F5 Capital v. Pappas

16-530-cv F5 Capital v. Pappas et al. UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT August Term, 2016 (Argued: December 6, 2016 Decided: April 26, 2017) Docket No. 16-530-cv F5 CAPITAL, a Cayman Islands Corporation, Plaintiff-Appellant, — v. — PETROS PAPPAS, MILENA MARIA PAPPAS, ROGER SCHMITZ, TOM SOFTELAND, SPYROS CAPRALOS, KOERT ERHERDT, RENEE KEMP, RAJATH SOURIE, EMILY STEPHENS, STELIOS ZAVVOS, OAKTREE VALUE OPPORTUNITIES FUND, L.P., OAKTREE OPPORTUNITIES FUND IX DELAWARE, L.P., OAKTREE CAPITAL MANAGEMENT, L.P., OAKTREE OPPORTUNITIES FUND IX (PARALLEL 2), L.P., MONARCH ALTERNATIVE SOLUTIONS MASTER FUND LTD., MONARCH CAPITAL MASTER PARTNERS II-A L.P., MONARCH CAPITAL MASTER PARTNER II L.P., MONARCH DEBT RECOVERY MASTER FUND, LTD., MONARCH OPPORTUNITIES MASTER FUND, LTD., P MONARCH RECOVERY LTD., STAR SYNERGY L.L.C., STAR OMAS L.L.C., OAKTREE OBC HOLDINGS L.L.C., OAKTREE DRY BULK HOLDINGS L.L.C., MILLENNIA L.L.C., MILLENNIA HOLDINGS L.L.C., MIRABEL SHIPHOLDING & INVEST LIMITED, MIRACH SHIPPING COMPANY LIMITED, HERON VENTURES LTD., OCEANBULK CARRIERS L.L.C., BLUESEA INVEST AND HOLDING LIMITED , MONARCH ALTERNATIVE CAPITAL LP, STAR BULK CARRIERS CORP., Defendants-Appellees, BLUESEA OCEANBULK SHIPPING L.L.C. , Defendant. Before: CALABRESI, RAGGI, and LYNCH, Circuit Judges. F5 Capital (“F5”) brought a shareholder derivative action on behalf of Star Bulk Carriers Corp. (“Star Bulk”), alleging that individual members of Star Bulk’s board and affiliated entities improperly exploited their control of the corporation in entering into three separate self-dealing transactions. F5’s complaint included four causes of action, three of which were derivative and one of which purported to be a direct class-action claim for wrongful equity dilution. F5 did not seek intracorporate remedies by making a pre-suit demand on Star Bulk’s board of directors. In dismissing F5’s complaint, the district court concluded that the dilution claim was properly derivative under Delaware law and that F5 failed to plead demand futility under Rule 23.1(b)(3)(B), Fed. R. Civ. P., as to any of the claims. For the reasons set forth in this opinion, we conclude that (1) F5’s dilution claim was properly derivative, not direct; (2) the district court had subject matter jurisdiction to adjudicate the non-class, derivative claims; and (3) F5 did not allege facts sufficient to excuse it from making a pre-suit demand. The judgment of the district court is therefore AFFIRMED. MARK C. RIFKIN (Benjamin Y. Kaufman, Michael Liskow, on the brief), Wolf Haldenstein Adler Freeman & Herz LLP, New York, NY, for plaintiff-appellant F5 Capital DAVID W. BROWN (Andrew J. Ehrlich, Gregory F. Laufer, on the brief), Paul, Weiss, Rifkind, Wharton & Garrison LLP, New York, NY, for Oatkree defendants-appellees. Tariq Mundiya, Matthew W. Edwards, Willkie Farr & Gallagher LLP, New York, NY, for Monarch defendants- appellees and defendant-appellee Roger Schmitz. Bruce G. Paulsen, Jeffrey M. Dine, Michael B. Weitman, Seward & Kissel LLP, New York, NY for Pappas 2 defendants-appellees and defendants-appellees Tom Softeland, Spyros Capralos, Koert Erhardt, Stelios Zavvos, Star Synergy LLC, Star Omas LLC, Millennia LLC, Millennia Holdings LLC, Mirabel Shipholding & Invest Limited, Mirach Shipping Company Limited, Heron Ventures Ltd., Oceanbulk Carriers LLC, and Bluesea Invest and Holding Limited GERARD E. LYNCH, Circuit Judge: F5 Capital (“F5”) brought this shareholder derivative action on behalf of Star Bulk Carriers Corp. (“Star Bulk”), alleging that individual members of Star Bulk’s board and affiliated entities improperly exploited their control over the corporation in executing three separate transactions. Those transactions, according to F5, were infected with self-dealing and were not undertaken to serve the corporation’s best interests. F5’s complaint included four causes of action, three of which were derivative and one of which purported to be a direct class-action claim for wrongful equity dilution. F5 did not seek intracorporate remedies by making a pre-suit demand on Star Bulk’s board of directors. In dismissing F5’s complaint, the district court concluded that the dilution claim was properly derivative under Delaware law and that F5 failed to plead demand futility under Rule 23.1(b)(3)(B), Fed. R. Civ. P., as to any of the claims. 3 For the reasons set forth in this opinion, we conclude that (1) F5’s dilution claim was properly derivative, not direct; (2) the district court had subject matter jurisdiction to adjudicate the non-class, derivative claims; and (3) F5 did not allege facts sufficient to excuse it from making a pre-suit demand. We therefore AFFIRM the judgment of the district court. BACKGROUND The following facts are taken from the complaint and we accept them as true for the purposes of this opinion. F5 is a Cayman Islands corporation that invests in international shipping companies. Star Bulk is a global shipping company that uses sea vessels to ship dry bulk cargos including iron ore, coal, and grains. Star Bulk is incorporated in the Marshall Islands and maintains its executive office in Athens, Greece. The owner of F5, Hsin Chi Su, was a minority shareholder in Star Bulk and served in management positions at Star Bulk until October 2008. After he and defendant Petros Pappas, another key player in Star Bulk’s management, had a falling out resulting from a business dispute, the defendants worked to exclude Su from a leadership role at Star Bulk through several self-dealing transactions that F5 claims harmed the corporation and its minority shareholders. 4 The allegedly offending transactions are as follows. First, Star Bulk acquired Oceanbulk Carriers LLC and its fleet of vessels in a merger (“Oceanbulk Merger”).1 Oceanbulk was a new company and, prior to the merger, it reported significant financial losses. F5 contends that the merger was an unwise business decision that allowed certain defendants to consolidate their control of Star Bulk to the detriment of the other shareholders.2 Specifically, the merger was “meant to reward the Pappas Defendants and their cohorts through increased shareholder control and new sweetheart management positions at Star Bulk.” Compl. ¶ 86. In consummating the merger, Star Bulk incurred $1.3 billion in debt and needed to raise an additional $614 million in capital. According to F5, those monetary commitments threatened Star Bulk’s financial health and risked other injuries to the minority shareholders.3 F5 voted against the merger, but 95.6% of 1 The merger had a complex structure that involved forming multiple subsidiaries and holding companies to effect the transaction. 2 The defendants include the nine members of Star Bulk’s board at the time F5 filed its complaint: Petros Pappas, Spyros Capralos, Tom Softeland, Koert Erhardt, Roger Schmitz, Renee Kemp, Rajath Sourie, Emily Stephens, and Stelios Zavvos, as well as various corporate entities with which some of those defendants are allegedly affiliated. The corporate defendants, and the groups into which the parties divide the various defendants, are identified at pp. 6-7 below. 3 In its brief, F5 claims that Star Bulk’s stock price has dropped precipitously since the merger. That fact was not in the complaint and, as far as we can tell, was 5 Star Bulk’s shareholders approved the transaction. Second, Star Bulk purchased 34 dry bulk vessels from Excel Maritime Carriers Ltd. (“Excel Transaction”), at what F5 claims was a dramatically inflated price. Because the Excel Transaction was not structured as a merger, Star Bulk’s board voted on the transaction, but its shareholders did not. Third and finally, F5 alleges on information and belief that Star Bulk entered into service contracts with entities affiliated with Pappas at three times the going rate for the ship maintenance services included in the contracts (“Service Contracts”). More specific facts concerning each transaction will be discussed as necessary below. A further introductory word about the parties in this action is warranted. The complaint names as defendants not only the nine members of Star Bulk’s board, but also several corporate and other entities with which certain of those defendants are affiliated. As the parties do, we divide those entities into three groups. The first group is the “Pappas Defendants,” which includes Petros not asserted until F5’s brief on appeal. Thus, as a formal matter, that fact is not part of the record before us. See Fed. R. App. P. 10(a). In any event, F5 attributes that drop in large part to the Oceanbulk Merger, but it does not explain what other factors may have contributed to Star Bulk’s apparent financial decline. 6 Pappas, his daughter Milena Pappas, and several entities that they own.4 See Compl. ¶¶ 42-47. The second group of defendants, the “Oaktree Defendants,” includes Oaktree Capital Management, L.P. and several related entities.5 See Compl. ¶¶ 27-33. Three of the individual defendants—Sourie, Kemp, and Stephens—were Oaktree employees who were appointed to Star Bulk’s board after the Oceanbulk Merger. The third group of defendants, the “Monarch Defendants,” includes Monarch Alternative Capital LP and affiliated entities.6 See Compl. ¶¶ 34-41. Schmitz, an individual defendant, is a Monarch employee. F5 originally filed its complaint in the Supreme Court of the State of New York in New York County. The complaint asserted the following causes of action: (1) a derivative claim against the individual defendants for breach of fiduciary duty; (2) a derivative claim against all other defendants for aiding and abetting 4 Those entities are: Millennia LLC, Millennia Holdings LLC, Mirabel Shipholding & Invest Limited, Mirach Shipping Company Limited, Bluesea Oceanbulk Shipping LLC, and Heron Ventures Limited. 5 The Oaktree-related entities are: Oaktree Value Opportunities Fund, L.P., Oaktree Opportunities Fund IX Delaware, L.P., Oaktree Opportunities Fund IX (Parallel 2) L.P., Oaktree OBC Holdings LLC, Oaktree Dry Bulk Holdings LLC. 6 The other Monarch entities are: Monarch Alternative Solutions Master Fund Limited, Monarch Capital Master Partners II-A L.P., Monarch Capital Master Partner II L.P., Monarch Debt Recovery Master Fund Limited, Monarch Opportunities Master Fund Limited, and P Monarch Recovery Limited. 