In the

United States Court of Appeals For the Seventh Circuit ____________________ No. 23-1073 APPVION, INC. RETIREMENT SAVINGS AND EMPLOYEE STOCK OWNERSHIP PLAN, by and through GRANT LYON in his capacity as the ESOP Administrative Committee of Appvion, Inc., Plaintiff-Appellant,

v.

DOUGLAS P. BUTH, et al., Defendants-Appellees. ____________________

Appeal from the United States District Court for the Eastern District of Wisconsin No. 1:18-cv-01861 — William C. Griesbach, Judge. ____________________

ARGUED DECEMBER 1, 2023 — DECIDED APRIL 23, 2024 ____________________

Before WOOD, ST. EVE, and LEE, Circuit Judges. WOOD, Circuit Judge. Appvion, Inc., is a Wisconsin-based paper company. In the 1990s, Appvion was owned by French conglomerate Arjo Wiggins Appleton. For many years, Ap- pvion struggled to stabilize its financial footing, but this proved to be difficult. Unable to find a suitable buyer as of late 2000, Arjo decided to explore a sale to Appvion’s employees. 2 No. 23-1073

That sale would take the form of an “employee stock owner- ship plan,” or ESOP, and would be financed by the employ- ees’ retirement savings. That is the path Appvion took, but ultimately it was not enough to save the company, which de- clared bankruptcy in 2017. The bankruptcy court authorized Grant Lyon to act on behalf of the Plan. After conducting an internal investigation, Lyon raised an avalanche of claims against dozens of individuals and corporations. His central theory is that the defendants fraudulently inflated the price of Appvion in 2001 and that the price remained inflated until Appvion’s bankruptcy. All the defendants filed motions to dismiss, and the dis- trict court granted almost all the relief they sought. We com- mend the court for its careful review of this complex matter. We find no error in many of its rulings, but we conclude that reversal is necessary for a subset of the counts based on the Employee Retirement Income Security Act (ERISA). We there- fore remand for further proceedings. I. Background Because this case comes to us at the motion-to-dismiss stage, we accept all well-pleaded facts in Lyon’s complaint as true and draw all reasonable inferences in his favor. Hernan- dez v. Illinois Institute of Tech., 63 F.4th 661, 666 (7th Cir. 2023). A. The Plan Appvion (formerly known as Appleton Papers) is a com- pany specializing in two types of paper, both rather dated in today’s digital age: carbonless paper, which transfers infor- mation written on a top sheet to sheets beneath it, and thermal paper, now used mostly for receipts and lottery tickets. Arjo began trying to sell Appvion in 1998, but it struggled to find No. 23-1073 3

a buyer on the open market. It then offered a contingent fee of up to $10 million to Appvion’s officers, including its then- CEO Douglas Buth, if they managed to sell Appvion for over $700 million. Buth thought that he could meet that goal through a sale of Appvion to its employees, and Arjo gave him a green light to move forward. Buth hired investment banking firm Houlihan Lokey to organize the sale, agreeing to pay Houlihan 1% of the final sale price if the deal closed. Houlihan in turn engaged State Street Bank and Trust Company to be the trustee of the em- ployees’ ownership interest in Appvion. Houlihan also hired Willamette Management Associates to appraise Appvion for purposes of the sale; Willamette determined that Appvion was worth $810 million. Houlihan then issued an opinion finding that to be a fair price for Appvion, despite the fact that its fee was contingent on closing and even though an Appvion newsletter had warned that no firm with a contingent fee was qualified to issue a fairness opinion. State Street and Appvion’s officers began trying to con- vince Appvion’s employees to accept the deal at that price, relying in part on Houlihan’s fairness opinion. In a series of road shows and letters to the employees, they repeatedly de- scribed Houlihan as an independent advisor, never mention- ing Houlihan’s contingent fee. Appvion’s officers also pre- sented the employees with a prospectus outlining the deal. The prospectus disclosed the officers’ contingent fees, but again failed to disclose Houlihan’s fee even as it touted Houli- han’s fairness opinion. In November 2001, the employees voted in favor of buying Appvion, and the deal closed. It was structured as a sale to a newly formed corporation, Paperweight Development 4 No. 23-1073

Company. The employees contributed $107 million from their retirement accounts to buy 100% of Paperweight’s stock; Pa- perweight used that money, plus over $700 million in loans, to buy 100% of Appvion’s stock for its appraised value of $810 million. Thus, Paperweight became the sole owner of Ap- pvion and was in turn fully owned by Appvion’s employees. (There was little practical difference between Paperweight and Appvion: their finances were consolidated and they had identical Boards. For simplicity, we refer to Appvion except where the distinction is relevant.) Houlihan received its $8.1 million fee, while Buth, general counsel Paul Karch, and two other Appvion officers also received fees totaling around $4 million for helping the sale go through. After the sale was finalized, Appvion formed the Em- ployee Stock Ownership Plan Administrative Committee (“the ESOP Committee”). The ESOP Committee, whose mem- bers were Board-appointed officers of Appvion, was the named fiduciary of the Plan. It was responsible for the Plan participants’ purchases of Appvion stock going forward. It also was in charge of selecting a trustee to hold the Plan par- ticipants’ shares of Appvion. Over the next 16 years, the ESOP Committee retained three trustees: State Street, from 2001 to 2013; Reliance Trust Company, from 2013 to 2014; and Argent Trust Company, from 2014 to 2017. To facilitate sales and purchases of Appvion shares by the employees, the trustees recalculated the fair market value of Appvion twice a year. The Plan’s documents required the trustees to hire an independent appraiser to help with that task. State Street initially retained Willamette as Appvion’s appraiser. In 2004, however, the Willamette employees who had worked on the Appvion account left to become managing No. 23-1073 5

directors at the investment-banking firm Stout Risius Ross. The trustees moved their business over to Stout, which con- tinued Willamette’s valuation practices. Each valuation proceeded as follows: The appraiser calcu- lated the fair market value of Appvion, based in part on finan- cial projections provided by Appvion’s directors and officers and in part on Appvion’s assets and liabilities. The appraiser gave that valuation to the trustee, which used it to set the new price of a share of Appvion. The ESOP Committee then re- viewed and approved the price set by the trustee, reported it to the Plan participants, and used it to approve purchases and sales of Appvion’s shares. The price varied over time, starting at $10 per share in the 2001 sale, reaching a high of $33.62 by the end of 2006, and gradually falling to a low of $6.85 in 2016. Four relevant events took place between the 2001 sale and Appvion’s bankruptcy in 2017. First, shortly after the sale, State Street ceded to Appvion’s CEO most of the trustee’s power over nominating and removing directors of Appvion. Second, Appvion’s Board adopted various incentive plans that awarded bonuses to the directors and officers. These bo- nuses initially rewarded increases in Appvion’s price from one valuation to the next. But after the stock price began de- clining, the Board switched to incentive programs that were calculated solely on the basis of Appvion’s share price, mean- ing that the directors and officers reaped bonuses even if Ap- pvion’s value had fallen since the last valuation. Third, in 2003, Appvion bought BemroseBooth, an English ticketing company, for $63.5 million. The transaction was a flop: by 2007 Appvion’s financial statements disclosed that, after off- setting BemroseBooth’s assets by its unfunded pension debt, BemroseBooth was worth $0. And fourth, the directors and 6 No. 23-1073

officers repeatedly tried and failed to sell Appvion. The clos- est they came was in early 2012, but the deal soon fell through. As noted, Appvion declared bankruptcy in 2017. In con- nection with that bankruptcy, Lyon became the sole member of the ESOP Committee, replacing Appvion’s officers. The bankruptcy court authorized Lyon to bring claims on behalf of the Plan. Upon reviewing Appvion’s internal documents, Lyon discovered that in the 2001 sale and for every valuation thereafter, Willamette (and then Stout) had adopted appraisal methods that, in Lyon’s view, fraudulently inflated the price of Appvion’s shares. Three of these methods are particularly important. First, the appraisers did not question the insiders’ financial projections when using them to calculate Appvion’s enterprise value, even though year after year those projections proved overly optimistic. Second, the appraisers added a con- trol premium to Appvion’s enterprise value—at first 15% of Appvion’s calculated value, and later 10%—to account for the Plan’s control of Appvion, even though the Plan participants had very little power over the Board’s members or its deci- sions. And third, when offsetting Appvion’s enterprise value by its liabilities, the appraisers excluded substantial portions of Appvion’s debt, particularly its unfunded pension debt. Had they included those debts as liabilities, Lyon alleges, the Plan’s equity in Appvion would have been worth much less— as little as $0 for most of the relevant years. B. Procedural History Lyon filed a First Amended Complaint (FAC) against Houlihan, the Plan’s three trustees, its two independent ap- praisers, various individuals associated with those corporate defendants, and the directors and officers and their spouses. Spanning more than 200 pages, the FAC charged a total of 59 No. 23-1073 7

