TABLE OF CONTENTS
I. Private Right of Action under Section 14(d)(7) OI ÍD
II. The Wasserman Transaction Oí to
III. The Sheinberg Payment Cí
IV. The Preclusive Effect of the Settlement of the Delaware Class Action Cí ÍO
A. The Delaware Settlement (Oí «O
B. The Full Faith and Credit Question Cí K
1. The Jurisdiction of State Courts to Release Exclusively Federal Claims in a Class Settlement Oí oí Hi
2. The Disparity Between the State and Federal Claims O) oí ÜI
C. The Contract Bar Argument O) oí Oí
D. Conclusion Oí oí OO
V. Class Certification Oí oí OO
VI. The Motion to Amend the Complaint Oí oí ífi
VII. Conclusion CT) oí ÍD
In 1990, Matsushita Electrical Co. Ltd. (“Matsushita”) acquired MCA, Inc. (“MCA”) for $6.1 billion. The acquisition was accomplished through a tender offer of $71 per share of MCA common stock.1
Lew Wasserman, MCA’s chairman and chief executive officer at the time, owned 4,953,927 shares of MCA common stock worth $351,728,817 at the tender price of $71 per share. His cost basis was 3® per share. Rather than tender his shares at the tender offer price, Wasserman entered into a separate agreement with Matsushita, known as the “Capital Contribution and Loan Agreement,” pursuant to which Wasserman exchanged his shares for preferred stock in a *648wholly-owned Matsushita subsidiary called “MEA Holdings.”2 Matsushita agreed to fund MEA Holdings by contributing 106% of the tender price multiplied by the number of MCA shares Wasserman exchanged. The MEA Holdings preferred stock Wasserman received pays a dividend of 8.75% annually, is secured by letters of credit, and is redeemable upon the death of either Wasserman or his wife, but in no event earlier than five years from the date of the exchange. Was-serman was 77 at the time. It is not disputed that the transaction was designed to be a tax-free exchange of Wasserman’s MCA stock under Internal Revenue Code § 351(a), 26 U.S.C. § 351(a) (1994).3
Sidney Sheinberg, MCA’s chief operating officer at the time of Matsushita’s tender offer, owned approximately 1,179,635 shares of MCA common stock. He tendered these shares pursuant to Matsushita’s $71 per share offer and received in exchange consideration worth approximately $83,754,085. Two days after Matsushita accepted all tendered shares, Sheinberg received an additional $21 million in cash, ostensibly in exchange for unexercised MCA stock options.
These consolidated appeals arise out of actions brought in the United States District Court for the Central District of California by former MCA shareholders4 who tendered their shares for the $71 tender price. They claim that Matsushita violated SEC Rule 14d-10, 17 C.F.R. § 240.14d-10 (1994), by treating Wasserman and Sheinberg differently from other shareholders in the tender offer. Rule 14d-10, known as the “all-holder, best-price” rule, requires bidders to treat all shareholders on equal terms.5
The district court denied plaintiffs’ motion for summary judgment on their claim that Matsushita’s agreement to pay Wasserman consideration that was different from the $71 per share tender offer violated Rule 14d-10, and later granted Matsushita’s motion for summary judgment on this claim.6 We reverse, instruct the district court to grant plaintiffs’ motion for partial summary judgment, and remand for further proceedings to determine the amount of damages, if any, that plaintiffs are entitled to recover as a result of the Wasserman transaction.
The district court granted Matsushita’s motion for summary judgment on all of plaintiffs’ claims. As noted above, with respect to plaintiffs’ claim that the Wasserman transaction violated Rule 14d-10, we reverse. With respect to plaintiffs’ claim that the Sheinberg payment violated Rule 14d-10, we vacate and remand for further proceedings to determine whether the $21 million Sheinberg payment was in fact a premium paid to encourage Sheinberg to tender his shares.7
*649We also reverse the district court’s orders denying the Epstein plaintiffs’ motions for class certification and leave to amend their complaint.
During the pendency of these consolidated appeals, the Delaware Court of Chancery entered a judgment approving the settlement of a state class action that released all claims arising out of Matsushita’s tender offer for MCA stock, including the Williams Act claims raised in the Epstein class action. Matsushita argues that the settlement of the Delaware class action precludes the federal claims raised in the Epstein action. We disagree and hold that the settlement of the Delaware class action does not preclude the Epstein class action.
I. Private Right of Action under Section 14(d)(7)
The SEC’s statutory authority to promulgate Rule 14d-10 derives from sections 14(d)(6) and 14(d)(7) of the 1968 Williams Act Amendments to the Securities Exchange Act of 1934. 15 U.S.C. § 78n(d)(6), (7) (1981).8 Matsushita9 makes the threshold argument that it cannot be sued by MCA shareholders for violating Rule 14d-10 because Congress did not intend sections 14(d)(6) and 14(d)(7) to be privately enforceable.10 In advancing this argument, Matsushita asks us to create a conflict with the Second and Third Circuits, both of which have held that Congress intended to create a private right of action under section 14(d)(7). See Polaroid Corp. v. Disney, 862 F.2d 987, 996 (3d Cir.1988); Field v. Trump, 850 F.2d 938, 946 (2d Cir.1988), cert. denied, 489 U.S. 1012, 109 S.Ct. 1122, 103 L.Ed.2d 185 (1989); cf. Pryor v. United States Steel Corp., 794 F.2d 52, 57-58 (2d Cir.) (holding that section 14(d)(6) also contains a private right of action), cert. denied, 479 U.S. 954, 107 S.Ct. 445, 93 L.Ed.2d 393 (1986). We find the reasoning of the Second and Third Circuits to be persuasive.
*650In Field and Pryor, the Second Circuit applied the traditional four-factor Cort v. Ash test in deciding whether, in enacting sections 14(d)(6) and 14(d)(7), “Congress intended to create ... by implication ... a private cause of action.” Pryor, 794 F.2d at 57 (quoting Touche Ross & Co. v. Redington, 442 U.S. 560, 575, 99 S.Ct. 2479, 2488-89, 61 L.Ed.2d 82 (1979)). In holding that Congress did intend to create a private right of action the Second Circuit reasoned, first, that both 14(d)(6) and 14(d)(7) “identify] [their] beneficiaries and, unlike the bulk of federal securities regulation, confer[] a substantive right on those beneficiaries.” Pryor, 794 F.2d at 57 (cited in Field, 850 F.2d at 946). Second, the court found that a private damages remedy was totally consistent with the statutory purpose of protecting injured investors and provided a particularly effective means of enforcing sections 14(d)(6) and 14(d)(7). Field, 850 F.2d at 946; Pryor, 794 F.2d at 56, 58. Finally, the Second Circuit noted that these claims are not those “traditionally relegated to state law.” Field, 850 F.2d at 946; Pryor, 794 F.2d at 58.