7 the breach of fiduciary duty; (3) a derivative claim against the individual defendants for corporate waste; and (4) a direct class-action claim for equity dilution. The defendants timely removed to federal district court in the Southern District of New York. According to the notice of removal, there was federal jurisdiction over F5’s direct, class action dilution claim under the Class Action Fairness Act (“CAFA”), 28 U.S.C. § 1332(d), and supplemental jurisdiction over the non-class, state law claims, 28 U.S.C. § 1367(a). F5 did not contest removal. The defendants jointly moved to dismiss the complaint under Rule 23.1, Fed. R. Civ. P., for failure to plead demand futility. The district court granted the motion, dismissed the complaint with prejudice, and denied F5’s implied request for leave to amend the complaint.7 This appeal followed. DISCUSSION We review de novo the district court’s dismissal of a derivative action for failure to satisfy Rule 23.1, Fed. R. Civ. P., accepting all facts in the complaint as 7 Certain groups of defendants also filed individual motions to dismiss on other grounds, such as lack of personal jurisdiction. See Fed. R. Civ. P. 12(b)(2). Because the district court granted the joint motion in its entirety, it denied the other motions as moot. 8 true. Espinoza ex rel. JPMorgan Chase & Co. v. Dimon, 797 F.3d 229, 234-36, 239 (2d Cir.), certified question answered, 124 A.3d 33 (Del. 2015). The parties agree that Delaware law applies.8 See Chau v. Lewis, 771 F.3d 118, 126 (2d Cir. 2014). For organizational clarity, this opinion proceeds as follows. First, we discuss whether F5’s class action dilution claim is properly considered derivative or direct. We conclude that the dilution claim is properly considered derivative, and therefore Rule 23.1’s demand requirements apply. We next turn to the two issues of subject matter jurisdiction that arise in connection with the derivative claims. On those two issues, we hold (1) that the district court properly retained jurisdiction over the case after it became clear that CAFA, 28 U.S.C. § 1332(d)(2), could no longer anchor jurisdiction; and (2) that the exception to supplemental jurisdiction in certain diversity cases, 28 U.S.C. § 1367(b), did not preclude the district court from exercising jurisdiction over the derivative state-law claims. 8 As explained, Star Bulk is incorporated in the Marshall Islands. The Marshall Islands have adopted Delaware corporate law: “This Act shall be applied and construed to make the [corporate] laws of the Republic . . . uniform with the laws of the State of Delaware and other states of the United States of America with substantially similar legislative provisions. Insofar as it does not conflict with any other provision of this Act, the non-statutory law of the State of Delaware and of those other states of the United States of America . . . is hereby declared to be and is hereby adopted as the law of the Republic.” 52 Marsh. Is. Rev. Code Part I, § 13. 9 Finally, we conclude that F5 failed to plead demand futility as Rule 23.1 requires. We therefore affirm. I. The Class Action Dilution Claim is Derivative. F5 Capital brought a putatively direct class-action claim for dilution of its (and the potential class members’) ownership interest in Star Bulk. Specifically, F5 alleges that, “[a]s a result of the Oceanbulk Merger and the Excel Transaction . . . the interest of Plaintiff in Star Bulk has been diluted and the interest of the Defendants has been increased.” Compl. ¶ 141. Before the Oceanbulk Merger, F5 owned 1.4% of Star Bulk. After the merger, that interest diminished to 0.5%. F5’s share of Star Bulk then decreased to 0.36% after the Excel Transaction. At the same time, the Oaktree, Monarch, and Pappas defendants increased their collective ownership of Star Bulk from 43.9% before the Oceanbulk Merger to 80.2% after that transaction. In dismissing the complaint, the district court agreed with the defendants that the dilution claim is properly considered a derivative claim that belongs to Star Bulk, not to its shareholders. In order to determine whether a claim is derivative or direct, courts rely on the following two inquiries: (1) “[w]ho suffered the alleged harm—the corporation or the suing stockholder individually”; and (2) “who would receive 10 the benefit of the recovery or other remedy.” Tooley v. Donaldson, Lufkin & Jenrette, Inc., 845 A.2d 1031, 1035 (Del. 2004). “If the corporation alone, rather than the individual stockholder, suffered the alleged harm, the corporation alone is entitled to recover, and the claim in question is derivative.” Feldman v. Cutaia, 956 A.2d 644, 655 (Del. Ch. 2007), aff’d, 951 A.2d 727 (Del. 2008) (“Feldman I”).9 “Alternatively, if the stockholder suffered harm independent of any injury to the corporation that would entitle him to an individualized recovery, the cause of action is direct.” Id. Typically, a “claim for wrongful equity dilution is premised on the notion that the corporation, by issuing additional equity for insufficient consideration, made the complaining stockholder’s stake less valuable.” Feldman I, 956 A.2d at 655. Such claims are “not normally regarded as direct” under Delaware law. Feldman v. Cutaia, 951 A.2d 727, 732 (Del. 2008) (“Feldman II”) (internal quotation marks omitted). This is so because “the alleged injury is to the corporation” and it “falls upon all shareholders equally and falls only upon the individual 9 The Feldman case is a leading Delaware authority on the distinction between direct and derivative dilution claims. Both the Chancery Court opinion and the Supreme Court opinion affirming it are instructive, and we will have occasion to refer to both. 11 shareholder in relation to his proportionate share of stock as a result of the direct injury being done to the corporation.” Feldman I, 956 A.2d at 655 (internal quotation marks omitted). Moreover, “[m]ere claims of dilution . . . cannot convert a claim traditionally understood as derivative into a direct one” because “a corporation is free to enter into . . . numerous transactions, all of which may result legitimately in the dilution [of the stake of present equity holders].” Id. at 656 (internal quotation marks omitted). In certain instances, however, stockholders may bring a claim of equity dilution directly. Direct dilution claims are a “species of corporate overpayment” claim, and arise where: (1) [A] stockholder having majority or effective control causes the corporation to issue “excessive” shares of its stock in exchange for assets of the controlling stockholder that have a lesser value; and (2) the exchange causes an increase in the percentage of the outstanding shares owned by the controlling shareholder, and a corresponding decrease in the share percentage owned by the public (minority) shareholders. Gentile v. Rossette, 906 A.2d 91, 99-100 (Del. 2006) (emphasis added). When those two requirements are met, the minority “stockholders also have a separate, and direct, claim.” Id. at 100. “Because the shares representing the ‘overpayment’ 12 embody both economic value and voting power, the end result of this type of transaction is an improper transfer . . . of economic value and voting power from the public shareholders to the majority or controlling stockholder.” Id. Importantly, Gentile permits a direct claim for equity dilution only in the “limited circumstances involving controlling stockholders.” Feldman II, 951 A.2d at 729. A controlling stockholder exists in one of the following circumstances: someone (1) “owns more than 50% of the voting power,” or (2) “exercises control over the business and affairs of the corporation.” Feldman I, 956 A.2d at 657 (internal quotation marks omitted). See Kahn v. Lynch Commc’n Sys., Inc., 638 A.2d 1110, 1114 (Del. 1994). Other cases have relied on that test, observing that, “where a significant or controlling stockholder causes the corporation” to engage in an offending transaction, “a separate and distinct harm” may allow shareholders to “seek relief in a direct action.” Gatz v. Ponsoldt, 925 A.2d 1265, 1274 (Del. 2007). F5 asks us to conclude that the Pappas, Oaktree, and Monarch defendants collectively constitute a control group for purposes of the Oceanbulk Merger and Excel transactions.10 That is, F5 contends that we may combine those defendants’ 10 F5 has not argued that any of the three sets of defendants individually acted as controlling stockholders prior to the Oceanbulk Merger. Indeed, it cannot since none of the three owned more than about 20% of the shares. 