named defendants with violations of ERISA, federal securities law, and various state laws. The district court dismissed the FAC in its entirety, with- out prejudice, pursuant to Federal Rule of Civil Procedure 12(b)(6). It held that the pre-2012 ERISA claims were barred by ERISA’s statute of repose and that the post-2012 ERISA claims were improperly group-pleaded and lacked particu- larity. It also held that the federal securities claims failed for lack of scienter. The court dismissed the state-law theories as preempted by ERISA and, as to the claims against Houlihan, also for failure to state a claim. Lyon took advantage of his opportunity to replead and filed a prolix Second Amended Complaint (SAC), which drops some of the FAC’s defendants and adds some new ones. The SAC is around 400 pages, and it responds to the dis- trict court’s demand for individualized allegations about each defendant’s role in the alleged fraud. In total, it raises 37 the- ories for recovery, against everyone in sight: the directors and officers, Houlihan, the Plan’s three trustees (State Street, Reli- ance, and Argent), and the Plan’s second independent ap- praiser (Stout). The district court referred the SAC to a mag- istrate judge, who determined that Lyon had not fixed most of the problems the district court had identified in the FAC. The magistrate judge recommended that the court again dis- miss all counts except two ERISA theories against Argent, the Plan’s most recent trustee. The district court adopted the report almost in its entirety, dismissing with prejudice all but those two counts. Lyon moved for entry of final judgment on the dismissed claims, and the district court obliged with a judgment under Federal Rule of Civil Procedure 54(b). That order lay the foundation 8 No. 23-1073

for the present appeal, even as Lyon continues to pursue the claims against Argent in district court. On appeal, Lyon has abandoned his prohibited-transac- tions theories against the trustees (SAC Counts 22–23) and his ERISA arguments against Houlihan (SAC Counts 36–37 and part of SAC Count 21). He is appealing the dismissal of the following parts of the case: (1) all of his ERISA claims against the directors and officers and most of his ERISA claims against State Street and Reliance; (2) his federal securities- fraud claims against Reliance, nine of the directors and offic- ers, and Stout and two of Stout’s managing directors; and (3) his state-law theories against the directors and officers, Stout, and Houlihan. We take up the arguments in that order, considering each of those rulings de novo. Chaidez v. Ford Motor Co., 937 F.3d 998, 1004 (7th Cir. 2019). II. ERISA ERISA is a “comprehensive statute designed to promote the interests of employees and their beneficiaries in employee benefit plans.” Shaw v. Delta Air Lines, Inc., 463 U.S. 85, 90 (1983). The statute does not require an employer to provide a benefit plan, but if an employer chooses to do so, ERISA im- poses “various uniform standards, including rules concerning reporting, disclosure, and fiduciary responsibility,” on that employer. Id. at 91. Most of Lyon’s claims arise under ERISA, based on the relationship between the directors, officers, and trustees (as fiduciaries) and the Plan participants (as benefi- ciaries). A. Pre-2012 Claims ERISA contains a statute of repose, which reflects “a legis- lative judgment that a defendant should be free from liability No. 23-1073 9

after the legislatively determined period of time.” California Public Employees’ Retirement System v. ANZ Securities, Inc., 582 U.S. 497, 505 (2017) (“CalPERS”) (quoting CTS Corp. v. Wald- burger, 573 U.S. 1, 9 (2017)). The relevant part of the statute provides that: No action may be commenced under this subchapter with respect to a fiduciary’s breach of any responsibil- ity, duty, or obligation under this part, or with respect to a violation of this part, after the earlier of— (1) six years after (A) the date of the last action which constituted a part of the breach or violation, or (B) in the case of an omission, the latest date on which the fiduciary could have cured the breach or violation. 29 U.S.C. § 1113. Lyon filed his first complaint on November 26, 2018. Sec- tion 1113(1) therefore normally would bar liability for any ac- tions or omissions prior to November 26, 2012. But there is an exception to the statute of repose, which states that “in the case of fraud or concealment, [an] action may be commenced not later than six years after the date of discovery of such breach or violation.” Id. If Lyon can show fraud or conceal- ment, that exception saves his ERISA claims, because he filed suit fewer than six years after discovering the breaches. In Radiology Center, S.C. v. Stifel, Nicolaus & Co., we held that ERISA’s use of the phrase “fraud or concealment” adopts the fraudulent-concealment doctrine. 919 F.2d 1216, 1221 (7th Cir. 1990). That doctrine refers to “steps taken by wrongdoing fiduciaries to cover their tracks”—that is, it focuses on “steps taken by the defendant to hide the fact of the breach rather than … the underlying nature of plaintiffs’ claim.” Id. at 1220. 10 No. 23-1073

Under Radiology Center, Lyon must allege more than just the underlying fraud to get the benefit of the fraud-or-conceal- ment exception. Lyon points to language in some of our cases that he ar- gues opens the door to “self-concealing frauds.” He is correct that, after Radiology Center, we acknowledged that fraud and concealment is not “limited to active concealment that is sep- arate from the underlying wrongdoing”; it “can include gen- uine acts of concealment committed in the course of the un- derlying wrong.” Martin v. Consultants & Administrators, 966 F.2d 1078, 1095 (7th Cir. 1992). We were careful to note, how- ever, that “fraud claims do not receive the benefit of ERISA’s six-year statute of limitations simply because they are fraud claims.” Id. The plaintiff must allege “actual concealment— i.e., ‘some trick or contrivance intended to exclude suspicion and prevent inquiry.’” Id. (quoting Wood v. Carpenter, 101 U.S. 135, 143 (1879)); see also id. at 1103 (Posner, J., concurring). In Martin, for example, we found fraudulent concealment when the defendant channeled the proceeds of a fraudulent kick- back scheme through dummy corporations. The kickback scheme was the underlying fraud; the use of dummy corpo- rations, though “occur[ing] in the course of the fraud itself ra- ther than independently of it,” was an additional step of con- cealment taken separately from the fraud. Id. at 1095. We therefore must ask whether Lyon alleges some trick or contrivance separate from the underlying fraud. For the 2001 sale, Lyon’s underlying fraud allegation is that the defendants misrepresented the fair market value of Appvion to trick the would-be Plan participants into buying shares at an inflated price. Conflicted by their contingent fees, Houlihan and Ap- pvion’s officers conspired with State Street and Willamette to No. 23-1073 11

exclude Appvion’s substantial pension debt from Willamette’s valuations of Appvion and to add a phony con- trol premium to the equation. Houlihan represented to the employees that Willamette’s valuation was fair, and it con- vinced them to believe that representation. Buth and Karch falsely claimed that Houlihan was independent and did not disclose its conflict of interest. For the post-2001 misconduct, Lyon’s underlying allegation is that the directors, officers, and trustees continued the fraud by having Willamette (and later Stout) biannually issue inflated valuations of Appvion, which the trustees then used (with approval from the ESOP Com- mittee) to set fraudulently high prices for future stock pur- chases by the employees. The problem for Lyon is that these allegations of fraud are the same as his allegations of fraudulent concealment. Lyon argues that the defendants concealed their fraud in the 2001 sale through their misleading representations that Houlihan was independent. These are the same representations making up his fraud claim for that sale. And to show concealment from 2001 to 2012, Lyon points to the defendants’ continued approval of inflated prices for Appvion, the identical actions constituting the underlying fraud. Lyon pushes back on that conclusion, but the only exam- ple to which he points is the filing of regulatory forms. These filings, including 10-Qs and 10-Ks filed with the Securities and Exchange Commission and Form 5500s filed with the De- partment of Labor, are mandatory forms in which the defend- ants reported the appraised price of Appvion shares. Lyon in essence argues that the defendants concealed their fraud by not reporting Appvion’s real value in those filings. But the doctrine of fraudulent concealment requires “positive acts,” 12 No. 23-1073