In Polaroid, the Third Circuit relied on Pryor and Field, but added that the “ ‘contemporary legal context’ informing what Congress perceived itself to be doing when it acted,” supported an inference that Congress intended to create a private remedy for violations of section 14(d)(7). Polaroid, 862 F.2d at 995 (quoting Cannon v. Univ. of Chicago, 441 U.S. 677, 698-99, 99 S.Ct. 1946, 1958-59, 60 L.Ed.2d 560 (1979)). In 1968, when the Williams Act was enacted, various lower federal courts had construed section 10(b) of the 1934 Act as providing a private cause of action, and the Supreme Court, in J.I. Case v. Borak, 377 U.S. 426, 84 S.Ct. 1555, 12 L.Ed.2d 423 (1964), had announced a liberal policy toward inferring private rights of action for securities law violations generally. See Polaroid, 862 F.2d at 995-96. Because Congress enacted the Williams Act in this context, “it is reasonable to conclude that Congress passed the Williams Act with an understanding that courts would construe the Act as creating private remedies that would enforce the provisions of the Act effectively.” Id. at 996.
Matsushita argues that we should not follow Pryor, Field, and Polaroid because their rationale is inconsistent with our decision in In re Washington Pub. Power Supply Sys. Sec. Litig., 823 F.2d 1349 (9th Cir.1987) (en banc) (hereinafter ‘WPPSS”). We reject this argument. In WPPSS, we held that section 17(a) of the Securities Act of 1933 did not create a private right of action, id. at 1358, but in doing so we hardly sounded the death knell for implied rights of action generally. Instead, like the Second and Third Circuits in Pryor, Field, and Polaroid, we applied the traditional four-factor Court v. Ash test, emphasizing two prongs in particular: whether there is an implicit indication, in the statute’s language, legislative history, or structure, of legislative intent to create a private remedy, and whether such an implication would be consistent with the underlying legislative scheme.11 Id. at 1353. Like the Third Circuit, further, we focused on the context in which the statute was enacted. See id. at 1357 (looking to “contemporary legal context prevailing at the time of any comprehensive reexamination of or significant amendment to” the securities laws in evaluating whether Congress intended to create private right of action under section 17(a) of the 1933 Act). In WPPSS we not only found no evidence whatsoever of congressional intent to create a private right of action under section 17(a), but we also indicated that creating an implied private right of action would frustrate the purposes of the legislative scheme. Id. at 1355-56. In Pryor, Field, and Polaroid, just the opposite was true.
While we stated in WPPSS that we “will not engraft a remedy on a statute, no matter how salutary, that Congress did not intend to provide,” id. at 1353 (quoting California v. Sierra Club, 451 U.S. 287, 297, 101 S.Ct. 1775, 1781, 68 L.Ed.2d 101 (1981)), we made equally clear that “a private remedy may be inferred from the plain language of the statute, the statutory structure, or some other *651source.” Id. The principal difference between this case and WPPSS is the clarity of the statutory language. In WPPSS, the statute at issue, section 17(a) of the 1933 Act, states that “[i]t shall be unlawful for any person in the offer or sale of any securities ... (1) to employ any device, scheme, or artifice to defraud, or (2) to obtain money or property by means of any untrue statement of a material fact_” 15 U.S.C. § 77(q)(a) (1981). Section 14(d)(7), in contrast, specifically endows shareholders with the right to receive any increased consideration offered to other shareholders. It states that whenever any person “varies the terms of a tender offer” by “increasing the consideration offered” to some security holders, “such person shall pay the increased consideration to each security holder whose securities are taken up and paid for pursuant to the tender of-fer_” (emphasis added). This language, of course, constitutes far more than the “general censure of fraudulent practices” we found insufficient to create an implied right of action in WPPSS, 823 F.2d at 1353. Instead, it is “legislation with an unmistakable focus on the benefitted class.” Id. at 1354.
The structure of section 14(d)(7) also points to Congress’ intent to give shareholders the right to sue. If shareholders were not permitted to sue for damages for violations of section 14(d)(7), there would be no way, once a violation has occurred, to enforce the express statutory command that a bidder “shall pay to each security holder” any increased consideration paid to any other security holder.12 Under section 21(d)(1) of the 1934 Act, the SEC’s authority to enforce the provisions of the 1934 Act is limited to bringing injunctive actions.
Finally, the legal context of section 14(d)(7) of the 1968 Williams Act differed markedly from that of section 17(a) of the 1933 Act. It was not until 1964 that the Supreme Court announced the implied right of action doctrine in J.I. Case v. Borak, 377 U.S. 426, 84 S.Ct. 1555, 12 L.Ed.2d 423 (1964). Thus, at the time of passage of the Williams Act, the Court’s implied right of action jurisprudence gave Congress reason to think it need not decide the private right of action question itself.
Matsushita also urges us not to follow Pryor, Field, and Polaroid on the ground that these cases are inconsistent with the implied cause of action jurisprudence of the Supreme Court as it exists today. As Matsu-shita would have us read the Supreme Court’s cases, a private right of action would not be recognized unless Congress explicitly created one. For example, Matsushita asserts that the Supreme Court would not interpret section 14(d)(7) as creating a damages remedy for MCA shareholders because Congress did not say explicitly that all “security holders shall have the right to receive” any increased consideration offered to select shareholders. Wasserman Brief at 56.
Matsushita’s argument that there has been a recent sea change in the Court’s implied right of action jurisprudence is based on wishful thinking, not ease law.13 The argu*652ment is essentially the same one advanced by Justice Scalia in his concurring opinion in Thompson v. Thompson, 484 U.S. 174, 188, 108 S.Ct. 513, 520-21, 98 L.Ed.2d 512 (1988) (Scalia, J., concurring in the judgment), where he urged his colleagues to “get out of the business of implied private rights of action altogether.” Id. at 192, 108 S.Ct. at 522-23. The problem is that no other justice of the Court has yet to endorse the “get out of the business” approach to implied causes of action. To the contrary, the Court has only recently reaffirmed the vitality of the Cort v. Ash test — which the Second and Third Circuits applied in holding that section 14(d)(7) is privately enforceable — as an aid in “determin[ing] ‘whether Congress intended to create the private remedy asserted’ for the violation of statutory rights.” Wilder v. Virginia Hospital Assn., 496 U.S. 498, 508 n. 9, 110 S.Ct. 2510, 2517 n. 9, 110 L.Ed.2d 455 (1990) (quoting Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S. 11, 15-16, 100 S.Ct. 242, 245-46, 62 L.Ed.2d 146 (1979)); Suter v. Artist M., 503 U.S. 347, 363-64, 112 S.Ct. 1360, 1370, 118 L.Ed.2d 1 (1992) (applying “the familiar test of Cort v. Ash" in concluding that Congress did not intend to make a private remedy available to enforce the Adoption Assistance and Child Welfare Act). Indeed, in Thompson v. Thompson, the case in which Justice Scalia called for an end to implied private rights of action, the eight members of the Court who comprised the majority stated that the existence of a private right of action does not require “evidence that Members of Congress, in enacting the statute, actually had in mind the creation of a private cause of action.” Thompson, 484 U.S. at 179, 108 S.Ct. at 516. Rather, the Court reaffirmed its long-standing view that “as an implied cause of action doctrine suggests,” Congress’ “intent may appear implicitly in the language or structure of the statute, or in the circumstances of its enactment.” Id. (internal citation and quotation omitted).14
We therefore reject Matsushita’s argument that the rationale of Pryor, Field, and Polaroid is inconsistent with Supreme Court implied private right of action jurisprudence as it exists today.15 Like the Second and Third Circuits, we believe that the statutory language — that a bidder “shall pay the increased consideration” to a shareholder who was paid less than another shareholder — is more than adequate under existing Supreme Court case law to support a finding of congressional intent to create a private right of action for violations of section 14(d)(7).