13 individual interests in Star Bulk in considering whether there was a controlling shareholder. In general, defendants’ holdings may not be aggregated “to satisfy the control test absent a voting agreement among the group or a blood pact to act together.” Feldman I, 956 A.2d at 657 (internal quotation marks omitted). In other words, a direct action may lie “if the complaint pleads that a number of shareholders, each of whom individually cannot exert control over the corporation collectively form a control group and are connected in some legally significant way—e.g., by contract, common ownership, agreement, or some other arrangement—to work together toward a shared goal.” Carsanaro v. Bloodhound Techs., Inc., 65 A.3d 618, 659 (Del. Ch. 2013) (internal quotation marks and alterations omitted). Here, F5 has not alleged sufficient facts to make out a plausible claim that the three groups of defendants constituted a single control group. Of the pre- merger 43.9% ownership, Pappas owned 3.3%, Oaktree owned 19.6%, and Monarch owned 21%. After the merger, Pappas and his related entities increased their ownership share to 12.6% and Oaktree’s shot up to 61.3%. Monarch’s ownership interest, however, decreased to 7.4%. Although F5 does plead the existence of a voting agreement in which Monarch promised to vote in favor of 14 the Oceanbulk Merger, the dramatic decrease in Monarch’s ownership interest after the Oceanbulk Merger is inconsistent with the purported goal of the voting agreement: to “obtain unfettered control of Star Bulk” at the expense of the minority shareholders. Appellant Br. 25. The two shareholders whose interest increased, Pappas and Oaktree, held only about 20% of the shares before the merger, too little to control the outcome—hence the need to add Monarch’s substantial 21% stake to allege anything approximating a control group. It is implausible, however, to allege that Monarch committed with Pappas and Oaktree to expand their ownership stake while shrinking its own. (In fact, Monarch’s percentage ownership share declined in approximately the same proportion as F5’s.) Further, 95.6% of Star Bulk shareholders voted in favor of the merger,11 a fact that renders still more implausible F5’s claim that Monarch, Oaktree, and Pappas defendants colluded to take over the corporation to the other shareholders’ detriment.12 A shared interest in the successful 11 The district court took judicial notice of Star Bulk’s SEC filing for this fact. See Kramer v. Time Warner Inc., 937 F.2d 767, 773 (2d Cir. 1991). F5 does not dispute its accuracy or question the propriety of noticing it. 12 F5 does not alleged that any of the materials distributed to the shareholders in connection with the vote were deceptive in any way. Indeed, F5 includes in its complaint a selection of the press release announcing the Oceanbulk Merger in 15 consummation of the transaction is not sufficient to aggregate their shares to allege that Pappas, Oaktree, and Monarch formed a control group as necessary to plead a direct claim, nor does F5 plausibly allege the type of relationship that Delaware law requires to combine the defendants’ shares for this purpose. Moreover, even assuming arguendo that we can aggregate the defendants’ shares of Star Bulk, their collective interest is insufficient to satisfy the definition of a controlling stockholder. As explained, before the Oceanbulk Merger, the three sets of defendants collectively owned 43.9% of Star Bulk’s shares. That is plainly less than 50%, and F5 has not plausibly alleged that the group collectively “exercise[d] control over the business and affairs of the corporation.” Feldman I, 956 A.2d at 657 (internal quotation marks omitted). Thus, even if we aggregate the defendants’ ownership interests for purposes of assessing their control over the Oceanbulk Merger, F5 has not pled that they acted as a control group under Gentile. Finally, after the Excel Transaction, during which Oaktree certainly was a controlling stockholder, Oaktree’s stake in Star Bulk decreased from 61.3% to which Star Bulk discloses the fact that Oaktree and Pappas would increase their ownership in the corporation after the transaction. Specifically, the press release provides that: “Upon completion of the Transaction, the Oaktree Investors will own 61.3% of Star Bulk’s shares of common stock and the Pappas Investors will own 12.5% of Star Bulk’s common stock.” Compl. ¶ 53. 16 57.3%. Although the shareholders did not vote on the Excel Transaction, that decrease is inconsistent with a direct claim of equity dilution because the transaction did not improperly transfer voting power from a minority to a controlling stockholder. F5’s additional arguments to the contrary are not persuasive. F5 contends that, although the defendants collectively owned only 43.9% of Star Bulk, the defendants had effective control over the particular transactions at issue. The complaint does not so allege, however. As explained above, 95.6% of Star Bulk shares were voted in favor of the Oceanbulk Merger, and the complaint alleges no facts evidencing that the defendant shareholder groups dominated the operation of Star Bulk, or even exercised disproportionate influence over the Oceanbulk Merger. Further, in order to complete the merger, a majority of Star Bulk shares that were not affiliated with Oaktree or Pappas needed to, and did, vote in favor of the transaction. See Compl. ¶ 53. Those facts undermine F5’s claim that the defendants had sufficient control over the transaction to be deemed controlling shareholders under Gentile, even assuming that their ownership shares could be aggregated for purposes of determining the extent of their control of Star Bulk. 17 F5’s contentions concerning the Excel Transaction are significantly weaker even than its inadequate allegations concerning the Oceanbulk Merger. First, as explained, F5’s share of Star Bulk decreased from 0.5% to 0.36% as a result of the Excel Transaction. F5 concedes, however, that the defendants’ percentage ownership of Star Bulk did not correspondingly increase in connection with the Excel Transaction—indeed, Oaktree’s majority interest in Star Bulk decreased from 61.3% to 57.3%. To combat this problem, F5 claims that the defendants “improperly increase[d] their economic value through the transaction at the expense of” F5 and the unaffiliated shareholders. Appellant Br. 52. But that is not the circumstance that Gentile contemplates, since the Gentile framework requires “an increase in the percentage of the outstanding shares owned by the controlling stockholder, and a corresponding decrease in the share percentage owned by the public (minority) shareholders.” Gentile, 906 A.2d at 100 (emphasis added). In other words, there must be “an improper transfer—or expropriation—of economic value and voting power” for the claim to be treated as direct. Id. (emphasis added). The complaint does not identify such an impermissible transfer of voting power associated with the Excel Transaction. To the extent F5 alleges that the defendants obtained an unfair economic benefit by using 18 corporate funds for the transaction, it alleges a classically derivative claim. In sum, F5’s dilution claim does not fall into the small group of direct dilution claims that Gentile describes. It is therefore derivative and subject to the same demand requirements as F5’s other claims. See Fed. R. Civ. P. 23.1(b). Those requirements will be discussed in detail below, after we turn to two issues related to our subject matter jurisdiction. II. The District Court Had and Retained Subject Matter Jurisdiction Over All Claims in This Case. Before we can address the merits of F5’s derivative claims, we first must resolve two threshold questions concerning our subject matter jurisdiction. Although neither party has challenged our jurisdiction, “[f]ederal courts have a duty to inquire into their subject matter jurisdiction sua sponte, even when the parties do not contest the issue.” D’Amico Dry Ltd. v. Primera Mar. (Hellas) Ltd., 756 F.3d 151, 161 (2d Cir. 2014). To frame these issues, we note first that the complaint raises only issues of state law and that the parties are not completely diverse under 28 U.S.C. § 1332(a)(2). “[D]iversity is lacking within the meaning of [§ 1332(a)(2)] where the only parties are foreign entities, or where on one side there are citizens and aliens 19 and on the opposite side there are only aliens.” Universal Licensing Corp. v. Paola del Lungo S.p.A., 293 F.3d 579, 581 (2d Cir. 2002). F5 is a Cayman Islands corporation, many of the defendants are citizens of foreign jurisdictions, and at least one individual defendant is a citizen of the United States. Thus, the plaintiff is a foreign entity, and the defendants include citizens of both foreign countries and the United States. When the defendants removed the case from state court, they explained that CAFA, 28 U.S.C. § 1332(d)(2), which demands only minimal diversity, was the basis for subject matter jurisdiction. The defendants also argued that the district court could properly exercise supplemental jurisdiction over the derivative claims under 28 U.S.C. § 1367(a). CAFA’s role as the anchor of federal jurisdiction in this case raises two issues: (1) whether, now that the class action claim cannot proceed as such, the district court properly retained jurisdiction over the complaint, and (2) whether the district court could exercise supplemental jurisdiction over the non-class, state law derivative claims despite the restrictions in 28 U.S.C. § 1367(b). We conclude that the district court had—and retained—jurisdiction. 20 A. Subject Matter Jurisdiction Over the Derivative Dilution Claim We have now held, as did the district court, that the dilution claim may not proceed as a class action because the claim belongs to Star Bulk, not its shareholders. Thus, CAFA arguably no longer provides a jurisdictional anchor for F5’s complaint. We therefore must decide whether district courts may retain jurisdiction over state-law claims with minimally diverse parties where the class- action component of the complaint is dismissed after the case is removed to federal court. We conclude that they may. As a general matter, the Supreme Court has “consistently held that if jurisdiction exists at the time an action is commenced, such jurisdiction may not be divested by subsequent events.”13 Freeport-McMoRan, Inc. v. K N Energy, Inc., 498 U.S. 426, 428 (1991) (per curiam). At the time of removal, F5’s complaint contained a class-action claim that met CAFA’s other jurisdictional requirements, including a $5 million amount in controversy and minimal diversity. See 28 U.S.C. § 1332(d)(2); Purdue Pharma L.P. 13 There are, of course, exceptions to this general rule that do not apply here. For example, a complaint may become moot, thereby ending the case or controversy for Article III purposes, during the course of litigation. See Keepers, Inc. v. City of Milford, 807 F.3d 24, 44-45 (2d Cir. 2015). 