not “[c]oncealment by mere silence.” Wood, 101 U.S. at 143; see generally Macquarie Infrastructure Corp. v. Moab Partners, L.P., No. 22-1165, 2024 WL 1588706 (S. Ct. Apr. 12, 2024) (pure omissions are insufficient to show a violation of SEC Rule 10b- 5, which prohibits fraud). Lyon’s approach would make the exception to ERISA’s statute of repose apply whenever the de- fendants failed to come clean about their fraud. ERISA re- quires more. See Wolin v. Smith Barney Inc., 83 F.3d 847, 851 (7th Cir. 1996). Because Lyon failed to allege a trick or contrivance sepa- rate from the fraud, his ERISA claims are time-barred to the extent they seek recovery for misconduct before November 26, 2012. The district court therefore correctly dismissed Lyon’s breach-of-duty claims against the directors and offic- ers who left Appvion before 2012 (SAC Counts 4–7, 11, and 19); his prohibited-transaction claim related to the 2001 sale of Appvion (SAC Count 21); and the portions of the remaining ERISA counts that relate to pre-2012 conduct (SAC Counts 1, 3, 8–10, 12–18, and 25–28). B. Post-2012 Conduct The statute of repose does not end our ERISA inquiry, be- cause some of the alleged misconduct occurred after Novem- ber 26, 2012. Lyon asserts that the directors, officers, and trus- tees continued to accept fraudulent valuations and to approve purchases at inflated prices until Appvion’s bankruptcy. He puts those allegations into three buckets: breach of duty, pro- hibited transactions, and co-fiduciary liability. The district court allowed the breach-of-duty and prohibited-transaction claims against Argent to go forward, holding that they could be read to allege imprudence rather than fraud. (Recall that Argent entered the picture in 2014, when it bought Reliance’s No. 23-1073 13

ESOP business and assumed the Appvion account.) The court dismissed the rest of the claims. We reverse the dismissal of the post-2012 ERISA claims. 1. Breach of Duty We first take up the breach-of-duty theories, which re- quire proof of three elements: that the defendants were plan fiduciaries, that they breached a fiduciary duty, and that their breach resulted in harm to the Plan. Kenseth v. Dean Health Plan, Inc., 610 F.3d 452, 464 (7th Cir. 2010). All agree that Lyon adequately pleaded harm; if the defendants misled the em- ployees into overpaying for Appvion’s stock, they harmed the Plan. He also needed, however, plausibly to allege that the defendants bore some responsibility for that harm. The directors and officers first contend that the complaint fails to point to anything indicating that they were acting as fiduciaries in approving Stout’s valuations. A person is an ERISA fiduciary “to the extent” that she “exercises any discre- tionary authority or discretionary control respecting manage- ment of such plan” or “has any discretionary authority or dis- cretionary responsibility in the administration of such plan.” 29 U.S.C. § 1002(21)(A). It is not enough to allege merely that a defendant is a fiduciary; the defendant must have been “act- ing as a fiduciary … when taking the action subject to com- plaint.” Pegram v. Herdrich, 530 U.S. 211, 226 (2000). Lyon clears that hurdle with ease. His complaint alleges that, through their role in the ESOP Committee, the officers reviewed and approved the valuations. The complaint even lists specific edits the ESOP Committee requested at several meetings with Stout’s valuation team, including adjustments to the appraised price. The outside directors on Appvion’s 14 No. 23-1073

Board were not as directly involved in the valuation process as the officers. Nonetheless, the complaint alleges that they met with the independent appraisers twice a year, reviewed the valuations at Board meetings, and could appoint or re- move members of the ESOP Committee if they were dissatis- fied with the committee’s work on the valuations. Those are not “clerical, mechanical, ministerial” functions, Pohl v. Nat’l Benefits Consultants, Inc., 956 F.2d 126, 129 (7th Cir. 1992); they are discretionary acts at the core of the directors’ and officers’ responsibilities as ERISA fiduciaries. That leaves the harder question: does the complaint con- tain enough to put the breach question in play? ERISA lays out the duties of fiduciaries in 29 U.S.C. § 1104(a)(1). As rele- vant here, a fiduciary must act for the exclusive purpose of benefiting the plan’s participants (the duty of loyalty), 29 U.S.C. § 1104(a)(1)(A), and “with the care, skill, prudence, and diligence” of a prudent fiduciary (the duty of prudence), 29 U.S.C. § 1104(a)(1)(B). These duties are “the highest known to the law.” Halperin v. Richards, 7 F.4th 534, 546 (7th Cir. 2021) (quoting Donovan v. Bierwirth, 680 F.2d 263, 272 n.8 (2d Cir. 1982)). Two rules of pleading are relevant to this analysis. The first is Federal Rule of Civil Procedure 8(a)(2), which requires a pleader to provide “a short and plain statement of the claim showing that the pleader is entitled to relief.” As the Supreme Court did in Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), we focus here on the final clause of that rule: the need to allege “enough facts to state a claim to relief that is plausible on its face,” id. at 570, meaning facts that “raise a right to relief above the speculative level,” id. at 555. No. 23-1073 15

The second rule is Federal Rule of Civil Procedure 9(b), which requires that for allegations of fraud or mistake, the pleader must “state with particularity the circumstances con- stituting fraud or mistake.” This supplants the usual “short and plain statement” requirement of Rule 8. The idea is that because even unfounded accusations of fraud can cause seri- ous harm, a plaintiff must “have some basis” for those accu- sations before making them. Uni*Quality, Inc. v. Infotronx, Inc., 974 F.2d 918, 924 (7th Cir. 1992). Rule 9(b) stands as a barrier against such “strike suits” by requiring plaintiffs to identify the precise contours of the fraud. Vicom, Inc. v. Harbridge Mer- chant Services, Inc., 20 F.3d 771, 777 (7th Cir. 1994). The district court dismissed all of Lyon’s breach-of-duty claims. But its reasons for doing so do not hold up under ex- amination. The magistrate judge had determined that Lyon had premised his claims against the directors and officers, State Street, and Reliance entirely on fraud, and that those claims therefore were subject to the “heightened pleading standard” found in Rule 9(b). The magistrate judge then rec- ommended dismissing those claims for failing to meet that standard. In footnote six, however, the magistrate judge said that he “believe[s] any singular [sic] defendant has adequate notice of the allegations against themselves to satisfy the Twombly pleading standard.” The district court adopted the recommendation as to all claims, but it refused to endorse footnote six, without specifying which standard it thought was applicable: Rule 9(b) or Twombly. We are thus unsure on what ground the district court dismissed the breach-of-duty arguments. We see three possible explanations for the district court’s ruling, but we have problems with each one. 16 No. 23-1073

First, if the court dismissed the claims because it thought Lyon failed to plead intent with particularity, it misread Rule 9(b). That rule requires plaintiffs to plead with particularity the “who, what, when, where, and how” of the fraud. DiLeo v. Ernst & Young, 901 F.2d 624, 627 (7th Cir. 1990). Notably absent from that list is the word “why.” That is because Rule 9(b) explicitly excludes from its heightened pleading standard allegations of “[m]alice, intent, knowledge, and other condi- tions of a person’s mind,” stating that they “may be alleged generally.” (Emphasis added.) In the portion of the recom- mendation that the district court adopted, the magistrate judge determined that “[t]he biggest hole in Lyon’s theory” was his failure to explain why all the defendants, many of whom had “nothing to gain and everything to lose,” would conspire to commit fraud. This reasoning reads the exclusion of intent out of Rule 9(b). Using the claim against CEO Kevin Gilligan (SAC Count 13) as a representative example, we can see that Lyon pleaded with particularity the “who, what, when, where, and how” of the fraud: Who: Kevin Gilligan, CEO of Appvion from Au- gust 2015 to October 2017. SAC at ¶ 950. What: Adopting inflated prices for Appvion’s stock. Id. at ¶ 951–53. When: On July 11, 2016; January 18, 2017; and July 17, 2017. Id. at ¶ 952. Where: At the ESOP Committee’s meetings. Id. How: Providing faulty projections to Stout and moving to approve Stout’s resulting valuations, No. 23-1073 17