In sum, we hold that in enacting section 14(d)(7) of the Williams Act, Congress intended to provide a damage remedy as a means of enforcing its command that every security holder who tenders his shares be paid any “increased consideration” offered to others.
II. The Wasserman Transaction
Captioned the “Equal treatment of securities holders,” Rule 14d-10 prohibits a bidder *653from making a tender offer that is not open to all shareholders or that is made to shareholders at varying prices.16 The gist of plaintiffs’ claims is that Matsushita violated the antidiscrimination requirements of the Rule by paying Wasserman and Sheinberg premiums pursuant to the tender offer.
Negotiations between Matsushita and MCA began in August 1990, when a representative of Matsushita telephoned MCA’s financial advisor to express interest in acquiring MCA. During the course of the talks, Matsushita stressed that it wanted Wasserman and Sheinberg to commit their shares to Matsushita in advance of the friendly takeover and to remain in MCA’s employment for five years. On the morning of November 26, 1990, Matsushita and Was-serman entered into the Capital Contribution and Loan Agreement, pursuant to which Wasserman agreed to exchange his MCA shares for preferred stock in a subsidiary Matsushita would create called “MEA Holdings.”
Performance of the Capital Contribution and Loan Agreement was conditioned on the tender offer in several respects. First, neither Matsushita nor Wasserman was obligated to perform the Agreement if any of the conditions of the tender offer were not satisfied. See Capital Contribution and Loan Agreement § 7(c), ER 357 Exhibit S at 12. If, for example, Matsushita did not acquire 50% of MCA’s common stock as a result of the tender offer, the Wasserman deal would be off. Second, the timing of performance was tied to the tender offer. The Wasser-man exchange was scheduled to take place “immediately following the time at which shares of MCA Common Stock are accepted for payment pursuant to and in accordance with the terms of the offer....” Id. § 2(a), ER 357 Exhibit S at 4. Third, the amount of cash Matsushita was required to contribute in order to fund MEA Holdings was dependent upon the tender price, with Matsushita agreeing to contribute to MEA Holdings 106% of the “highest price paid ... for any shares of MCA Common Stock pursuant to the [tender] offer.” Id. at § 1(c), ER 357 Exhibit S at 2. Finally, the redemption value of Wasserman’s preferred stock was set as the tender price. Id. Thus if Matsushita increased the tender price at the last minute in response to a competitive bid for MCA, Matsushita would have been required to increase both its funding of MEA Holdings and the amount paid upon the redemption of the Wasserman preferred stock.
Moments after signing the Capital Contribution and Loan agreement, Matsushita and MCA announced the $71 per share tender offer. Shareholders were given from the time of the announcement until 12:01 a.m. on December 29, 1990 to tender their shares. The owners of 91% of MCA’s common stock did so, and at 12:05 a.m., Matsushita accepted all tendered shares for payment. At 1:25 a.m., Matsushita exchanged Wasserman’s shares for MEA Holdings preferred stock pursuant to the Capital Contribution and Loan Agreement. MCA was merged into Matsushita as a wholly owned subsidiary on January 3, 1991.
*654Whether the Wasserman transaction violated Rule 14d-10 depends upon whether Wasserman received greater consideration than other MCA shareholders “during such tender offer,” Rule 14d — 10(a)(2), or whether he received a type of consideration not offered to other MCA shareholders “in a tender offer.” Rule 14d-10(c).17
Matsushita argues that the Wasserman transaction falls outside the Rule’s ambit because it closed after the tender offer period expired. The tender offer period, Matsushita contends, ended when it accepted the tendered MCA shares for payment — at 12:05 a.m. on December 29, 1990, one hour and 20 minutes before Wasserman’s shares were exchanged. In Matsushita’s view, liability under Rule 14d-10 boils down to a pure question of timing: the Rule is simply a “mechanical provision” concerned with “payments to shareholders of a target corporation only during a specifically-defined tender offer period.” Brief of Matsushita at 23 (emphasis in original). Outside that period, Matsushita insists, “Rule 14d-10 is without effect” because the Rule “is engaged (or not engaged) depending upon when payment is made. ” Id. at 23, 35 (emphasis in original).
Although Matsushita argues that Rule 14d-10 is designed to operate only during a “specifically-defined tender offer period,” neither the phrase “tender offer period” nor a specific time frame is to be found in the Rule’s text. To be sure, section (a)(2) of the Rule prohibits paying one security holder more than another “during such tender offer.” But the term “tender offer,” as used in the federal securities laws, has never been interpreted to denote a rigid period of time. On the contrary, in order to prevent bidders from circumventing the Williams Act’s requirements, Congress, the SEC, and the courts have steadfastly refused to give the term a fixed definition. Instead, we have, held that “[t]o serve the purposes of the Williams Act, there is a need for flexibility in fashioning a definition of a tender offer.” SEC v. Carter Hawley Hale Stores, Inc., 760 F.2d 945, 950 (9th Cir.1985).
Even if the language of Rule 14d-10(a)(2) were to provide a measure of support for Matsushita’s timing argument, Matsushita would still be unable to account for the language of Rule 14d-10(c), which prohibits a bidder from “offering] ... more than one type of consideration in a tender offer” if shareholders are not permitted to choose between the different types of consideration offered. (Emphasis added). Section (c)(1) makes no mention of payment and no mention of timing. Instead, Rule 14d-10(c)(l) prohibits a bidder such as Matsushita from offering a shareholder, such as Wasserman, not only consideration of greater value than that offered to other shareholders, but also consideration that is different from that of-, fered to other shareholders. It endows each *655shareholder with the “equal right to elect among each of the types of consideration offered,” regardless of when actual payment is made.
The administrative history of Rule 14d-10 underscores how strained Matsushita’s timing argument is. It suggests anything but the notion that the SEC intended the Rule to be a mechanical provision concerned not with discriminatory tender offers, but with the timing of payment to favored shareholders. In promulgating Rule 14d-10, the SEC emphasized the need for “equality of treatment among all shareholders who tender their shares.” Id., 1985 WL 61507, 1985 SEC LEXIS 1175 at *15 (quoting S.Rep. No. 550, 90th Cong., 1st Sess. 10 (1967)). It further characterized Rule 14d-10 as a substantive provision necessary to achieve the “investor protection purposes of the [1934] Exchange Act” and the Williams Act. SEC Release No. 34-22198, 1985 WL 61507, 1985 SEC LEXIS 1175 at *2 (July 1, 1985). At no point in its discussion of the Rule’s purposes did the SEC suggest that the Rule’s sole focus is the timing of payments.
