingly, it is ORDERED that Plaintiffs’ Motion to Strike, ECF No. 51, is GRANTED IN PART and DENIED IN PART. The Motion to Strike is GRANTED to the extent that seeks to strike the Wedbush Report, ECF No. 50-4; the Motion to Strike is DENIED in all other regards. Further, it is ORDERED that Defendants’ Motion to Dismiss, ECF No. 44, is GRANTED IN PART and DENIED IN PART. The Motion to Dismiss is GRANTED with respect to dismissing Defendants Julie Booth and Robert Walters; the Motion to Dismiss is DENIED in all other regards. Further, it is ORDERED that Defendants Julie Booth and Robert Walters are DISMISSED WITH PREJUDICE. Further, it is ORDERED that the Scheduling Order is SET as follows: Defendant’s Expert Disclosures: September 4, 2023; Class Certification Motion: September 4, 2023; Rule 26(a)(3)(B) Disclosures: September 6, 2023; Plaintiff’s Expert Disclosures: November 6, 2023; Class Certification Response: November 6, 2023; Class Certification Reply: November 20, 2023; Plaintiff’s Reply to Expert Report: November 20, 2023; Settlement Conference: November 21, 2023, at 9:00 AM EST; Motions Challenging Experts: December 22, 2023; Discovery Cutoff: February 5, 2024; Dispositive Motions: March 4, 2024; Motions in limine: May 6, 2024; Pretrial Submissions: June 3, 2024; Final Pretrial Conference: TBD; and Jury Trial Date: TBD.

This is not a final order and does not close the above-captioned case. Dated: March 8, 2023 s/Thomas L. Ludington THOMAS L. LUDINGTON United States District Judge