then communicating the approved price to the Plan participants. Id. at ¶ 952–53. The SAC therefore pleads “the identity of the person making the misrepresentation, the time, place, and content of the mis- representation, and the method by which the misrepresenta- tion was communicated to the plaintiff.” U.S. ex rel. Hanna v. City of Chicago, 834 F.3d 775, 779 (7th Cir. 2016) (quoting U.S. ex rel. Grenadyor v. Ukrainian Vill. Pharmacy, Inc., 772 F.3d 1102, 1106 (7th Cir. 2014)). That is all Rule 9(b) requires. Second, if the court dismissed these contentions because it thought Lyon failed to plead some other element of his claim with particularity, it was too quick to conclude that any such failure automatically warranted dismissal of those claims. Rule 8 allows plaintiffs to “plead statements of a claim … al- ternatively or hypothetically,” FED. R. CIV. P. 8(d)(2), and when “both fraudulent and nonfraudulent conduct violating the same statute or common law doctrine is alleged, only the first allegation can be dismissed under Rule 9(b),” Kennedy v. Venrock Associates, 348 F.3d 584, 493 (7th Cir. 2003). Thus, if a court thinks a plaintiff has failed to plead fraud, it must “dis- regard averments of fraud not meeting Rule 9(b)’s standard and then ask whether a claim has been stated.” Lone Star La- dies Inv. Club v. Schlotzsky’s, Inc., 238 F.3d 363, 368 (5th Cir. 2001). The district court correctly followed this rule as to Lyon’s claims against Argent, holding that those claims could be read as alleging imprudence rather than fraud. It held, however, that the claims against the directors and officers were “entirely premised on their knowing and self-motivated participation in fraud” and that “[t]here is no other way to read the allegations against” them. (Emphasis in original). 18 No. 23-1073

We do not read the complaint that narrowly. It is true that Lyon’s pre-2012 allegations, particularly as they relate to the 2001 sale of Appvion, rest entirely on a theory of fraud, but his post-2012 allegations are not so limited. For example, Count 28 of the SAC asserts that Board member Mark Suwyn knowingly allowed the ESOP Committee to approve inflated valuations so that his compensation would increase. That al- legation, a charge that Suwyn breached his duty of loyalty, sounds in fraud. Were we to excise the allegation for failure to meet Rule 9(b), however, Count 28 would survive as a claim for breach of the duty of prudence. It says that “[i]f Suwyn claims he did not know about these, and other, valuation ir- regularities and the related ESOP Committee and ESOP Trus- tee failures described herein, then he is admitting to a com- plete lack of prudence and loyalty for not knowing these basic, yet hugely material facts.” SAC at ¶ 1042. That “if-then” structure is the type of hypothetical pleading envisioned by Rule 8(d)(2). See Charles A. Wright & Arthur R. Miller, 5 Fed. Practice and Procedure § 1282 (4th ed. 2023). And Lyon’s hypo- thetical pleading—that the defendants failed to carry out their duties in a prudent way—comfortably qualifies as a violation of the duty of prudence under 29 U.S.C. § 1104(a)(1)(B). The same is true of the claims against the trustees; were we to ex- cise the fraud allegations against them, imprudence claims would remain. See, e.g., SAC at ¶ 591 (“In sum, Reliance failed to prudently investigate the value of PDC’s stock and there- fore breached its duty of prudence and loyalty under ERISA.”). Third, if the district court dismissed the claims for lack of plausibility, it raised the bar too high. Parts of the district court’s opinion indicate that it saw Rule 9(b) as imposing not just a heightened particularity standard, but a heightened No. 23-1073 19

plausibility standard as well. But Rule 9(b) does no such thing. All it does is call for more than a “short and plain statement” of the claim; it leaves the requirement of plausibility un- touched. See Loreley Financing No. 3 Ltd. v. Wells Fargo Securi- ties, LLC, 797 F.3d 160, 174 (2d Cir. 2015) (“While Rule 9(b) requires that the circumstances constituting fraud be stated with particularity, it does not require factual pleadings that demonstrate the probability of wrongdoing.” (cleaned up)). The Court acknowledged as much in Twombly; it dismissed the claims not because “the allegations in the complaint were insufficiently ‘particularized’” but because they were implau- sible, and it noted that its decision does not “broaden the scope of” Rule 9(b). 550 U.S. at 569 n.14. (In this respect, as we note below, ERISA and securities claims differ markedly; the latter must indeed satisfy a higher likelihood standard.) Applying Twombly, we find that Lyon adequately pleaded breach of the duty of loyalty by State Street, Reliance, and the ten post-2012 directors and officers. Lyon alleged that the di- rectors and officers had an incentive artificially to raise Ap- pvion’s price so that they would benefit from increased bo- nuses. They accomplished this goal by providing Stout with exaggerated projections and conspiring with it to exclude pension debt and to add a fraudulent control premium. Lyon alleged that State Street and Reliance had an incentive to keep the directors and officers happy in order to retain Appvion’s business, and so they looked the other way and accepted val- uations they knew to be inflated. Perhaps Lyon will have trou- ble proving those allegations, but this case comes to us on a motion to dismiss, and “the plausibility standard does not al- low a court to question or otherwise disregard nonconclusory factual allegations simply because they seem unlikely.” Fire- stone Financial Corp. v. Meyer, 796 F.3d 822, 827 (7th Cir. 2015); 20 No. 23-1073

see also Swanson v. Citibank N.A., 614 F.3d 400, 404 (7th Cir. 2010) (“‘Plausibility’ in this context does not imply that the district court should decide whose version to believe, or which version is more likely than not.”). Even if the fraud aspects of Lyon’s claim were implausible, Lyon has stated claims for breach of the duty of prudence. He pleaded that to the extent the directors, officers, and trustees did not intentionally inflate Appvion’s price, they were care- less in failing to scrutinize Stout’s valuation methods. We acknowledge at the outset that the fact that Appvion eventually failed is not proof of imprudence, Albert v. Oshkosh Corp., 47 F.4th 570, 579 (7th Cir. 2022), and that “rel[ying] on a financial advisor is evidence of prudence,” Fish v. GreatBanc Trust Co., 749 F.3d 671, 681 (7th Cir. 2014). But hiring an expert is not a complete shield against an imprudence claim. ERISA’s duty of prudence requires fiduciaries to “investigate the ex- pert’s qualifications, provide the expert with complete and ac- curate information, and make certain that reliance on the ex- pert’s advice is reasonably justified under the circumstances.” Keach v. U.S. Trust Co., 419 F.3d 626, 636–37 (7th Cir. 2005) (in- ternal quotation marks omitted) (quoting Howard v. Shay, 100 F.3d 1484, 1489 (9th Cir. 1996)). This duty “will necessarily be context specific.” Fifth Third Bancorp v. Dudenhoeffer, 573 U.S. 409, 425 (2014). A prudent fiduciary would more closely scru- tinize an expert’s advice if red flags indicated that the expert’s methods might be unsound. Lyon makes several allegations that, taken as true, indicate that a prudent fiduciary would have questioned Stout’s valu- ations. The most troubling allegation is that Stout used a con- trol premium in appraising the price of a share of Appvion. From 2012 to 2014, Stout added a 10% control premium to the No. 23-1073 21

company’s enterprise value to account for the Plan’s control- ling interest. The control premium had an outsized effect on Appvion’s appraised value. For example, in the December 2014 valuation, the control premium accounted for roughly 61% of the appraised value of the Plan’s equity in Appvion. That translated directly into much higher share prices for Plan participants. Yet these price increases were not justified by ac- tual control. State Street had long ago ceded to Appvion’s CEO most of the Plan participants’ power over the composi- tion of Appvion’s Board, and the Plan participants were al- lowed to vote only on extraordinary actions by the Board (such as a sale of the company). For all other matters, the trus- tee was required to vote the Plan’s shares as directed by the ESOP Committee. The control premium alone might be enough to support further proceedings in this case. But it did not stand alone: Lyon added several other allegations of imprudence to the mix. For instance, he asserted that the directors and officers consistently gave Stout overly rosy projections that caused Stout to overestimate Appvion’s future earnings. He also al- leged that Appvion’s leadership repeatedly tried and failed to sell Appvion, suggesting that the market did not agree with Stout’s valuations. Further, when Stout included Bemrose- Booth’s pension debt in calculating its fair market value, that value shot down to zero—indicating that the directors, offic- ers, and trustees should have ensured that Stout was account- ing for Appvion’s own, much larger pension debt. 1