Matsushita implicitly acknowledges that Rule 14d-10 does not contain a rigid time frame of its own when it urges us to read Rule 14d-10 as incorporating, sub silentio, the time frame set out in Rule 10b-13, which prohibits side purchases “from the time [a] tender offer or exchange offer is publicly announced or otherwise made known ... [to security holders] until the expiration of the period ... during which securities tendered pursuant to such tender offer or exchange offer may by the terms of such offer be accepted or rejected.” Matsushita justifies fusing elements of two separate regulations by asserting that “for purposes of the Williams Act and the SEC Rules promulgated thereunder,” Rule 10b-13 defines the time period of a tender offer. Brief of Matsushita at 24.
Matsushita offers no authority for incorporating Rule 10b-13’s time frame into Rule 14d-10. In promulgating the all-holders, best-price Rule in 1986, the SEC gave no hint that its new regulation would be governed by Rule 10b-13’s time clock. Nor does the Williams Act fully incorporate Rule 10b-13’s timing provisions. Rule 10b-13 prohibits bidders from making side deals during a fixed period of time; it does not purport to serve as a general definition of when a tender offer begins and ends. In fact, in Rule 14d-2, the SEC rejected the notion that Rule 10b-13 timing determines when tender offers start for purposes of section 14(d)(7) and the rules promulgated thereunder.18
We therefore reject Matsushita’s timing argument. Indeed, if adopted, it would drain Rule 14d-10 of all its force. Under Matsu-shita’s reading, even the most blatantly discriminatory tender offer — in which large shareholders were paid twice as much as small shareholders — would fall outside Rule 14d-10’s prohibition, so long as the bidder waited a few seconds after it accepted all of the tendered shares before paying the favored shareholders. Rule 14d-10’s equality requirements, which “expressly preclude bidders from discriminating among holders of the class of securities that is the subject of the offer, either by exclusion from the offer or by payment of different consideration,” SEC Release No. 34-23421, 1986 WL 71340, 1986 SEC LEXIS 1179 at *10 (July 11,1986), cannot be so easily circumvented.
An inquiry more in keeping with the language and purposes of Rule 14d-10 focuses not on when Wasserman was paid, but on whether the Wasserman transaction was an integral part of Matsushita’s tender offer. If it was, Matsushita violated Rule 14d-10 because it paid him, pursuant to the tender offer, different, and perhaps more valuable *656consideration than it offered to other shareholders.19
Matsushita contends that the Wasserman transaction cannot be deemed a part of the tender offer because it was merely a private exchange of stock, structured to take place after the tender offer finished. But Matsushita’s assumption that the Wasserman transaction was private, rather than a part of the tender offer, begs the question. In Field v. Trump, 850 F.2d at 944, the Second Circuit held that “[w]hether [an] acquisition of shares in a corporation is part of the tender offer for purposes of the Act cannot be determined by rubber-stamping the label used by the acquiror.” To hold otherwise, the court stated, would render “virtually all of the provisions of the Williams Act, including its filing and disclosure requirements,” subject to evasion “simply by an offeror’s announcement that offers to purchase ... stock were private purchases.” Id. Thus, the court observed that because the Williams Act and its implementing regulations do not define the term “tender offer,” “[cjourts faced with the question of whether purchases of a corporation’s shares are privately negotiated or are part of a tender offer have applied a functional test that scrutinizes such purchases in the context of various salient characteristics of tender offers and the purposes of the Williams Act.” Id. at 943-44. See also SEC Release No. 34-22198, 1985 WL 61507, 1985 SEC LEXIS 1175, at *7 (July 1, 1985) (“[TJhe fact that ... different consideration is offered to different holders of the same class of securities, does not mean that a tender offer has not been made under the Williams Act. Rather, if such a transaction is found to be a tender offer, then the tender offer would not have been made in compliance with the all-holders requirement”).
Because the terms of the Wasserman Capital Contribution and Loan Agreement were in several material respects conditioned on the terms of the public tender offer, we can only conclude that the Wasserman transaction was an integral part of the offer and subject to Rule 14d-10’s requirements. Two facts compel this conclusion: first, the redemption value of Wasserman’s preferred stock incorporated the tender offer price by reference, and second, the Capital Contribution and Loan Agreement was conditioned on the tender offer’s success. If the tender offer failed, Wasserman would have remained the owner of his MCA stock. This is precisely the arrangement Matsushita made with its shareholders through its public tender offer: if an insufficient number of shares were tendered, each shareholder too would have retained ownership of her MCA stock. The deal Matsushita made with Wasserman thus differed from the tender offer in only one material respect — the type (and possibly the value) of consideration provided. Rule 14d — 10(c)(1) forbids just such a transaction.
To be sure, the fact that a private purchase of stock and a public tender offer are both part of a single plan of acquisition does not, by itself, render the purchase a part of a tender offer for purposes of Rule 14d-10. Rule 14d-10 does not prohibit transactions entered into or effected before, or after, a tender offer — provided that all material terms of the transaction stand independent of the tender offer. Thus a bidder who purchases shares from a particular shareholder before a tender offer begins does not violate Rule 14d-10. See, e.g., Kahn v. Virginia Retirement Systems, 13 F.3d 110 (4th Cir.1993) (bidder’s unconditional private purchase of target’s shares two days before tender offer was formally announced did not violate section 14(d)(7) and Rule 14d-10), cert. denied, — U.S. -, 114 S.Ct. 1834, 128 L.Ed.2d 462 (1994).
If, in advance of the tender offer, Wasser-man had become unconditionally obligated to exchange his MCA shares, the transaction *657would not have violated Rule 14d-10, even if Matsushita believed that acquiring Wasser-man’s shares was a first step in acquiring MCA. In such a case, both Wasserman and Matsushita would have assumed the burdens of their agreement despite the risk that an anticipated tender offer would fail, or command a different price. But such a course was not followed. Matsushita sought to acquire MCA without purchasing the holdings of individual shareholders block by block and accordingly subjecting itself to the risk that it would end up with a huge investment in MCA stock, but without control. The tender offer device is designed to avoid this risk, but only if holders of the same security are offered precisely the same consideration.
Matsushita argues, in the alternative, that the summary judgment should be affirmed because plaintiffs cannot prove that they suffered injury from the Wasserman deal. This argument, however, confuses the question of whether Matsushita complied with Rule 14d-10 with the question whether plaintiffs may recover damages. By paying Wasserman consideration that it failed to offer to other shareholders, Matsushita violated Rule 14d — 10(c)(1). It remains to be determined on remand whether the preferred stock Wasserman received had a value greater than $826,959,182, the value of his 4,953,927 shares at the cash tender price of $66 per share.20
Accordingly, we reverse the summary judgment in favor of Matsushita on plaintiffs’ claim that the Wasserman transaction violated Rule 14d-10, reverse the district court’s order denying plaintiffs’ motion for partial summary judgment that the Wasserman transaction did violate Rule 14d-10, and remand for further proceedings on the question whether the consideration Wasserman received in exchange for his MCA stock had a greater value than what plaintiffs’ received and, if so, how much greater.