21 v. Kentucky, 704 F.3d 208, 213-14 (2d Cir. 2013). It therefore follows that the later failure of the class claim did not divest the district court of subject matter jurisdiction because CAFA anchored jurisdiction at the time of removal. When faced with the analogous but analytically distinct question of whether “jurisdiction under CAFA is secure even though, after removal, the plaintiffs amended their complaint to eliminate the class allegations,” we have answered in the affirmative. In Touch Concepts, Inc. v. Cellco P’ship, 788 F.3d 98, 102 (2d Cir. 2015) (internal quotation marks omitted). We observed that, “for the purpose of analyzing statutory subject-matter jurisdiction, the Supreme Court has treated amended complaints in removal cases with flexibility.” Id. at 101. Thus, “when a defendant removes a case to federal court based on the presence of a federal claim, an amendment eliminating the original basis for federal jurisdiction generally does not defeat jurisdiction.” Id. (internal quotation marks omitted). Similarly, “in cases removed on the basis of diversity[,] the filing of a post-removal amended complaint that reduces the amount in controversy below the statutory threshold does not impair diversity jurisdiction.” Id. Accordingly, we concluded that the district court did not lose jurisdiction even though the plaintiff amended its complaint to eliminate the class claim after removal. Id. at 22 102. See also Cedar Lodge Plantation, L.L.C. v. CSHV Fairway View I, L.L.C., 768 F.3d 425, 426-27, 429 (5th Cir. 2014) (holding that CAFA jurisdiction is assessed at the time of removal, and that post-removal amendments to the complaint do not defeat jurisdiction). We recognize that post-removal amendments to complaints differ from a judicial determination that the purported class claim could not properly proceed as such under the substantive law that applies to that claim. Nevertheless, we note that other circuits addressing the more closely related question of whether a failure of class certification prevents district courts from retaining jurisdiction over an individual claim that was originally brought as a class action under CAFA have uniformly held that “federal jurisdiction under [CAFA] does not depend on certification.” Cunningham Charter Corp. v. Learjet, Inc., 592 F.3d 805, 806 (7th Cir. 2010). This is so because “jurisdictional facts are assessed at the time of removal,” and “post-removal events (including non-certification, de- certification, or severance) do not deprive federal courts of subject matter jurisdiction.” Vega v. T-Mobile USA, Inc., 564 F.3d 1256, 1268 n.12 (11th Cir. 2009); accord Metz v. Unizan Bank, 649 F.3d 492, 500 (6th Cir. 2011) (holding that “denial of class certification does not divest federal courts of jurisdiction” (internal 23 quotation marks omitted)); Buetow v. A.L.S. Enters., Inc., 650 F.3d 1178, 1182 n.2 (8th Cir. 2011) (same); United Steel, Paper & Forestry, Rubber, Mfg., Energy, Allied Indus. & Serv. Workers Int’l Union, AFL-CIO, CLC v. Shell Oil Co., 602 F.3d 1087, 1091-92 (9th Cir. 2010) (same, because “post-filing developments do not defeat jurisdiction if jurisdiction was properly invoked as of the time of filing”). Similarly, the Fifth Circuit has held that “federal jurisdiction is properly exercised over” individual cases that were severed from a class action complaint because “post-removal events do not oust CAFA jurisdiction.” Louisiana v. Am. Nat. Prop. Cas. Co., 746 F.3d 633, 639, 640 (5th Cir. 2014). Because jurisdictional facts are assessed at the time of removal, and because at that time the complaint here appeared to plead in good faith the class claim necessary for jurisdiction, the fact that the court subsequently determined that the case could not proceed as a class action under CAFA did not deprive it of subject matter jurisdiction.14 14 We do not suggest that any class action pleading—even one lacking a good faith basis in law and fact—can support the continued exercise of CAFA jurisdiction. Nor do we conclude that the district courts, on finding that a case cannot proceed as a class action, must adjudicate state law claims rather than remand them to state court. They can also, of course, dismiss them without prejudice for consideration in state courts. 24 B. Supplemental Jurisdiction Over Other Derivative Claims Next we must examine whether the district court properly exercised supplemental jurisdiction over the claims that F5 pled as derivative, non-class claims. In their notice of removal, defendants conceded that there is no independent basis for subject matter jurisdiction over F5’s derivative, non-class claims. Rather, citing 28 U.S.C. § 1367, the defendants argued that the court “has supplemental jurisdiction over all” claims over which CAFA did not provide federal jurisdiction. J.A. 34. In general, “in any civil action of which the district courts have original jurisdiction, the district courts shall have supplemental jurisdiction over all other claims that are so related to claims in the action . . . that they form part of the same case or controversy under Article III.” 28 U.S.C. § 1367(a). See United Mine Workers of Am. v. Gibbs, 383 U.S. 715, 725 (1966). That broad grant of supplemental jurisdiction is significantly curtailed, however, by 28 U.S.C. § 1367(b), which provides that: In any civil action of which the district courts have original jurisdiction founded solely on section 1332 of this title, the district courts shall not have supplemental jurisdiction under subsection (a) over claims by plaintiffs against persons made parties under Rule 14, 25 19, 20, or 24 of the Federal Rules of Civil Procedure, . . . when exercising supplemental jurisdiction over such claims would be inconsistent with the jurisdictional requirements of section 1332. The parties here are not completely diverse as 28 U.S.C. § 1332(a) requires for diversity cases that do not meet CAFA’s specifications. The question therefore arises whether § 1367(b) precluded the district court from exercising supplemental jurisdiction over the derivative non-class claims, because the case apparently satisfies that statute’s three criteria for exclusion. First, original jurisdiction was founded solely on CAFA, which is codified at § 1332(d), and thus jurisdiction over the case is “founded solely on section 1332“ of Title 28. Second, the non-class state law claims are asserted by the plaintiff against defendants joined either compulsorily or permissively under Rules 19 or 20 of the Federal Rules of Civil Procedure. Third, there is no independent basis for subject matter jurisdiction over the non-class claims because the parties are only minimally diverse, thus suggesting that exercising supplemental jurisdiction would be “inconsistent with the jurisdictional requirements of section 1332.” 28 U.S.C. § 1367(b). After identifying this potential problem with our jurisdiction, we ordered 26 supplemental briefing. Specifically, we asked the parties to “address[] the question of subject matter jurisdiction over plaintiff’s derivative non-class state law claims in light of the express language of 28 U.S.C. § 1367(b)” and cases interpreting that statute. The parties’ perfunctory submissions did not shed much light on the question that we posed. Nevertheless, after considering the issue and the parties’ submissions, we conclude that, on the facts of this case, the district court properly exercised supplemental jurisdiction over the derivative, non-class claims. 1. Background on § 1367(b) Congress enacted § 1367(b) in 1990 “in response to the Supreme Court’s decision in Finley v. United States, [490 U.S. 545 (1989)].” 13D C. Wright & A. Miller, Federal Practice and Procedure § 3567.2 (3d ed. Jan. 2017 Update) (“Wright & Miller”). In Finley, a California plaintiff sued the federal government under the Federal Tort Claims Act and, in the same case, brought state law claims against two California defendants. 490 U.S. at 546. The Supreme Court ruled that the federal courts lacked jurisdiction over the state law claims against the non- diverse parties because there was no independent basis for jurisdiction over those claims. Id. at 555-56. In enacting the supplemental jurisdiction statute, 27 Congress overruled Finley and “thus embrace[d] pendent parties jurisdiction in federal question cases.” Wright & Miller § 3567.2. Congress, however, did not want parties to “wreak havoc in diversity of citizenship cases” by “overrid[ing] the two statutory limitations” on diversity jurisdiction: the amount-in- controversy and complete diversity requirements. Id. “Thus, although Congress wanted . . . to enable parties to resolve in one action all of their disputes arising from the same core of facts and thereby to conserve judicial resources, it did not want plaintiffs to be able to plead a complaint craftily so as to force a nondiverse case into federal court.” United Capitol Ins. Co. v. Kapiloff, 155 F.3d 488, 493 (4th Cir. 1998). Section 1367(b) therefore eliminated supplemental jurisdiction over certain claims brought by plaintiffs, where exercising jurisdiction over those claims would result in a lack of complete diversity in the underlying dispute. The statute’s application only to plaintiffs “reflects Congress’ intent to prevent original plaintiffs—but not defendants or third parties—from circumventing the requirements of diversity.” Viacom Int’l, Inc. v. Kearney, 212 F.3d 721, 726-27 (2d Cir. 2000) (citing legislative history). Thus, “any effort by the plaintiff to use supplemental jurisdiction to join a non-diverse defendant under Rule 20 is 28 thwarted by § 1367(b).” Wright & Miller § 3567.2. In considering the meaning of the word “plaintiff” in § 1367(b), we have held that the statute refers to the original plaintiff in the action, but not to a defendant who brings a counterclaim or cross-claim, thereby becoming a third party plaintiff. Viacom Int’l, Inc., 212 F.3d at 726-27. So too have the other circuits that have considered the question. See Allstate Interiors & Exteriors, Inc. v. Stonestreet Constr., LLC, 730 F.3d 67, 73 (1st Cir. 2013); State Nat’l Ins. Co. Inc. v. Yates, 391 F.3d 577, 580 (5th Cir. 2004); Grimes v. Mazda N. Am. Operations, 355 F.3d 566, 572 (6th Cir. 2004). In a typical case that was originally filed in federal court, that reading serves Congress’s purpose because “defendants are involuntarily brought into court,” whereas a plaintiff is “master of [its] complaint.” Viacom Int’l, 212 F.3d at 727 (internal quotation marks omitted). That distinction has significant consequences, as