1 Because the defendants’ actions related to BemroseBooth took place

outside of the repose period, Lyon cannot seek recovery for those actions. 22 No. 23-1073

We assume that, taken separately, any or all of these addi- tional red flags would not have been enough to put the de- fendants on alert. But our job is to read the complaint as a whole, not to parse it “piece by piece to determine whether each allegation, in isolation, is plausible.” Braden v. Wal-Mart Stores, Inc., 588 F.3d 585, 594 (8th Cir. 2009). Taking all of Lyon’s allegations together (with emphasis on the control pre- mium), we conclude that the complaint plausibly pleads that the directors, officers, and trustees breached their duty of pru- dence by failing to ensure that Stout’s methods were sound. Compare Howard, 100 F.3d at 1490 (finding a breach of a duty when the defendant failed to “make an honest, objective effort to read the valuation, understand it, and question the meth- ods and assumptions that do not make sense”), with Keach, 419 F.3d at 637 (finding no breach of duty when the defend- ants, “rather than blindly accept[ing]” the financial projec- tions of the company’s officers, “challenged a number of as- sumptions in [the appraiser’s] evaluation before developing their own independent evaluation”). Our finding of plausibility finds support in a recent—and quite similar—case from the Fourth Circuit. In Brundle v. Wil- mington Trust, N.A., 919 F.3d 763 (4th Cir. 2019), an ESOP par- ticipant alleged that the trustee breached its fiduciary duties by allowing the ESOP to overpay for the sponsor company’s shares. After a multi-day bench trial, the district court found in the participant’s favor. The trustee appealed, arguing in part that it justifiably relied on the valuation it received from an independent appraiser—none other than Stout Risius Ross. The Fourth Circuit affirmed. Of particular relevance

That does not prevent him, however, from pointing to BemroseBooth as a red flag that the defendants should have been aware of after 2012. No. 23-1073 23

here, the court noted that Stout had included a 10% control premium in its calculations even though the ESOP lacked “the power to appoint a majority of the [company’s] board, a key indicator of control.” Id. at 777. The court also faulted the trus- tee for failing to probe irregularities in the projections that the company’s management provided to Stout. The Fourth Cir- cuit concluded that the trustee therefore should have been aware of possible flaws in Stout’s methodology. Granted, the Brundle plaintiff also pointed to other red flags that Lyon does not allege, such as that “at least one [trustee] official knew of” a prior, much lower valuation of the company by another in- dependent appraiser. Id. at 774. But that should come as no surprise; unlike the Brundle plaintiff, Lyon has yet to engage in discovery. As this litigation progresses, Lyon might (or might not) find further indications of the defendants’ fraud or imprudence. We find the Fourth Circuit’s reasoning persuasive. It rein- forces our conclusion that Lyon has “present[ed] a story that holds together.” Swanson, 614 F.3d at 404. We naturally take no position on any future developments in the case. Lyon will still have to show that the defendants fraudulently or impru- dently misrepresented Appvion’s value. See Twombly, 550 U.S. at 556 (“[A] well-pleaded complaint may proceed even if it strikes a savvy judge that actual proof of those facts is im- probable, and ‘that a recovery is very remote and unlikely.’” (quoting Scheuer v. Rhodes, 416 U.S. 232, 326 (1974))). And, of course, the defendants will have the opportunity to prove that they acted prudently and in good faith, including by intro- ducing evidence that they did in fact scrutinize Stout’s valua- tion methods. 24 No. 23-1073

2. Prohibited Transactions Lyon’s arguments for allowing his prohibited-transaction claim against the post-2012 officers (SAC Count 25) to go for- ward are even stronger. The district court grouped this claim with the breach-of-duty claim against the directors and offic- ers, reasoning that although fraud is not an element of pro- hibited transactions, Lyon’s theory of the case depended on fraud. We disagree with that conclusion. For the reasons we outlined above, we do not read Lyon’s claim so narrowly. ERISA’s background rule forbids a plan fiduciary from causing the plan to purchase shares in the plan’s own com- pany. See 29 U.S.C. § 1106(a)(1)(E). Were that prohibition ab- solute, it would make all ESOPs illegal. The key is 29 U.S.C. § 1108(e), which allows a plan to buy the employer’s stock “for adequate consideration,” defined in 29 U.S.C. § 1002(18)(B) as the fair market value of the stock. But Lyon need not plead facts showing that the transaction was for more than fair market value. As we explained in Allen v. Great- Banc Trust Co., the exceptions in section 1108 “are affirmative defenses for pleading purposes, and so the plaintiff has no duty to negate any or all of them.” 835 F.3d 670, 676 (7th Cir. 2016). The officers argue that Lyon took the burden upon himself by alleging inadequate consideration in the SAC, but the same was true in Allen. See First Amended Complaint at ¶ 42, Allen v. GreatBanc Trust Co., No. 1:15-cv-3053 (N.D. Ill. Jan. 31, 2017) (“[T]he prohibited transaction did not meet the conditional exemption requirements of ERISA. … The facts indicate that [the trustee] did not perform a good faith valuation of the Per- sonal-Touch stock purchased by the Plan and that the Plan paid substantially more than fair market value for Personal- No. 23-1073 25

Touch stock.”). We did not ascribe any significance to the Al- len plaintiff’s choice to plead the absence of an exemption, and we will not do so here either. Allen mandates reversal. Whether Lyon contends in the end that the officers caused the Plan to enter the transaction out of fraud or imprudence, all he had to allege was “that the transaction was a prohibited one,” 835 F.3d at 675—or in other words, that the officers caused the Plan to purchase Ap- pvion’s shares, 29 U.S.C. § 1106(a)(1)(E). Lyon meets that re- quirement. He alleges that the officers in the ESOP Committee had (and exercised) the power to direct the trustees to pur- chase Appvion’s stock, and that the trustees then did pur- chase that stock. That is enough to save his claims from a mo- tion to dismiss. 3. Co-Fiduciary Liability That leaves Lyon’s claims for co-fiduciary liability against the directors, officers, State Street, and Reliance (SAC Counts 26–29). Because the district court dismissed all other ERISA claims, it dismissed these as well; one cannot be liable as a co- fiduciary without another fiduciary’s underlying violation. But, having reversed the district court’s dismissal of the post- 2012 ERISA claims for direct violations, we must now deter- mine whether that also means reviving the co-fiduciary claims. We conclude that it does. ERISA creates three avenues for imposing liability on a fi- duciary for her co-fiduciary’s violation of ERISA. Two of these require that the fiduciary knew of her co-fiduciary’s breach, and that the fiduciary either knowingly participated in or concealed that breach, 29 U.S.C. § 1105(a)(1), or that the fiduciary failed to make reasonable efforts to remedy the 26 No. 23-1073

breach, 29 U.S.C. § 1105(a)(3). In contrast, the third avenue al- lows for co-fiduciary liability when the fiduciary, by failing to comply with her duties under ERISA, enabled her co-fiduci- ary’s breach. 29 U.S.C. § 1105(a)(2). This provision does not require knowledge of the other fiduciary’s breach; all it re- quires is that the fiduciary’s breach of duty gave the co-fidu- ciary the opportunity to breach. See Free v. Briody, 732 F.2d 1331, 1336 (7th Cir. 1984) (“A fiduciary also is to be liable for the loss caused by the breach of fiduciary responsibility by another fiduciary of the plan if he enables the other fiduciary to commit a breach through his failure to exercise prudence.” (quoting H. R. Conf. Rep. No. 93-1280 (1974), reprinted in 1974 U.S.C.C.A.N. 5038, 5080)). The defendants argue that Lyon fails to allege that they knew of each other’s breaches. We need not decide if Lyon’s allegations of knowledge are sufficient, however, because the third avenue saves Lyon’s claims. Lyon has plausibly alleged liability under 29 U.S.C. § 1105(a)(2). The directors and offic- ers had the power to appoint or remove the trustees and the concomitant duty to monitor them. Lyon alleges that if the di- rectors and officers did not knowingly inflate Appvion’s price, then their failure to monitor or remove the trustees al- lowed the trustees to continue accepting Stout’s inflated val- uations (whether willfully or carelessly). The same is true in reverse. Lyon alleges that by approving Stout’s valuations without sufficient scrutiny, the trustees allowed the ESOP Committee to accept those valuations and to use them to au- thorize purchases of Appvion shares by the Plan. Cf. Free, 732 F.2d at 1336 (“ERISA does not make a trustee an insurer against a co-trustee’s misconduct, but the Act does require the trustee to use reasonable care to prevent such a breach.”). No. 23-1073 27