III. The Sheinberg Payment
Unlike Wasserman, Sheinberg tendered his MCA shares for the $71 per share tender price. Pursuant to an amended employment agreement executed shortly before the tender offer was announced on November 26, 1990, MCA agreed, with Matsushita’s approval, to pay Sheinberg $21 million if the tender offer succeeded. Two days after Matsushita accepted all tendered MCA shares, Sheinberg received the promised payment.
The parties dispute the purpose of the Sheinberg payment. Plaintiffs contend that it constituted a covert premium of $17.80 per share designed to induce Sheinberg to tender his shares. Matsushita contends that the payment was instead designed both to cash out stock options that MCA had given Shein-berg because of his performance as its chief operating officer and to compensate Shein-berg for agreeing to amend his employment contract with MCA.
According to the minutes of the MCA Board of Director’s meeting of November 25, 1990, the day before the tender offer was announced, Wasserman explained that before discussions with Matsushita began, he had intended to grant Sheinberg options for 1,000,000 shares of MCA stock. ER 357 *658Exhibit D. He stated, according to the minutes, that when merger negotiations began, “there had been a freeze put on further stock option grants,” and he was therefore unable to propose the Sheinberg options at the next meeting. Id. at 31. According to the minutes, Wasserman said he believed that it would be unfair to deprive Sheinberg of his legitimate expectations simply because the tender offer was nearly finalized. Id. He recommended that the options be granted at an exercise price of $50 per share. Id.
Wasserman also told the Board that Mat-sushita had approved of the Sheinberg options and that it had agreed to give Shein-berg “cash payments if the transaction went forward, measured by the difference between the transaction price” and the $50 option price. Id. at 31-32. Martin Lipton, counsel for MCA, Wasserman, and Sheinberg in this transaction, told the Board that Sheinberg’s payment would amount to $21 million and that, as further consideration for Matsushita’s promise to pay him this sum, Sheinberg would be required to waive his right to severance pay in the event Matsushita acquired MCA and to agree to five years of continued employment. Id. at 33. The Stock Awards Plan Committee approved the “arrangements” Wasserman proposed for Sheinberg, “subject to the Board’s approval of the proposed merger agreement.” Id. at 68.
Plaintiffs argue that the Sheinberg payment was entirely conditional upon the success of the tender offer. If the tender offer did not succeed, it is undisputed that Shein-berg would not have received the $21 million. Matsushita points to no evidence in the record indicating that the grant of options to Sheinberg was anything more than hypothetical in the event that the tender offer did not succeed. At the MCA Board meeting, according to the minutes, Wasserman stated that if the merger “did not occur, he would present the arrangements to the Board for approval and implementation as stock-based awards.” Id. at 32. From this record a trier of fact could infer that Wasserman’s statement was merely a declaration of future intention, not a binding commitment to give Sheinberg stock options regardless of the outcome of the tender offer.
Plaintiffs also point to the absence of any evidence even suggesting that Sheinberg might be in line for an award of stock options before November 25, 1990, the eve of the tender offer. Moreover, they cite the deposition testimony of MCA’s financial advisor, Felix Rohatyn, that he could not recollect any discussions of a stock option award to Sheinberg until Wasserman brought up the matter at the November 25 Board Meeting. ER 357 Exhibit A at 32.
Plaintiffs also argue that Matsushita’s account of the Sheinberg payment is internally inconsistent. If Matsushita is correct that Sheinberg owned — prior to the day the tender offer was announced — options for 1,000,-000 shares of MCA stock, each with an exercise price of $50 per share, it is unclear why Matsushita would claim that the Sheinberg payment was also made in consideration of an agreement to amend his employment contract. The $21 million is ostensibly derived from the difference between the $50 exercise price and the $71 tender price, multiplied by the 1,000,000 option shares.
We hold that plaintiffs’ circumstantial evidence gives rise to a disputed issue of material fact and therefore precludes summary judgment for Matsushita on the claim that the $21 million payment to Sheinberg violated Rule 14d-10. A trier of fact could decide, on the evidence cited by plaintiffs, not to credit Matsushita’s explanation for the Shein-berg payment.
Matsushita argues, without controverting any of plaintiffs’ evidence, that the Sheinberg payment did not violate Rule 14d-10 because (1) the Sheinberg payment was made after Matsushita accepted the tendered MCA shares, and (2) the payment was made by MCA, not Matsushita.
We rejected Matsushita’s timing argument in the context of the Wasserman transaction, see Part II, supra, and the fact that Matsu-shita repeats the argument in defending the Sheinberg payment only brings into sharp relief its lack of merit.
Moreover, Matsushita’s assertion that MCA rather than Matsushita paid Shein-*659berg21 begs the question. The central issue regarding the legality of the Sheinberg payment, after all, is whether it constitutes incentive compensation that MCA wanted to give Sheinberg independently of the Matsu-shita deal, or a premium that Matsushita wanted to give Sheinberg as an inducement to support the tender offer and tender his own shares. The mere fact that MCA cut Sheinberg’s $21 million check just before it was formally merged into Matsushita may be a relevant fact, but it does not establish as a matter of law that the $21 million Sheinberg payment did not violate Rule 14d-10.
Because Matsushita’s liability under Rule 14d-10 for the Sheinberg payment turns on disputed questions of material fact, we vacate the summary judgment for Matsushita on this claim and remand for further proceedings. See Fed.R.Civ.P. 56(c).
IV. The Preclusive Effect of the Settlement of the Delaware Class Action
Before addressing the Epstein plaintiffs’ argument that the district court abused its discretion in denying their motion for class certification, we consider Matsushita’s claim that the Epstein class action is barred by the judgment of the Delaware Court of Chancery settling the state class action. Although the Delaware class action was based exclusively on state law claims, the settlement agreement expressly stated that the federal securities law claims asserted in the “Epstein action ... are hereby compromised, settled, released and discharged with prejudice....” In re MCA, Inc. Shareholder Litig., Order and Final Judgment (No. 11740) (Feb. 22, 1993), at 3-4, 1993 WL 43024, at *1.
Matsushita makes a two-pronged preclusion argument. First, it argues that the Delaware judgment releasing the federal claims is entitled to preclusive effect as a matter of full faith and credit. Second, Mat-sushita argues that the release of the federal claims bars the Epstein action as a matter of contract.22
A. The Delaware Settlement
The Delaware class action was filed on September 26, 1990, the day the financial press reported that Matsushita was negotiating to buy MCA. The suit named as defendants MCA and its directors, including Was-serman and Sheinberg. The essence of the complaint was that MCA’s directors had breached their fiduciary duties by failing to implement a market check mechanism to maximize shareholder value upon a change of corporate control, as required by Revlon, Inc. v. MacAndrews & Forbes Holdings Inc., 506 A.2d 173, 182 (Del.1986).
On December 2, 1990, three days after public disclosure of the terms of the tender offer, the Epstein class action was filed in the Central District of California. Unlike the state action, the Epstein complaint named Matsushita as a defendant and claimed that the tender offer violated SEC Rules 14d-10 and 10b-3.