discussed below, for a case in which the defendants assert federal jurisdiction via removal. Finally, § 1367(b) applies only in cases where § 1332 is the sole basis for federal court jurisdiction, including alienage cases under § 1332(a)(2). Franceskin v. Credit Suisse, 214 F.3d 253, 258 n.2 (2d Cir. 2000). Because CAFA is codified at § 1332(d), it initially appears that § 1367(b) would also apply to cases in which federal jurisdiction is predicated on CAFA. 29 2. “Contamination Theory” of Complete Diversity Under § 1367(b) The caselaw on § 1367(b) is sparse, and virtually nonexistent in cases that involve CAFA. In general, however, courts adhere to its requirements fairly strictly in order to preserve the complete diversity requirement. Along those lines, “the Supreme Court [has] articulated a ‘contamination theory’ governing the interaction of Sections 1332 and 1367.” Pa. Pub. Sch. Employees’ Ret. Sys. v. Morgan Stanley & Co., 772 F.3d 111, 118 (2d Cir. 2014) (discussing Exxon Mobil Corp. v. Allapattah Servs., Inc., 545 U.S. 546, 560, 562 (2005)). Under that theory, “the Court noted that . . . the view that the inclusion of a non-diverse party somehow contaminates every other claim in the complaint . . . can make some sense in the special context of the complete diversity requirement.” Id. (internal quotation marks omitted). Thus, the Supreme Court’s “expansive interpretation of § 1367[(a)] does not extend to additional parties whose presence defeats diversity.” Id. (internal quotation marks omitted). This is so because “[a] failure of diversity . . . contaminates the action . . . and takes away any justification for providing a federal forum.” Id. (internal quotation marks omitted). “It follows that a defect [in diversity] eliminates every claim in the action, including any jurisdictionally 30 proper action that might otherwise have anchored original jurisdiction, and removes the civil action from the purview of § 1367 altogether.” Id. (internal quotation marks omitted). Further, “it is clear that a diversity-destroying party joined after the action is underway may catalyze loss of jurisdiction.” Merrill Lynch & Co. Inc. v. Allegheny Energy, Inc., 500 F.3d 171, 179 (2d Cir. 2007). Thus, “the contamination theory furnishes limitations on joinder in certain circumstances that may well extend beyond the restrictions listed in § 1367(b).” Id. Otherwise, “[w]henever a claim is brought by one diverse party against another, § 1367 would authorize the court to hear, as part of the same case, claims brought by nondiverse parties.“ In re Lorazepam & Clorazepate Antitrust Litig., 631 F.3d 537, 541 (D.C. Cir. 2011).15 15 We are not persuaded by the parties’ argument that § 1367(b) applies only where non-diverse parties are joined after the litigation commences. In their view, § 1367(b) would not prevent the district court from exercising supplemental jurisdiction where, as here, the non-diverse parties were named in the original complaint. The parties are plainly incorrect. As the Fourth Circuit has observed, § 1367(b) “does not merely speak to the addition of parties. It also contains an exception for Rule 20, which authorizes permissive joinder of parties. And permissive joinder can certainly be utilized at the beginning of the action, not just for an ongoing diversity action.” Rosmer v. Pfizer Inc., 263 F.3d 110, 116 (4th Cir. 2001). As a practical matter, the application of § 1367(b) in cases where the parties are not diverse from the beginning receives little attention, since it appears self-evident that diversity jurisdiction does not exist in such cases. Nevertheless, a literal reading of § 1367(a) might suggest that the existence of one 31 Importantly, the Supreme Court held in Exxon that the “contamination theory” applies only to the complete diversity requirement and “makes little sense with respect to the amount-in-controversy requirement, which is meant to ensure that a dispute is sufficiently important to warrant federal-court attention.” Exxon, 545 U.S. at 562. In other words, the amount-in-controversy requirement is claim-specific, but the complete diversity requirement applies to the entire action. Wright & Miller § 3567.2. To put it yet another way, adding a non-diverse party to a case in which federal jurisdiction is predicated on diversity defeats federal jurisdiction over the entire case, but adding a claim seeking less than the jurisdictional amount in controversy to a diversity claim that does satisfy the amount-in-controversy requirement does not. We have explained that the distinction makes sense in light of the different purposes of each requirement. “The purpose of the amount-in-controversy requirement . . . is fulfilled by a claim with complete diversity would support the exercise of supplemental jurisdiction over other claims where non-diverse parties are present. Section 1367(b) makes clear that supplemental jurisdiction does not exist in such cases. It would completely undermine the purpose of § 1367(b) and the Supreme Court’s “contamination theory” to decline to apply its exclusion where the non-diverse parties were included in the initial complaint. Thus, we reject the parties’ suggestion that § 1367(b) only applies where parties are added after the action is commenced. 32 single claim of sufficient importance to warrant a federal forum and is not negated by additional, smaller claims.” Pa. Pub. Sch. Emps.’ Ret. Sys., 772 F.3d at 118 (internal quotation marks omitted). “A failure of diversity, on the other hand, . . . takes away any justification for providing a federal forum.” Id. (internal quotation marks omitted). Ultimately, therefore, “while the amount-in-controversy requirement is somewhat malleable, complete diversity of all parties is an absolute, bright-line prerequisite to federal subject matter jurisdiction.” Id. at 119. That is indisputably true in the typical case, where complete diversity is essential to getting the matter in federal court in the first instance. CAFA is an independent anchor of jurisdiction, however, that requires only minimal diversity in certain state-law cases that satisfy the requirements of a class action. 3. Application to CAFA “In order to determine the scope of supplemental jurisdiction authorized by § 1367 . . . we must examine the statute’s text in light of context, structure, and related statutory provisions.” Exxon, 545 U.S. at 558. Although it appears from a superficial reading of the statute that § 1367(b) would preclude the district court from exercising supplemental jurisdiction here, we conclude that the district 33 court properly adjudicated the non-class, derivative state law claims. CAFA was enacted in 2005, approximately 15 years after Congress passed the supplemental jurisdiction statute. CAFA itself “dramatically expanded federal jurisdiction over class actions.” Greenwich Fin. Servs. Distressed Mortg. Fund 3 LLC v. Countrywide Fin. Corp., 603 F.3d 23, 32 (2d Cir. 2010). It did so with the “primary objective” of “ensuring Federal court consideration of interstate [class action] cases of national importance.” Standard Fire Ins. Co. v. Knowles, —U.S.—, 133 S. Ct. 1345, 1350 (2013) (internal quotation marks omitted). By bringing large class actions within the jurisdiction of the federal courts, CAFA also sought to “curb perceived abuses of the class action device which . . . had often been used to litigate multi-state or even national class actions in state courts.” Shell Oil Co., 602 F.3d at 1090. In other words, CAFA embodies Congress’s judgment that complete diversity is not essential in class actions that meet its requirements, which are intended to cover class actions of sufficient importance. It would make little sense to apply the “contamination theory” of supplemental jurisdiction to a CAFA- eligible case even though the underlying action as a whole is sufficiently diverse to make jurisdiction proper under at least one provision of § 1332. Such a result 34 would not “enable parties to resolve in one action all of their disputes arising from the same core of facts and thereby to conserve judicial resources,” Kapiloff, 155 F.3d at 493, nor would it further § 1367(b)’s goal of preventing plaintiffs from “plead[ing] a complaint craftily so as to force a nondiverse case into federal court,” id. Indeed, other courts have observed that “CAFA was clearly designed to prevent plaintiffs from artificially structuring their suits to avoid federal jurisdiction.” Freeman v. Blue Ridge Paper Prods., Inc., 551 F.3d 405, 407 (6th Cir. 2008) (emphasis added). Legislative history supports that view: according to a Senate Report, CAFA “was necessary because the previous law ‘enable[d] lawyers to game the procedural rules and keep nationwide or multi-state class actions in state court.’” Id. at 408, quoting S. Rep. No. 109–14, at 4 (2005), as reprinted in 2005 U.S.C.C.A.N. 3, 5. CAFA thus made “it harder for plaintiffs’ counsel to . . . try[] to defeat diversity” to stay in state court. Id. (internal quotation marks omitted). CAFA’s purpose is therefore utterly at odds with the purpose of § 1367(b), which seeks to prevent plaintiffs from strategically using the law of supplemental jurisdiction to destroy the complete diversity requirement and get into federal court. Given that CAFA is the jurisdictional 35 anchor for the complaint and that the evident purpose of that statute was to expand federal jurisdiction over suits that meet its requirements, it would be inconsistent with that purpose to interpret § 1367(b) to keep the pendent claims out of federal court while allowing the class claims to proceed, where the anchor of jurisdiction plainly does not contemplate such a result. See S.E.C. v. Rosenthal, 650 F.3d 156, 162 (2d Cir. 2011) (discussing the canon of statutory construction that counsels avoiding an interpretation that would have absurd results). Moreover, in stripping away the complete diversity requirement, CAFA dramatically increased the amount-in-controversy requirement to $5 million and took other measures to ensure that only large class actions are entitled to a federal forum. See 