C. The July 2012 Valuation At this point, we have explained why we are affirming the dismissal of the pre-2012 ERISA claims as time-barred but al- lowing the post-2012 claims to go forward. That leaves one minor line-drawing issue regarding the valuation issued by Stout on July 16, 2012. Lyon argues that the part of his breach- of-duty claim against State Street (SAC Count 3) related to this valuation falls within the statute of repose because State Street used that valuation to purchase shares of Appvion in Decem- ber 2012. He presents several arguments to support that the- ory, none of which we find convincing. As we stated earlier, ERISA’s six-year statute of repose be- gins running on “the date of the last action which constituted a part of the breach or violation, or [ ] in the case of an omis- sion the latest date on which the fiduciary could have cured the breach or violation.” 29 U.S.C. § 1113(1). Lyon contends that the “last action” constituting the breach was State Street’s approval of stock purchases in December at the July-ap- proved price. To identify the relevant date, we must “consider[ ] the na- ture of the fiduciary duty” underlying the claim, Tibble v. Ed- ison International, 575 U.S. 523, 528 (2015). That requires us to “isolate and define the underlying violation upon which [the plaintiff’s] claim is founded.” Meagher v. International Ass’n of Machinists & Aerospace Workers Pension Plan, 856 F.2d 1418, 1422 (9th Cir. 1988). The complaint is clear: Lyon’s breach-of- duty claim against State Street is entirely about State Street’s approval of Willamette’s and Stout’s inflated valuations. Lyon brought a separate prohibited-transaction claim against State Street for its role in approving purchases by the Plan using those valuations (SAC Count 22), but the district court 28 No. 23-1073

dismissed that claim because State Street did not have discre- tionary authority over the purchases. Lyon did not appeal that decision, and he cannot now combine the claim he ap- pealed with the one he surrendered. For similar reasons, we are not convinced by Lyon’s at- tempt to fit his claim within the statute of repose’s language for omissions. To be sure, it is often easy to recharacterize an action as an omission (or vice versa); approving an inflated valuation can be painted as failing to approve an accurate val- uation or failing to correct an earlier inaccurate valuation. But again, Lyon’s argument is belied by his complaint; the breach- of-duty claim against State Street seeks to impose liability for what State Street did, not for what it did not do. Lyon’s other argument is that State Street’s breach of duty was not complete until its breach caused the Plan participants’ damages. Lyon is correct that the employees who purchased shares in December suffered no damage from the inflated July price—and therefore had no claim—until they made pur- chased shares at that price. Were section 1113(1) a statute of limitations, Lyon would have a point; statutes of limitations do not start to run until a claim accrues. See CalPERS, 582 U.S. at 504–05. But section 1113(1) is a statute of repose, id. at 507, and it therefore starts running not when the claim accrues, but “on ‘the date of the last culpable act or omission of the de- fendant,’” id. at 505 (quoting CTS Corp., 573 U.S. at 8). III. Securities Fraud Lyon also brought four federal securities-fraud claims. Three of these claims charge violations of section 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b), and SEC Rule 10b-5, 17 C.F.R. § 240.10b-5. They allege that Stout, No. 23-1073 29

Reliance, and the officers who sat on the ESOP Committee falsely represented the value of Paperweight stock through their roles in setting the price of that stock. The fourth claim is for control-person liability under 15 U.S.C. § 78t(a), alleging that the Board members controlled the ESOP Committee and are therefore liable for the four officers’ misconduct. Because securities-fraud claims are subject to a five-year statute of re- pose, 28 U.S.C. § 1658(b)(2), Lyon’s claims reach only conduct occurring on or after November 26, 2013. A. Rule 10b-5 Claims We first take up the 10b-5 claims. While the Court has rec- ognized private securities-fraud lawsuits as essential to en- forcing federal securities law, both it and Congress have acknowledged that “such lawsuits have also been subject to abuse.” Amgen Inc. v. Connecticut Retirement Plans & Trust Funds, 568 U.S. 455, 475 (2013). Both institutions therefore have erected barriers—including heightened pleading stand- ards—against meritless securities-fraud claims. A plaintiff bringing a claim under Rule 10b-5 must plead: “(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.” Pugh v. Tribune Co., 521 F.3d 686, 693 (7th Cir. 2008). The district court dismissed the claims solely based on Lyon’s failure adequately to plead scienter, and we can affirm on that basis without discussing the other elements. For most claims, including those under ERISA, we do not weigh competing inferences at the pleading stage. Securities- fraud actions are different. Under the Private Securities Liti- gation Reform Act (PSLRA), a plaintiff adequately pleads 30 No. 23-1073

scienter only if she “state[s] with particularity facts giving rise to a strong inference that the defendant acted with the re- quired state of mind.” 15 U.S.C. § 78u-4(b)(2)(A). In Tellabs, Inc. v. Makor Issues & Rights, Ltd., the Supreme Court inter- preted the phrase “strong inference” to require that “a reason- able person would deem the inference of scienter cogent and at least as compelling as any opposing inference one could draw from the facts alleged.” 551 U.S. 308, 324 (2007). We therefore must take the complaint as a whole and dis- miss unless the inference that the defendants acted with sci- enter—that is, with the intent to deceive or with reckless indif- ference as to the truth of their statements—is at least as com- pelling as the inference that the defendants merely made care- less mistakes. Pension Trust Fund for Operating Engineers v. Kohl’s Corp., 895 F.3d 933, 936–37 (7th Cir. 2018) (“Pension Trust”). We conclude that Lyon has not cleared that difficult hurdle with respect to any of the defendants. 1. Stout In SAC Count 35, Lyon alleges that Stout and two of its managing directors, Scott Levine and Aziz El-Tahch, contin- ued the fraud initiated by Willamette by giving Reliance in- flated valuations of Appvion. Though both Lyon and Stout devote considerable portions of their briefs to arguing the merits of Stout’s valuations, this case comes to us at the mo- tion-to-dismiss stage, and so we must assume that Stout’s val- uations overstated the value of Appvion. The only question before us is whether Lyon’s alleged explanation for that dis- crepancy (that Stout intentionally misrepresented Appvion’s value) is at least as likely as the alternative explanation (that Stout and its officers simply did a bad job). No. 23-1073 31

The biggest obstacle to Lyon’s claim against Stout is his failure to identify a motive. Stout received a flat fee for each of its valuations, without any bonus payments for using a par- ticular methodology or arriving at a particular valuation. We assume that Stout had an interest in continuing to receive those flat fees, and that it might have feared that Reliance would switch to a different appraiser if it stopped issuing in- flated valuations of Appvion. But we concluded, even before Tellabs, that continued business and flat fees, “standing alone,” typically “will not suffice to establish fraudulent in- tent.” Tricontinental Indus., Ltd., v. PricewaterhouseCoopers, LLP, 475 F.3d 824, 841 (7th Cir. 2007). All third-party contractors have an interest in retaining the business of their customers, and so any plaintiff could point to that interest to establish scienter. See id. Instead of trying to show that Stout, Levine, and El-Tahch had a motive beyond retaining Reliance’s business, Lyon seizes on language in Tellabs agreeing with our court’s conclu- sion “that the absence of a motive allegation is not fatal” to pleading scienter. 551 U.S. at 325 (quoting Makor Issues & Rights, Ltd. v. Tellabs, Inc., 437 F.3d 588, 601 (7th Cir 2006)). In that same paragraph, however, the Court told us that “the sig- nificance that can be ascribed to an allegation of motive, or lack thereof, depends on the entirety of the complaint.” Id. The context of Tellabs is informative. There, the securities- fraud claim was against the CEO of a company, and the plain- tiff did not allege that the CEO had sold any of his shares at the allegedly inflated price. Although the Court did not de- cide whether the plaintiff had adequately pleaded scienter, neither did it find the lack of a direct financial motive dispos- itive. Its reasoning is easy to follow. The CEO of a company has a strong incentive to make her company appear more 32 No. 23-1073