*660On December 4, 1990, the day after the Epstein action was filed in the Central District, lead counsel for the Delaware plaintiffs transmitted a letter to MCA’s counsel that was to serve as the basis of an amended complaint. In that letter, the Delaware plaintiffs stated their intention to add additional claims against all directors. In particular, the complaint was amended to allege that MCA was wasting corporate assets by increasing its exposure to liability for violations of Rules 10b-13 and 14d-10, that directors Wasserman and Sheinberg had breached their fiduciary duties by negotiating preferential deals with Matsushita, and that MCA failed to make full disclosure of the benefits MCA insiders would receive from the takeover. Finally, Matsushita was added as a defendant and charged with conspiring with and aiding and abetting MCA directors in violating Delaware law.
On December 11, 1990, a week after Delaware class counsel’s letter proposing to amend the complaint and add Matsushita as a defendant, the defendants reported to the Central District that they had reached a settlement of the Delaware class action “in principle.” On December 14, the amended complaint was filed in the Court of Chancery, and on December 17, the parties agreed on the terms of a settlement to be submitted to the Vice Chancellor for approval. The settlement provided for the payment of $1,000,000 in fees to class counsel, but no monetary benefit for class members. The only arguable benefit to class members was a proposed change in the poison pill of the new corporation that was to be formed to own MCA’s radio station if the tender offer succeeded. The Vice Chancellor found the value of this poison pill provision to class members to be “illusionary.” See In re MCA Shareholders Litig., Inc., 598 A.2d 687, 696 (1991). Finally, the settlement provided for the release of all claims arising out of the tender offer, both state and federal.
In defending the settlement before the Vice Chancellor on January 30, 1991, class counsel argued that the state claims, while not frivolous, had little merit. Counsel also argued that the federal claims were frivolous and should not be pursued. On April 22, 1991, the Vice Chancellor disapproved the proposed settlement. He agreed with class counsel that the state law claims being compromised were “at best, extremely weak and, therefore, have little or no value.” In re MCA, Inc. Shareholders Litig., 598 A.2d at 694. In particular, he found the state claim of preferential treatment for Wasserman and other directors to be “weak” because no such state cause of action existed. Id. Under Delaware law, the Vice Chancellor noted, “[a] shareholder is not prohibited, by his status as a director, from dealing in shares of the corporation,” and that a director breaches his duty of loyalty only if he uses inside, information for his personal benefit. Id. Finding that the value of the settlement to the class lacked any “real monetary benefit,” the Vice Chancellor rejected the settlement because of the “significant value” of the federal claims that would have been released along with the valueless state claims. Id. at 696, 690.
Despite the Vice Chancellor’s opinion that the state claims had no merit and no value, Matsushita and its co-defendants took no action to dismiss the Delaware action. In fact, the docket of the Court of Chancery reflects no action of any kind — no discovery, no motions, no settlement or status conferences— until after the Central District awarded summary judgment to the defendants in the federal actions in February, 1992, more than a year after the Vice Chancellor disapproved the original Delaware settlement.
On October 22, 1992, after the federal plaintiffs filed their notices of appeal of the summary judgment to the Ninth Circuit, the parties to the Delaware action entered into a new settlement agreement. It provided for the creation of a $2 million settlement fund, enough to pay shareholders two to three cents per share before payment of fees and costs.23 Like the first settlement agreement, the second agreement provided for the release of all claims arising out of Matsushita’s acquisition of MCA, both state and federal. Unlike the first agreement, the second settle*661ment agreement permitted class members to opt out.24
The Vice Chancellor approved the second settlement agreement, concluding that “it is in the best interests of the class to settle this litigation and the terms of the settlement are fair and reasonable — although the value of the benefit to the class is meager.” In re MCA, Inc. Shareholders Litig., 1993 WL 43024, at *1 (Del.Ch. Feb. 16, 1993). He approved the second settlement for the sole reason that the federal claims which he originally thought had “significant value,” had been reduced to “minimal economic value” by the summary judgment entered in the Central District. The Vice Chancellor uncritically accepted the summary judgment as having destroyed the value of the federal claims without considering the fact that the Ninth Circuit reviews summary judgments de novo. See T.W. Elec. Serv., Inc. v. Pacific Elec. Contractors Ass’n, 809 F.2d 626, 629 (9th Cir.1987). The Vice Chancellor downgraded the value of the federal claims even as he expressed reluctance to assess the merits of those claims because they were “outside the jurisdiction of this Court.” Id., 1993 WL 43024 at *4.
Objectors argued that the settlement was collusive because the defendants “cut a deal” with the named plaintiffs in the Delaware action and their attorneys in order to extinguish the claims pending in the federal litigation. In re MCA Inc. Shareholders Litig., WL 43024, at *5. The Vice Chancellor acknowledged that the potential for this type of abuse “clearly exists in representative litigation,” and that “suspicions abound” when “the settling parties have previously proposed a patently inadequate settlement in which the class would have received no monetary benefit but the attorneys would have received $1 million in fees.” Id. at *4. Nonetheless, he refused to make a finding of collusion on the record before him because objectors “offered no evidence of any collusion.” Id. at *5.
The judgment incorporating the terms of the settlement was summarily affirmed by the Delaware Supreme Court. In re MCA Inc. Shareholders Litig., 633 A.2d 370 (Del.1993).
B. The Full Faith and Credit Question
1. The Jurisdiction of State Courts to Release Exclusively Federal Claims in a Class Settlement
Matsushita first argues that the Epstein action is precluded because the Delaware judgment releasing the federal claims is entitled to full faith and credit. As a general rule, the Full Faith and Credit Act, 28 U.S.C. § 1738, “requires federal courts to give the same preclusive effect to state court judgments that those judgments would be given in the courts of the State from which the judgments emerged.” Kremer v. Chemical Constr. Corp., 456 U.S. 461, 466, 102 S.Ct. 1883, 1889, 72 L.Ed.2d 262 (1982). This rule does not apply, however, if the state court “did not have jurisdiction over the subject matter or the relevant parties.” Underwrirers Nat’l Assurance Co. v. North Carolina Life and Accidental Health Insurance Gty. Ass’n, 455 U.S. 691, 704-05, 102 S.Ct. 1357, 1365-66, 71 L.Ed.2d 558 (1982). See also Thomas v. Washington Gas Light Co., 448 U.S. 261, 283, 100 S.Ct. 2647, 2661-62, 65 L.Ed.2d 757 (1980); Durfee v. Duke, 375 U.S. 106, 110, 84 S.Ct. 242, 244-45, 11 L.Ed.2d 186 (1963); Grubb v. Public Util. Comm’n, 281 U.S. 470, 475, 50 S.Ct. 374, 376-77, 74 L.Ed. 972 (1930); Thompson v. Whitman, 85 U.S. (18 Wall) 457, 469, 21 L.Ed. 897 (1873).25 *662The Epstein plaintiffs argue that the Delaware class settlement is not entitled to full faith and credit to the extent it released federal claims which are beyond the subject matter jurisdiction of state courts.
Matsushita responds that courts have uniformly held that
a state court class action settlement properly may preclude, by release, continued litigation by members of the class of all claims — state or federal, whether or not within the exclusive jurisdiction of the federal courts — arising out of the same subject matter or transaction.