28 U.S.C. § 1332(d)(6). It therefore shares a purpose with the amount-in-controversy requirement in ordinary diversity cases, which “is meant to ensure that a dispute is sufficiently important to warrant federal-court attention.” Exxon, 545 U.S. at 562. That purpose, according to the Supreme Court, does not implicate the “contamination theory” of supplemental jurisdiction. Id. The fact that this is a removal case also favors exercising supplemental jurisdiction here. As described at length above, the main purpose of § 1367(b) is to prevent opportunistic plaintiffs from thwarting the complete diversity 36 requirement by joining claims against non-diverse parties in the same action with claims against diverse parties. This case does not implicate that concern. Here, the plaintiff chose a state court forum, and it is the defendants who invoked CAFA to bring the case in federal court. Section 1367(b)’s clear focus is on plaintiffs’ manipulation of § 1332's requirements, a fact that the “repetition of the word ‘plaintiffs’ at several rule-citing junctures in subdivision (b) makes [] clear.” Viacom Int’l, Inc., 212 F.3d at 727 (internal quotation marks omitted). Indeed, as a general matter, the Supreme Court tends to treat statutory jurisdictional questions in removal cases with flexibility. See In Touch Concepts, Inc., 788 F.3d at 101. In light of Congress’s clear purpose in enacting § 1367(b), and mindful of the Supreme Court’s instruction that “[w]e must not give jurisdictional statutes a more expansive interpretation than their text warrants, but it is just as important not to adopt an artificial construction that is narrower than what the text provides,” Exxon, 545 U.S. at 558 (citation omitted), we conclude that it was proper for the district court to exercise supplemental jurisdiction over the derivative state law claims. 37 III. F5 Has Not Alleged Facts Sufficient to Excuse it From Making a Pre-Suit Demand on the Board. Rule 23.1 outlines the requirements that apply when “one or more shareholders or members of a corporation . . . bring a derivative action to enforce a right that the corporation or association may properly assert but has failed to enforce.” Fed. R. Civ. P. 23.1(a). Among other prerequisites, a derivative complaint must “state with particularity: (A) any effort by the plaintiff to obtain the desired action from the directors or comparable authority, and, if necessary, from the shareholders or members; and (B) the reasons for not obtaining the action or not making the effort.” Fed. R. Civ. P. 23.1(b)(3)(A)-(B). In other words, “a shareholder seeking to assert a claim on behalf of the corporation must first exhaust intracorporate remedies by making a demand on the directors to obtain the action desired.” Espinoza ex rel. JPMorgan Chase & Co., 797 F.3d at 234 (internal quotation marks omitted). “Although Rule 23.1 sets forth the pleading standard for federal court, the substance of the demand requirement is a function of state law—here, Delaware law.” Id. “[D]ismissals under Rule 23.1 are reviewed de novo.” Id. at 236. A shareholder plaintiff who does not make a demand on the board must 38 allege that demand is excused. “To plead demand excusal under [Delaware Chancery] Rule 23.1, a plaintiff in a derivative action must plead particularized facts creating a reasonable doubt that either (1) the directors are disinterested and independent or (2) the challenged transaction was otherwise the product of a valid exercise of business judgment.” Del. Cty. Empl. Ret. Fund v. Sanchez, 124 A.3d 1017, 1020 (Del. 2015) (internal quotation marks omitted).16 “Although there is a heightened burden under [Chancery] Rule 23.1 to plead particularized facts, when a motion to dismiss for failure to make a demand is made, all reasonable inferences from the pled facts must nonetheless be drawn in favor of the plaintiff in determining whether the plaintiff has met its burden.” Id. Here, no demand was made. Thus, we must determine whether F5 has pled facts that, if true, cast a reasonable doubt on whether (1) the board on whom F5 would have been required to make its pre-suit demand was disinterested and independent with respect to that demand; or (2) the transaction is entitled to the protections of the business judgment rule. 16 Rule 23.1(a) of the Delaware Chancery Court Rules outlines the pleading requirements for demand futility under Delaware law. It substantially parallels Federal Rule of Civil Procedure 23.1(b)(3). 39 A. The Impartiality of the Board The interestedness of the board is assessed “as of the time the complaint [was] filed.” Sandys v. Pincus, 152 A.3d 124, 128 (Del. 2016). “A director will be considered unable to act objectively with respect to a presuit demand if he or she is interested in the outcome of the litigation or is otherwise not independent.” Beam ex rel. Martha Stewart Living Omnimedia, Inc. v. Stewart, 845 A.2d 1040, 1049 (Del. 2004) (footnote omitted). “Most obviously, a plaintiff can show that a given director is personally interested in the outcome of the litigation, in that the director will personally benefit or suffer as a result of the lawsuit.” In re infoUSA, Inc. Shareholders Litig., 953 A.2d 963, 985 (Del. Ch. 2007). “A plaintiff may also challenge a director’s independence by . . . rais[ing] a reasonable inference that a given director is dominated through a close personal or familial relationship or through force of will, or is so beholden to an interested director that his or her discretion would be sterilized.” Id. (internal quotation marks omitted). Moreover, “directors can neither appear on both sides of a transaction nor expect to derive any personal financial benefit from it in the sense of self-dealing, as opposed to a benefit which devolves upon the corporation or all stockholders generally.” Aronson v. Lewis, 473 A.2d 805, 812 (Del. 1984). In connection with this inquiry, 40 the “key principle . . . is that the directors are entitled to a presumption that they were faithful to their fiduciary duties. In the context of presuit demand, the burden is upon the plaintiff in a derivative action to overcome that presumption.” Beam ex rel. Martha Stewart Living Omnimedia, Inc., 845 A.2d at 1048-49 (emphasis in original, footnotes omitted). Under Delaware law, the interestedness of the board must be assessed in a “detailed, fact-intensive, director-by-director analysis.” In re infoUSA, 953 A.2d at 985; see Brehm v. Eisner, 746 A.2d 244, 254 (Del. 2000). “Delaware law does not permit the wholesale imputation of one director’s knowledge to every other for demand excusal purposes. Rather, a derivative complaint must plead facts specific to each director, demonstrating that at least half of them could not have exercised disinterested business judgment in responding to a demand.” Desimone v. Barrows, 924 A.2d 908, 943 (Del. Ch. 2007) (emphasis in original, footnotes omitted). Even so, the court must “consider all the particularized facts pled by the plaintiffs about the relationships between the director and the interested party in their totality and not in isolation from each other.” Del. Cty. Empl. Ret. Fund, 124 A.3d at 1019. Finally, “[a]t the pleading stage, a lack of independence turns on whether 41 the plaintiffs have pled facts from which the director’s ability to act impartially on a matter important to the interested party can be doubted because that director may feel either subject to the interested party’s dominion or beholden to that interested party.” Sandys, 152 A.3d at 128 (internal quotation marks and footnotes omitted). Ultimately, we view the “pled facts . . . in full context in making the . . . pleading stage determination of independence.” Id. (internal quotation marks omitted). At the time that F5’s complaint was filed, there were nine members of Star Bulk’s board: Petros Pappas, Roger Schmitz, Tom Softeland, Spyros Capralos, Koert Erhardt, Renee Kemp, Rajath Sourie, Stelios Zavvos, and Emily Stephens. In order to plead demand futility because of the board’s interestedness in the each challenged transaction, therefore, F5 must allege particularized facts that raise a reasonable doubt concerning the impartiality of five of the board members. We assume that Pappas was conflicted with respect to each transaction. We conclude, however, that, at a minimum, Zavvos, Kemp, Sourie, Stephens, and Capralos would have been able to consider a pre-suit demand 42 impartially.17 Thus, F5 has failed to plead that demand would have been futile. 1. Oceanbulk Merger The Oceanbulk Merger is the complaint’s principal focus. As explained, the Oceanbulk Merger was a complex transaction in which Star Bulk acquired Oceanbulk and its fleet of dry bulk vessels. According to F5, Oceanbulk lost almost $20 million in the 18 months prior to the merger, making it a poor investment for Star Bulk that did not have a legitimate business purpose. Although not directly relevant to discerning the five board members’ impartiality, we note that the Oceanbulk Merger was evaluated and approved by a special committee. That two-person committee included defendants Softeland and Schmitz, and F5 alleges that Schmitz was “hopelessly conflicted” because of “his relationship with Monarch, a large shareholder, and the fact that the Oceanbulk Merger was his idea.” Compl. ¶ 79 (emphasis omitted). That committee retained Evercore Group LLC (“Evercore”) to issue a fairness opinion concerning the merger. According to F5, Evercore’s opinion was inadequate 17 We do not mean to imply that F5’s allegations concerning any of the other directors are sufficient to call their impartiality into question. It may be that some of the other board members could also have voted impartially on F5’s pre-suit demand. Since we have found that a majority of the board is impartial, however, we need not address the impartiality of the others. 