successful than it is, even if she does not directly benefit finan- cially by selling her shares in the company. That logic makes considerably less sense in our case. Stout is a global investment banker with many clients. While it likely has some incentive to be seen as cooperative by those clients, its “greatest asset is its reputation for honesty fol- lowed closely by its reputation for careful work.” DiLeo, 901 F.2d at 629. The flat fees it received from Reliance and Argent “could not approach the losses [it] would suffer from a per- ception that it would muffle a client’s fraud.” Id. As for Levine and El-Tahch, although their interests “may have diverged from the firm’s, covering up fraud and imposing large dam- ages on the partnership will bring a halt to the most promis- ing career.” Id. Stout and its managing directors “shared none of the gain from any fraud,” id., and Lyon’s theory would re- quire us to credit the possibility that Stout would risk follow- ing in the footsteps of Enron simply to keep one of its many clients happy. Without a showing of motive or additional facts tending to show intent, we cannot conclude that it is at least as likely that Stout acted with intent as that Stout (per- haps mistakenly) thought its methods were sound. 2. Reliance Lyon similarly failed to show motive in his claim against Reliance (SAC Count 31), which also received only flat fees from its trusteeship over the Plan’s assets. Also cutting against scienter is the fact that Reliance was less familiar with the al- legedly fraudulent valuations than Stout; Stout prepared the valuations, while Reliance only reviewed them. True, we noted in the context of Lyon’s breach-of-duty claims that se- curing an independent appraisal is not a blank check; Reli- ance may have had a fiduciary duty to scrutinize Stout’s No. 23-1073 33

valuation methods. See Keach, 419 F.3d at 637. But the failure to abide by a fiduciary duty is not enough to show scienter. See Pugh, 521 F.3d at 694. Without more to show that Reliance should have been aware of Stout’s alleged fraud, the com- plaint does not do enough “to show why [Reliance] was reck- less, rather than just negligent.” Pension Trust, 895 F.3d at 940. Lyon contends that Reliance should have been aware of the fraud because the excluded debt and control premium had such large effects on the valuations that Reliance must have noticed them. But it is one thing to show that Reliance knew what price Stout recommended for Appvion’s shares, and another thing entirely to show that Reliance knew those recommendations to be faulty. See Cornielsen v. Infinium Cap- ital Mgmt., LLC, 916 F.3d 589, 602 (7th Cir. 2019) (“[A] com- plaint fails to satisfy the PSLRA’s particularity requirements by making conclusory allegations of scienter derived from a defendant’s mere access to information.”). From Reliance’s perspective, Stout’s valuations just as eas- ily could have looked like business as usual. When Reliance took over as trustee in 2013, Stout had previously appraised a share of Appvion as being worth $17.55. Stout gave Reliance two valuations during the latter’s tenure: the June 30, 2013, valuation of $17.85, and the December 31, 2013, valuation of $16.25. Only in December 2014, after the end of Reliance’s trusteeship, did Stout’s valuations of Appvion start to plum- met. To assume that Reliance must have been aware of “red flags” in Appvion’s price because that price later dropped sig- nificantly relies too heavily on hindsight. Our cases repeat- edly have rejected this reasoning. Pugh, 521 F.3d at 694. 34 No. 23-1073

3. ESOP Committee Officers Lyon’s third Rule 10b-5 claim (SAC Count 33) is against the officers on the ESOP Committee. They were in charge of approving Stout’s valuations and causing the Plan to pur- chase more shares of Appvion. Unlike his claims against Stout and Reliance, Lyon supports the claim against the officers with a substantial allegation of motive: under various “incen- tive plans,” the officers received cash bonuses based on Ap- pvion’s appraised price. (Although that price dropped in al- most every valuation during the relevant period, Lyon alleges that the price would have dropped even more had Stout properly appraised Appvion’s value, and so the officers’ pay- outs would have been even lower in the absence of the alleged fraud.) Tellabs indicates that “personal financial gain may weigh heavily in favor of a scienter inference,” 551 U.S. at 325, which is especially true here because verifying Stout’s valua- tions was such an integral part of the ESOP Committee’s mis- sion, cf. Makor Issues & Rights, Ltd. v. Tellabs, Inc., 513 F.3d 702, 711 (7th Cir. 2008) (holding, on remand from Tellabs, that the plaintiffs sufficiently pleaded scienter because it was “exceed- ingly unlikely” that the CEO was “unaware of the problems of his company’s two major products”). Although the officers’ motive makes this a close question, we nonetheless conclude that Lyon has failed to clear the sci- enter bar. Our post-Tellabs cases generally have rejected the notion that the exercise of options and receipt of bonuses is enough to show scienter. See Plumbers and Pipefitters Local Un- ion 719 Pension Fund v. Zimmer Holdings, Inc., 679 F.3d 952, 956 (7th Cir. 2012) (“Plaintiffs contend that we should infer scien- ter because … top managers had an incentive to make [their company] look good in order to keep their jobs, improve their No. 23-1073 35

bonuses, and increase the value of their stock options. This is too generic to satisfy Tellabs.”); Pugh, 521 F.3d at 695 (“[S]tock sales must generally be unusual or suspicious to constitute circumstantial evidence of scienter.”). Lyon alleges that the officers had a motive not to question Stout’s valuations, but without some allegation that they knew those valuations to be inflated, he cannot show that fraud is a more likely inference than incompetence. And Lyon does not state facts showing that the officers knew the valua- tions were inflated; the SAC charges that they “knew about the irregularities with the valuations because they reviewed and approved the valuations,” but again, knowing what the valuations are is not the same as knowing they are inflated. Higginbotham v. Baxter International, Inc., 495 F.3d 753, 758 (7th Cir. 2007); see also Cornielsen, 916 F.3d at 602. Lyon’s pro- posed approach would have us find scienter any time a corpo- rate officer stood to gain from a high share price and ap- proved financial statements that later turned out to be in- flated. That would not be consistent with Tellabs. B. Control-Person Liability Control-person liability claims require an allegation that the defendant controlled someone who is liable under an- other provision of federal securities law. 15 U.S.C. § 78t(a). Be- cause we have affirmed the dismissal of all of Lyon’s other federal securities-fraud claims, Lyon’s control-person liability claim against the Board members (SAC Count 34) also falls. IV. State Law We wrap up with Lyon’s state-law theories, which have been carried forward from the FAC. The district court dis- missed all nine of them, mainly as preempted by ERISA. Lyon 36 No. 23-1073

now presses six on appeal, even though he did not replead them in the SAC. It may have been better practice to reassert them (if only to give notice of which claims he meant to pre- serve), but the district court would have dismissed those claims on the same grounds as before. That is why, when a court dismisses a claim without prejudice, a plaintiff is not re- quired to replead that claim in a subsequent complaint. See Bastian v. Petren Resources Corp., 892 F.2d 680, 682–83 (7th Cir. 1990). Houlihan argues that Lyon has waived the state-law theo- ries, but the cases on which he relies do not stand for that proposition. They involve plaintiffs who voluntarily removed claims by amending their complaints, without the district court first dismissing the claims. See, e.g., Prymer v. Ogden, 29 F.3d 1208, 1215 & n.6 (7th Cir. 1994). Unlike Bastian and the present case, those cases therefore did not involve “a previous ruling of the district judge adverse to the appellant.” 892 F.2d at 682. Those cases instead stand for the unsurprising propo- sition that because an amended complaint supersedes the original complaint, a plaintiff who deletes a claim from her complaint (without allowing the district court to rule on it first) waives that claim. Lubin v. Chicago Title & Trust Co., 260 F.2d 411, 413 (7th Cir. 1958). Here, there was no waiver, and so we proceed to this last set of arguments. A. Houlihan Lyon asserts three state-law theories against Houlihan: breach of fiduciary duty, fraud, and negligent misrepresenta- tion (FAC Counts 17–19). The district court held that the causes of action were preempted by ERISA and that, even if they were not, they failed on the merits. Lyon argues that they No. 23-1073 37