Supplemental Brief of Matsushita at 13. In other words, Matsushita reads existing case law as sanctioning the release of exclusively federal claims in the settlement of state class actions as long as the state and federal claims arise out of the same subject matter or transaction. We believe Matsushita’s argument represents an overly expansive reading of the case law. As we read the cases, they support only a limited state court power to release exclusively federal claims in a class action settlement.
Of all the cases cited by Matsushita, only two, Grimes v. Vitalink Communications Corp., 17 F.3d 1553 (3d Cir.), cert. denied, — U.S. -, 115 S.Ct. 480, 130 L.Ed.2d 393 (1994) and Nottingham Partners v. Trans-Lux Corp., 925 F.2d 29 (1st Cir.1991), implicate the Supremacy Clause because in each, as here, a state court extinguished exclusively federal claims in settling a state class action.26
In every other case cited by Matsushita, the state court had subject matter jurisdiction to adjudicate all the claims that were released in the class settlement. Class Plaintiffs v. City of Seattle, 955 F.2d 1268, 1287-89 (9th Cir.), cert. denied, — U.S. -, 113 S.Ct. 408, 121 L.Ed.2d 333 (1992), for example, did not present the question of the preemptive effect of an exclusive federal jurisdiction statute on the power of a state court to release exclusively federal claims in settling a class action. Rather it involved the settlement of a federal class action based upon alleged violations of the federal securities laws. Our holding in Class Plaintiffs was that the Eleventh Amendment did not bar a federal district court from releasing claims against the State of Washington as part of a class settlement because the State had waived its sovereign immunity by participating in the settlement negotiations and consenting to be bound by the settlement agreement. Id. at 1289.
For the same reason, In re Corrugated Container Antitrust Litig., 643 F.2d 195, 196 (5th Cir.1981), did not implicate the Supremacy Clause because the jurisdictional competence of a state court to release exclusively federal claims was not at issue. The holding of the Fifth Circuit was that a federal district court, in approving a class settlement of federal claims, had jurisdictional competence to extinguish state claims that were not pleaded, but which it had pendent jurisdiction to adjudicate. 643 F.2d at 221 n. 39.
TBK Partners, Ltd. v. Western Union Corp., 675 F.2d 456 (2d Cir.1982), was also not a Supremacy Clause ease because it too involved the approval of a class action settlement by a federal district court. In TBK Partners, the Second Circuit upheld a class settlement that extinguished state claims *663that were not pleaded and arguably beyond the jurisdiction of the district court to adjudicate. Id. at 460. Although not directly on point because it was not a preemption case— it did not involve the release of exclusively federal claims by a state court — TBK Partners is useful authority because it announced a principled test for limiting the preclusive effect of a judgment based upon a class settlement. Writing for the court, Chief Judge Newman reasoned,
we see no reason why the judgment upon settlement cannot bar a claim that would have to be based on the identical factual predicate as that underlying the claims in the settled class action. We have previously “assume[d] that a settlement could properly be framed so as to prevent class members from subsequently asserting claims relying on a legal theory different from that relied upon in the class action complaint but depending on the very same set of facts.” National Super Spuds, Inc. v. New York Mercantile Exchange, 660 F.2d 9, 18 n. 7 (2d Cir.1981).
Id. at 460 (emphasis added). Thus the Second Circuit gave preclusive effect to the settlement judgment’s release of unpleaded state claims because the same facts were “at the core” of both the unpleaded state and the pleaded federal claims. Id. Had the judgment been based upon an adjudication rather than a settlement of the federal claims, the unpleaded state law claims would have been barred by the doctrine of issue preclusion because they turned on the “very same set of facts.”
In applying an issue preclusion test to limit the preclusive effect of a judgment approving a class settlement, TBK Partners followed Judge Friendly’s reasoning in National Super Spuds: “ ‘[i]f a judgment after trial cannot extinguish claims not asserted in the class action complaint, a judgment approving a settlement in such an action ordinarily should not be able to do so either.’ ” Id. at 462 (quoting National Super Spuds, Inc. v. New York Mercantile Exchange, 660 F.2d at 18). Thus, in TBK Partners, the Second Circuit upheld the release of unpleaded claims in a settlement judgment which would have been barred by the issue preclusive effect of a judgment based upon an adjudication of the pleaded claims.
As we read TBK Partners, Matsushita’s reliance on it as “dispositive” authority for its argument that we should give preclusive effect to the Delaware settlement judgment merely because the state and federal claims arise out of the same transaction is completely misplaced. First, TBK Partners is not a Supremacy Clause case. Second, TBK Partners announced and applied an issue preclusion test, not an “arising out of the same transaction” test, in defining the limits on a court’s power to release claims in a class settlement. As the Second Circuit summed up in TBK Partners:
[w]e therefore conclude that in order to achieve a comprehensive settlement that would prevent relitigation of settled questions at the core of a class action, a court may permit the release of a claim based on the identical factual predicate as that underlying the claims in the settled class action even though the claim was not presented and might not have been presentable in the class action.
Id. at 460 (emphasis added).
We now turn to the only two eases cited by Matsushita that involved the preclusive effect of a state court judgment approving a class action settlement that included a release of exclusively federal claims. In both eases TBK’s issue preclusion test was applied; in neither ease was Matsushita’s “arising out of the same transaction” test applied. In Nottingham Partners v. Trans-Lux Corp., 925 F.2d at 29, the First Circuit, citing TBK Partners, applied an issue preclusion test in holding that the release of exclusively federal claims in a state court settlement judgment was entitled to preclusive effect. Tracking the reasoning of both TBK Partners and National Super Spuds, the First Circuit’s rationale was that the federal claims depended on the same underlying facts as the state claims — the failure of a corporation to disclose allegedly material facts in a proxy statement. Id. at 30-31, 33.
In Grimes v. Vitalink Communications Corp., 17 F.3d at 1553, the Third Circuit followed the First Circuit’s lead in Nottingham Partners in applying TBK Partners’ *664issue preclusion test. The district court, citing TBK Partners and Nottingham Partners, had held “that this federal securities case is barred by the collateral estoppel effect of the Delaware judgment [releasing those claims as part of a class settlement]_” Id. at 1556. The Third Circuit affirmed because the “facts that underlie the federal non-disclosure claims were actually litigated during the state court proceeding and were conclusively resolved against the objectors, here the federal plaintiffs.” Id. at 1562.
Thus, the cases cited by Matsushita fail to support its contention that we should give preclusive effect to the Delaware settlement judgment because the federal claims it released arose out of the same transaction as the state claims. Instead, these cases stand for a more restrictive limitation on the power of state courts to extinguish exclusively federal claims in approving class action settlements: a state court may release exclusively federal claims that would have been extinguished by the issue preclusive effect of an adjudication of the state claims.
National Super Spuds and TBK Partners together provide the doctrinal framework for using issue preclusion in determining the limits of judicial authority to release unplead-ed claims in settling class actions. But though neither National Super Spuds nor TBK Partners involved the preemptive effect of an exclusive federal jurisdiction statute on the reach of state judicial power in settling class actions, both involved the release of unpleaded state claims by federal district courts which had or may have had pendent jurisdiction to adjudicate the claims. Nonetheless, in both cases the Second Circuit recognized the importance of limiting judicial authority generally to release unpleaded claims in class settlements.