43 because it failed to analyze various costs and potential pitfalls of the merger in sufficient detail. F5 further claims that the Evercore opinion was compromised because Evercore would receive $2.95 million if the merger was successful, but only $800,000 if the transaction failed. F5’s allegations against the five directors who, in our view, would be impartial in considering a pre-suit demand are as follows. First, F5 alleges that Zavvos served on the board of Zeus Capital Partners, L.P., a private equity real estate fund, with Pappas. Zavvos and Pappas also serve together on the board of West Invest AG, another private equity fund. According to F5, Zavvos’s “substantial business entanglements with Pappas” mean that he cannot “impartially consider a demand on the Board.” App. Reply Br. 10. Its allegations concerning the Zavvos-Pappas relationship are plainly insufficient, especially in light of the requirement to plead particularized facts, because under Delaware law a “mere personal friendship or . . . outside business relationship” does not “raise a reasonable doubt about a director’s independence.” Beam ex rel. Martha Stewart Living Omnimedia, Inc., 845 A.2d at 1050. Zavvos’s alleged business relationship with Pappas is entirely unlike relationships that the Delaware courts have found sufficiently close to raise a 44 reasonable doubt concerning director independence. For example, the Delaware Supreme Court recently held that co-ownership of a private plane with an admittedly conflicted director was sufficient to raise a reasonable doubt concerning the independence of a director. Co-ownership of the plane was an “unusual fact” that “signaled an extremely close, personal bond between” the two directors “and between their families.” Sandys, 152 A.3d at 130. Those circumstances were “suggestive of the type of very close personal relationship that, like family ties, one would expect to heavily influence a human’s ability to exercise impartial judgment.” Id. F5 does not allege such a bond between Zavvos and Pappas here; it merely pleads that they served together on two other boards of directors and includes no particularized allegations concerning the details of their business or personal relationship. F5 has therefore failed to raise a reasonable doubt concerning Zavvos’s independence. Next, F5 contends that Sourie, Kemp, and Stephens could not impartially consider a pre-suit demand because Oaktree designated them to Star Bulk’s board after the merger, and Oaktree itself had a disproportionate interest in the Oceanbulk Merger because the merger would cause its share of Star Bulk to increase dramatically. Moreover, all three of the board members used to work for 45 Oaktree, spending years in management positions there. To show demand futility, in general “it is not enough to charge that a director was nominated by or elected at the behest of those controlling the outcome of a corporate election.” Aronson, 473 A.2d at 816. Thus, “in the demand-futil[ity] context a plaintiff charging domination and control of one or more directors must allege particularized facts manifesting a direction of corporate conduct in such a way as to comport with the wishes or interests of the corporation (or persons) doing the controlling.” Id. (internal quotation marks and citation omitted). “[T]he mere fact that [Oaktree], an alleged controlling stockholder, played some role in the nomination process should not, without additional evidence, automatically foreclose a director’s potential independence.” Teamsters Union 25 Health Servs. & Ins. Plan v. Baiera, 119 A.3d 44, 60 (Del. Ch. 2015) (internal quotation marks omitted). That is so because “people normally get appointed to boards through personal contacts.” Benihana of Tokyo, Inc. v. Benihana, Inc., 891 A.2d 150, 177 (Del. Ch. 2005), aff’d, 906 A.2d 114 (Del. 2006). Ultimately, “[i]t is the care, attention and sense of individual responsibility to the performance of one’s duties, not the method of election, that generally touches on independence.” Aronson, 473 A.2d at 816. Moreover, although Sourie, Kemp, and Stephens maintained their 46 positions as Oaktree employees, the pleadings do not allege that they maintained dual status as fiduciaries for both Oaktree and Star Bulk. Cf. Carsanaro, 65 A.3d at 638 (deeming three directors not independent for purposes of demand because of their status as dual fiduciaries as directors of companies on both sides of a transaction). F5 has not pled any particularized facts concerning the three Oaktree-appointed directors’ conduct in their positions or the nature of their obligations to Oaktree that raise a reasonable doubt concerning their independence. Finally, F5 has not raised a reasonable doubt concerning Capralos’s impartiality. F5 only alleges that Capralos was Star Bulk’s CEO and Director, but that after the merger he became a non-executive chairman of the board and received 168,842 shares of Star Bulk as a change-of-control payment. In response, defendants highlight complaint allegations indicating that the shares were granted pursuant to pre-existing consultancy and employment agreements. Absent a suggestion that the compensation was excessive, the receipt of such payment does not disqualify Capralos from passing on the pre-suit demand. See Orman v. Cullman, 794 A.2d 5, 29 n.62 (Del. Ch. 2002). F5’s attempt to characterize the payment as an undue personal financial benefit is unavailing and 47 unsupported by particularized facts. Thus, at least a majority of the board was disinterested with respect to the Oceanbulk Merger. 2. Excel Transaction F5’s allegations concerning five board members’ partiality with respect to the Excel Transaction are even less compelling. As explained, in the Excel Transaction, Star Bulk purchased 34 dry bulk vessels from Excel, a bankrupt company. In exchange for the ships, Star Bulk transferred 29.917 million shares of common stock and $288.39 million in cash to Excel, yielding a total purchase price of $634.91 million. Defendants and their affiliated entities also lent Star Bulk approximately $213 million to purchase the ships, a fact that F5 claims produced an obvious conflict. Star Bulk’s board voted on the transaction, but its shareholders did not. As with the Oceanbulk Merger, F5 complains that Sourie, Stephens, and Kemp were beholden to Oaktree, which benefitted from the Excel Transaction because it was one of Excel’s largest secured lenders and therefore held a claim against the company in connection with its bankruptcy reorganization plan. F5 also alleges that Oaktree orchestrated the Excel Transaction in a way that would increase Oaktree’s share of Star Bulk. Again, the mere fact of the three directors’ 48 appointment by Oaktree is not sufficient to show demand futility.18 F5 does not allege any facts that raise a reasonable doubt concerning the impartiality of Zavvos and Capralos. Therefore, the five directors that we identified were impartial with respect to the Excel Transaction. F5’s only other allegations concerning the Excel Transaction are that unidentified entities associated with Monarch, which Schmitz manages, lent money to Star Bulk to pay for the Excel ships, thus conflicting Schmitz out of the pre-suit demand concerning the transaction. Maybe so, but even assuming that Schmitz was not impartial as to this transaction, F5 has not pled particularized facts that raise a reasonable doubt about the impartiality of a majority of Star Bulk’s board. Accordingly, F5 has not pled demand futility with respect to the Excel Transaction based on a lack of director independence. 3. Service Maintenance Contracts F5 alleges on information and belief that Star Bulk overpays for ship maintenance service contracts. Specifically, Star Bulk hired Pappas-related 18 In its brief, the defendants observe that the three Oaktree directors did not vote on the Excel Transaction. We do not know the reason for their recusal, and since F5 does not allege any facts concerning their decision not to vote on the Excel Transaction in its complaint, we decline to draw any inferences from that fact about their ability to assess a demand challenging the transaction. 49 entities to provide ship maintenance for $750 per ship per day, where the standard industry cost for such services is only $250 per ship per day. Because of the size of Star Bulk’s fleet, those contracts apparently cost the company approximately $28 million per year. Although those allegations raise some questions, F5 does not allege that any director other than Pappas was conflicted in entering into the service contracts. Thus, as with the other transactions, the plaintiff has failed to plead particularized facts that raise a reasonable doubt concerning the interestedness of a majority of board members. B. The Business Judgment Rule The other way that derivative plaintiffs can plead demand excusal is by alleging “particularized facts creating a reasonable doubt that . . . the challenged transaction was otherwise the product of a valid exercise of business judgment.” Delaware Cty. Employees Ret. Fund, 124 A.3d at 1020 (internal quotation marks omitted). The business judgment rule is the “presumption that in making a business decision the directors of a corporation acted on an informed basis, in good faith and in the honest belief that the action was taken in the best interests of the company.” Gantler v. Stephens, 965 A.2d 695, 705-06 (Del. 2009) (internal quotation marks omitted). That presumption “can be rebutted if the plaintiff 50 shows that the directors breached their fiduciary duty of care or of loyalty or acted in bad faith.” In re Walt Disney Co. Derivative Litig., 906 A.2d 27, 52 (Del. 2006). The “business judgment rule is [therefore] process oriented and informed by a deep respect for all good faith board decisions.” In re Citigroup Inc. S’holder Derivative Litig., 964 A.2d 106, 122 (Del. Ch. 2009) (emphasis in original). The rule requires substantial deference, and ultimately “protects decisions unless they cannot be attributed to any rational business purpose.” In re infoUSA, 953 A.2d at 972 (internal quotation marks omitted). To excuse demand under the business judgment rule prong of the Aronson test, the “plaintiffs must plead particularized facts sufficient to raise (1) a reason to doubt that the action was taken honestly and in good faith or (2) a reason to doubt that the board was adequately informed in making the decision.” In re J.P. Morgan Chase & Co. S’holder Litig., 906 A.2d 808, 824 (Del. Ch. 2005), aff’d, 906 A.2d 766 (Del. 2006) (internal quotation marks omitted); see In re infoUSA, 953 A.2d at 972. F5 does not give the business judgment rule prong of the demand futility test much attention until its reply brief. But even the arguments presented there fail to persuade us that there is reason to doubt that the challenged transactions 51 are entitled to the protections of the business judgment rule. Thus, demand was not excused. 