are not preempted, largely because Houlihan was not a fidu- ciary of the Plan. The problem with Lyon’s position is that his opening brief did not challenge the district court’s alternative holding on the merits of the claims. When a district court bases its ruling on two grounds and a plaintiff challenges only one on appeal, she “waive[s] any claim of error in that ruling.” Landstrom v. Illinois Dep’t of Children & Family Services, 892 F.2d 670, 678 (7th Cir. 1990). That is because even if we reversed the district court’s preemption holding as to the claims against Houlihan, its dismissal of those claims on the merits would remain un- touched. Lyon asks us to overlook his waiver. But our ability to overlook waiver in civil cases “is severely constricted,” and in most cases “a civil litigant ‘should be bound by his counsel’s actions.’” S.E.C. v. Yang, 795 F.3d 674, 679 (7th Cir. 2015) (quot- ing Deppe v. Tripp, 863 F.2d 1356, 1360 (7th Cir. 2010)). Lyon claims that this is not such a case, insisting that we can reach his argument under United States v. Blagojevich, 612 F.3d 558 (7th Cir. 2010). But Blagojevich is inapposite. There, the district court denied a third party’s motion to intervene on both pro- cedural and substantive grounds. The would-be intervenors appealed and challenged only the substantive ground. We noted that, under the usual rule, this “doomed their appeal, because if you lose for two independent reasons an appellate victory on one does not affect the judgment.” Id. at 560. But we also noted that the appellee’s brief “met the non-argument on the merits, and at oral argument” the appellee represented that it was not invoking the forfeiture doctrine. Id. We there- fore held that the appellee had “forfeited the benefit of appel- lants’ forfeiture.” Id. Not so here. Houlihan’s response brief 38 No. 23-1073

did argue the merits of the state-law theories against it, but it also argued that Lyon waived them. And at oral argument, Houlihan’s counsel made it clear that they were not giving Lyon the leeway that the Blagojevich appellee had given its op- posing counsel. We recognize the appeal of overlooking Lyon’s waiver. His opening brief had to address dozens of claims against five groups of defendants, and so some arguments may have slipped through the cracks. But while the waiver doctrine can seem harsh, there are also good reasons for enforcing it. Be- cause Lyon failed to raise all his arguments in his opening brief, he forced Houlihan to guess what claims it had to de- fend against. To cover all its bases, Houlihan dedicated over nine full pages of its brief to the ERISA claims against it, even though Lyon later stated in his reply brief that he had aban- doned those claims on appeal. The waiver doctrine exists in part to prevent that problem. Because Houlihan has unambig- uously invoked that doctrine, we must apply it. B. Directors and Officers Lyon also brought a state-law count against the directors and officers for breach of fiduciary duty (FAC Count 16). The district court dismissed this theory solely on grounds of ERISA preemption, and Lyon challenged that holding in his opening brief. We therefore may reach that issue. ERISA contains an expansive preemption clause; its pro- visions “supersede any and all State laws insofar as they may now or hereafter relate to any employee benefit plan” covered by ERISA. 29 U.S.C. § 1144. Of course, any case involving an ESOP might theoretically “relate to” an employee benefit plan. The Supreme Court, however, has not adopted such a No. 23-1073 39

broad interpretation, “caution[ing] against an ‘uncritical liter- alism’ that would make pre-emption turn on ‘infinite connec- tions.’” Egelhoff v. Egelhoff, 532 U.S. 141, 147 (2001) (quoting New York State Conf. of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U.S. 645, 656 (1995)). Instead, ERISA preempts a law if the law “has a connection with or reference to such a plan.” Shaw v. Delta Air Lines, Inc., 463 U.S. 85, 96–97 (1983). Wisconsin laws governing fiduciary duties are laws of general applicability, and so they do not in terms “reference” ERISA plans. See Rutledge v. Pharma. Care Mgmt. Assoc., 592 US. 80, 88 (2020) (“A law refers to ERISA if it acts immediately and exclusively upon ERISA plans or where the existence of ERISA plans is essential to the law’s operation.” (cleaned up) (quoting Gobeille v. Liberty Mut. Ins. Co., 577 U.S. 312, 319–20 (2016))). Whether a law has a connection with ERISA plans is usu- ally a difficult question, but another case arising from Ap- pvion’s bankruptcy gives us a helpful starting point. Recall that Paperweight, the company through which the Plan ac- quired Appvion, funded its purchase in part with contribu- tions from the retirement accounts of the Plan participants, but in much larger part with over $700 million in debt. In Halperin v. Richards, 7 F.4th 534 (7th Cir. 2021), Paperweight’s creditors brought claims for breaches of state-law duties against the directors and officers, as well as aiding-and-abet- ting claims against Stout and Argent. We held that ERISA preempted the claims against Stout and Argent, but not the claims against the directors and officers. Halperin dealt with the problem of dual-hat fiduciaries. ERISA’s exclusive-benefit rule requires that plan fiduciaries act solely in the interests of the plan’s participants. At the 40 No. 23-1073

same time, ERISA allows a company’s directors and officers to serve as plan fiduciaries. As a matter of corporate law, di- rectors and officers have a fiduciary duty to the corporation. Thus, a director or officer serving as an ERISA fiduciary owes “parallel but independent” duties to the plan’s participants and to the corporation. That explains why we found the claims in Halperin against the directors and officers not to be preempted by ERISA. Preempting those claims by the Halperin creditors—who could not sue under ERISA, see 29 U.S.C. § 1132(a)—would have left them without recourse for the (alleged) fraud by the directors and officers. This would have created a loophole, contrary to the goals of ERISA, for corporate malfeasance. Halperin expressly reserved the question presented here: whether “ERISA would preempt similar corporation-law claims brought by ERISA beneficiaries, participants, or fidu- ciaries who can sue Appvion’s directors and officers under ERISA for the same conduct.” 7 F.4th at 545. Perhaps ERISA would not preempt such claims if the directors and officers owed Appvion’s owners some state-law duty separate from the duties it owed them under ERISA, and perhaps the em- ployees could have sued over that duty as shareholders rather than as plan participants. But when asked at oral argument, Lyon was unable to point to any conduct that qualified as a breach of a state-law duty, but not as a breach of an ERISA duty, nor have we been able to identify any. On this record, Lyon’s state-law claims arise entirely from the Plan partici- pants’ ERISA-governed ownership of Appvion. The state-law duties that Lyon accuses the directors and officers of breach- ing are the same duties he charges them with breaching in his ERISA claims, and they are owed to the same people. That makes Lyon’s state-law claim against the directors and No. 23-1073 41

officers little more than an “alternative enforcement mecha- nism,” and alternative enforcement mechanisms have a con- nection with (and therefore relate to) ERISA plans. See Trus- tees of AFTRA Health Fund v. Biondi, 303 F.3d 765, 775 (7th Cir. 2002). C. Stout Lyon’s final state-law claims on appeal are against Stout for fraud and negligent misrepresentation (FAC Counts 10 and 12). Lyon did not waive his challenge to these claims, but they suffer the same fate as the claim against the directors and officers. ERISA allows the Secretary of Labor to sue non-fidu- ciaries such as Stout for knowingly participating in another fiduciary’s violation. See Harris Trust & Savings Bank v. Salo- mon Smith Barney, Inc., 530 U.S. 238, 248 (2000). We have re- served the question whether ERISA allows private parties to bring such suits as well. Halperin, 7 F.4th at 553 n.3. Even assuming that ERISA confers on Lyon a private right of action against Stout, however, he could seek only equitable relief, not damages. See Mertens v. Hewitt Associates, 508 U.S. 248, 255 (1993). But he has asked for damages. See FAC at 205. Thus, as in Halperin, “state-law liability for Stout could lead to a damages remedy that would arguably conflict with ERISA’s remedial limits on claims against non-fiduciaries.” 7 F.4th at 553; see also Pilot Life Ins. Co. v. Dedeaux, 481 U.S. 41, 54 (1987). Lyon’s claims against Stout therefore seem to us an “end run around ERISA’s more limited remedial scheme.” Halperin, 7 F.4th at 541. ERISA preempts such claims. V. Conclusion We AFFIRM the dismissal of FAC Counts 10, 12, and 16– 19 and SAC Counts 4–7, 11, 19–21, 31, and 33–35. We 42 No. 23-1073

REVERSE and REMAND SAC Counts 1, 3, 8–10, 12–18, and 25–29 to the extent they seek recovery for conduct taking place after November 26, 2012, and REMAND those claims to the district court for further proceedings consistent with this opinion.