1. Oceanbulk Merger F5 principally alleges that the Oceanbulk Merger was a bad business decision because Oceanbulk was in financial trouble at the time the transaction took place. F5 also argues that the transaction was a poor business decision because Oceanbulk was a troubled company, and that Star Bulk had to take on a lot of additional debt and raise significant capital to purchase it. Although we assume Pappas’s personal interest in the transaction, the allegations are insufficient to show that the other directors acted in bad faith or failed to inform themselves about the transaction before voting in its favor. The committee reviewing the transaction commissioned an expert report, undermining any attempt by F5 to characterize the board as uninformed about the merger. Although F5 contends that the report was flawed, those flaws do not raise a reasonable doubt that F5 will be able to rebut the presumptive protections of the business judgment rule. Mere disagreement with the merits of the transaction is insufficient for us to doubt that the “decision [could] be attributed to any rational business purpose.” Gantler, 965 A.2d at 706 (internal quotation marks omitted). 52 2. Excel Transaction Again, F5 does not advance any specific arguments demonstrating that the Excel Transaction was not the product of valid business judgment. There is no indication that Star Bulk overpaid for the ships, and the mere fact that Star Bulk had to borrow money to buy the vessels is not on its own suspicious. In its reply, F5 contends that incurring $213 million in debt to buy the ships demonstrates that the transaction was flawed, arguing that because Excel was bankrupt, Star Bulk should have gotten the ships at a discount. Based on F5’s complaint, however, we do not know whether Star Bulk did in fact receive a discount on the ships because F5 does not provide any pricing benchmark by which to assess the deal. More importantly for purposes of the business judgment rule, F5 does not allege any facts suggesting that the board entered into the Excel Transaction in bad faith or without sufficient information about its terms. Thus, the complaint does not raise a reasonable doubt that the Excel Transaction is entitled to the protection of the business judgment rule.19 19 F5 also brings a derivative claim alleging that the Oceanbulk Merger and the Excel Transaction constituted actionable corporate waste. The complaint does not mention the service contracts as a basis for the waste claim. The district court dismissed the corporate waste claim because F5 did not allege sufficient facts to overcome the business judgment rule’s presumption of validity, which is 53 3. Service Maintenance Contracts The service contracts are somewhat suspect, but F5 has not alleged particularized facts raising a reasonable doubt concerning the applicability of the business judgment rule. Taking the complaint as true, a Pappas-controlled entity is receiving three times the normal, market-rate payment for its ship maintenance services. That fact alone, however does not rebut the presumption of valid business judgment. F5 has not alleged that the majority of the board acted in bad faith with respect to those contracts or that the board was inadequately informed before contracting with the service providers, nor does F5 suggest in its briefing how it could amend the complaint to allege facts that would make such a conclusion plausible. In the context of demand futility, F5’s allegations concerning the service contracts are insufficiently particularized to satisfy Rule 23.1. especially strong in the corporate waste context. In order to make out a claim for corporate waste, a plaintiff must “shoulder the burden of proving that the exchange was so one sided that no business person of ordinary, sound judgment could conclude that the corporation has received adequate consideration.” In re Walt Disney Co. Derivative Litigation, 906 A.2d at 74 (internal quotation marks omitted). F5 must therefore plead that the “challenged transaction served no corporate purpose” or that “the corporation received no consideration at all.” White v. Panic, 783 A.2d 543, 554 (Del. 2001). As explained, Star Bulk received substantial consideration for the Oceanbulk Merger and the Excel Transaction, and there is no reason to doubt that the business judgment rule protects both transactions from a claim of corporate waste. 54 IV. The District Court Properly Denied Leave to Amend. The district court denied F5’s implied request to amend the complaint because an amendment would be futile.20 Rule 15(a)(2) allows a party to amend its pleading with the court’s leave, which should be “freely give[n] . . . when justice so requires.” “Leave to amend may properly be denied if the amendment would be futile,” however. Krys v. Pigott, 749 F.3d 117, 134 (2d Cir. 2014) (citation omitted). “A proposed amendment to a complaint is futile when it could not withstand a motion to dismiss.” Balintulo v. Ford Motor Co., 796 F.3d 160, 164-65 (2d Cir. 2015) (internal quotation marks omitted). “[W]hen denial of leave to file a revised pleading is based on a legal interpretation, such as futility, a reviewing court conducts a de novo review.” Id. at 164. In a footnote in its opposition to the motion to dismiss, F5 generally indicated its desire to amend the complaint should the district court decide to dismiss it. Although the informality of the request is not ordinarily a basis to 20 We refer to the request as “implied” because F5 did not actually request leave to amend its complaint in the event of dismissal. Instead, in a footnote in its opposition to the defendants’ joint motion to dismiss, the plaintiff wrote: “In the event this Court ultimately dismisses Plaintiff’s Complaint, or any claim therein, Plaintiff reserves its right to amend the Complaint pursuant to” Rule 15. F5 Capital v. Pappas, No. 14-cv-9356, ECF No. 74 at 19 n.10. 55 deny leave to amend, Loreley Fin. (Jersey) No. 3 Ltd. v. Wells Fargo Sec., LLC, 797 F.3d 160, 190 (2d Cir. 2015), the district court did not deny leave to amend on that basis. Instead, it determined, albeit after only a brief analysis, that any amendment would be futile, a rationale that our decision in Loreley “leaves unaltered.” Id. Even on appeal, F5 has still “give[n] no clue as to how the complaint’s defects would be cured” through an amendment. Id. (internal quotation marks and citation omitted). In light of F5’s failure to indicate how it might amend the complaint to eliminate its gaping holes, along with the heightened pleading requirements associated with demand futility under Rule 23.1, we see no error in the district court’s denial of leave to amend based on futility. F5’s principal argument on appeal is not persuasive. F5 contends that the district court’s individual practices in civil cases improperly limited its ability to amend the complaint, contravening Rule 15 and our decision in Loreley. At the time that the district court dismissed the complaint, its individual practices provided in relevant part that: “If a motion to dismiss is filed, the plaintiff has a right to amend its pleading within twenty-one days . . . . If the plaintiff elects not to amend its pleading, no further opportunities to amend to address the 56 deficiencies identified by the motion to dismiss will ordinarily be granted.”21 Appellant Br. 55. Whatever the possible problems with the district court’s individual practices might be as a general matter, in this case the district court was clearly justified in finding that F5’s implicit request for amendment was futile. In its reply, F5 claims that it was entitled to a judicial resolution of the complaint before it could be expected to propose specific amendments. In F5’s view, the pleading deficiencies were “latent,” “easily missed or misperceived without full briefing and judicial resolution,” and even “borderline, . . . subject to reasonable dispute.” Loreley Fin. (Jersey) No. 3 Ltd., 797 F.3d at 191. Thus, the plaintiff argues that, as in Loreley, the district court’s decision to deny leave to amend was error. The procedure that we rejected in Loreley, however, is distinct from what occurred in this case: there, the court required the parties to attend a pre-motion conference and to exchange three-page letters concerning the anticipated motion to dismiss. Id. at 190. “The impropriety occurred . . . when, in the course of the conference, [the court] presented Plaintiffs with a Hobson’s 21 In 2016, the district court amended its individual practices to add “absent good cause” to the end of this sentence. That amendment merely clarified what we find implicit in the word “ordinarily” in the prior rule. 57 choice: agree to cure deficiencies not yet fully briefed and decided or forfeit the opportunity to replead.” Id. The district court did not employ that prohibited procedure here, where F5 had an opportunity to amend in response to full briefing of the defendants’ motion to dismiss, and failed, in its own briefing on the motion or by motion for leave to amend after the district court dismissed the case, to explain how it proposed to amend the complaint to cure its defects. To this day, F5 has not explained how it could amend the complaint to plead demand futility or otherwise survive a motion to dismiss. Thus, we affirm the district court’s decision to deny F5 leave to file its unspecified amendments to the complaint. CONCLUSION For the foregoing reasons, we AFFIRM the judgment of the district court in all respects. 58