Nottingham Partners and Grimes, in contrast, did implicate the Supremacy Clause because, as in the instant ease, exclusively federal securities claims were extinguished by a state court judgment approving a class settlement. In both eases, the judgment was given preclusive effect because the state and federal claims arose out of the identical factual predicate. In other words, had the judgment followed an adjudication rather than a settlement, it would necessarily have resolved the federal claims as a matter of issue preclusion.
In deciding the preclusive effect of the Delaware judgment at issue in this case, we need not decide whether to follow Nottingham Partners and Grimes in giving preclu-sive effect to a class action settlement that releases exclusively federal claims that turn on the same underlying facts as the state claims. All we need decide today is whether to break new ground in giving preclusive effect to a state court judgment that extinguishes exclusively federal claims that are factually unrelated to the state claims pleaded in the class action.27 The question of first impression confronting us is whether Congress, in denying state courts subject matter jurisdiction over 1934 Act claims, intended to leave state courts with the power to extinguish exclusively federal claims by approving a class action settlement that could not have been extinguished by adjudicating the class action. We can imagine no reason for imputing such an intent to Congress. We recognize that Congress could reasonably have been concerned about accommodating the interests of state courts in “achiev[ing] a comprehensive settlement that would prevent re-litigation of settled questions at the core of a class action,” TBK Partners, 675 F.2d at 462, but that concern would not be served by authorizing state courts to settle questions not at the core of a class action.
The very cases Matsushita relies upon counsel against an expansive state court power to release exclusively federal claims. In applying an issue preclusion test rather than a “same transaction” test, the eases embrace Judge Friendly’s common sense reasoning that a court’s jurisdiction to extinguish claims by class settlement should not exceed its jurisdiction to extinguish claims by adjudication. Ignoring this reasoning, Matsushita asks us to impose a “same transaction” test without offering logic or precedent in support *665of such a test.28 As we shall now show, the federal claims extinguished by the Delaware judgment could not have been extinguished by the issue preclusive effect of an adjudication of the state claims because, although the federal and state claims arose out of the same transaction, they are based upon different underlying facts.
2. The Disparity Between the State and Federal Claims
Matsushita makes no attempt to argue that any of the Williams Act claims of the Epstein plaintiffs share any predicate facts in common with the state law claims that formed the basis of the Delaware class action. Instead, Matsushita rests its preclusion argument solely on the fact that the claims arose out of the same transaction— Matsushita’s acquisition of MCA.
The gravamen of the federal class action is that Matsushita violated SEC Rules 14d-10 and 1 Ob-13 by offering greater value for the stock of Wasserman and Sheinberg than it offered for the stock of other shareholders.29 Thus the legal theory underlying the federal claims is that Matsushita breached a duty to shareholders that is imposed on tender offer-ors by federal securities laws. The relevant factual inquiry is whether Wasserman and Sheinberg received greater consideration than other MCA shareholders during the tender offer, section 14d — 10(a)(2), or a type of consideration not offered to other MCA shareholders. § 14d — 10(c)(1).
The gravamen of the state class action is that MCA directors breached their fiduciary duty of care to MCA as imposed by Delaware law. First, the Delaware plaintiffs claimed that MCA’s directors breached their fiduciary duty by failing to take steps to maximize shareholder value upon a change of corporate control as required by Revlon, Inc. v. MacAndrews & Forbes Holdings, 506 A.2d at 182. Plainly, this Revlon claim turns on different fact issues than the federal securities claims do. The question whether the MCA directors took the steps required by Revlon to assure maximum shareholder value is completely distinct from the question whether Matsushita violated the Williams Act by extending preferential treatment to some of the shareholders in making the tender offer.
The second state claim was that MCA’s directors breached their duty of candor by failing to disclose to the shareholders information concerning the merger such as the compensation packages MCA officers received under the terms of the merger agreement. But whether or not the MCA directors breached their fiduciary duty by failing to make disclosures has no bearing on whether directors received preferential treatment from Matsushita in violation of the Williams Act.30
The third state claim was that the defendants breached their fiduciary duty of loyalty by making preferential arrangements for themselves as part of the tender offer, and that Matsushita aided and abetted the MCA *666defendants in this breach. At first glance, this claim appears to be at least in the same ball park as the Rule 14d-10 claim. However, although this state claim is no doubt artfully drafted to resemble the 14d-10 claim, there is in fact no Delaware statutory or common law rule that prohibits a shareholder from obtaining the best deal for himself as part of a change of corporate control.31 Thus, adjudication of the claim would not have raised a question of fact whether Wasserman or Sheinberg received preferential treatment from Matsushita.
In sum, the state and federal claims are completely disparate. The only thing they share in common is that they arise out of the same transaction, which is the sole nexus Matsushita relies upon in arguing that the Delaware judgment’s release of the federal claims is entitled to full faith and credit.
Since the Delaware class action settlement judgment in this case attempted to extinguish exclusively federal claims which it could not have extinguished through adjudication, we hold that the decree exceeds the jurisdiction of the state court and, therefore, is not entitled to full faith and credit.
C. The Contract Bar Argument
Matsushita argues that regardless of the “technical res judicata effect” of the Delaware judgment, the Epstein plaintiffs are barred from litigating their Williams Act claims in federal court as a matter of contract. As Matsushita puts it, “just as an individual plaintiff may, as a part of a settlement in state court, agree to release exclusively federal claims, so, too, may a class give such a release as part of a judicially approved class settlement.” Brief of Matsushi-ta at 20.
This attempt to equate class settlements with the settlement of traditional litigation by individual parties falls short. Matsushita is correct, of course, that individual plaintiffs may release whatever claims they choose in settling traditional non-class litigation, whether or not related to the claims asserted in the pleadings. Because court approval of ordinary settlements is not required, the amount and form of consideration a party is willing to give or receive to settle a case is a matter of judicial indifference. Whether the forum court has jurisdiction over the claims a party chooses to release in settling traditional litigation is irrelevant. See Green v. Ancora-Citronelle Corp., 577 F.2d 1380, 1383 (9th Cir.1978) (an individual litigant who releases exclusively federal claims as part of a settlement of a state action may not relitigate them in federal court). Indeed, general releases are commonly exchanged between individual litigants in settlements.
However, the settlement of a class action is fundamentally different from the settlement of traditional litigation.32 First, Matsushita *667is simply wrong to assert that a class” may give a release as part of a settlement. A class is not an entity that has rights; the rights belong to the class members. But class members, unlike individual litigants in traditional lawsuits, are bound by the settlement even though they do not individually consent to its terms. Instead, consent is given by class representatives, who derive authority to represent members not by obtaining their consent, but by obtaining a court order designating them the representatives.
Second, class members may only give a release as part of a judicially approved settlement. Class representatives lack the “power to give a release of the class rights” on their own, absent judicial approval of the release and entry of a judgment. National Super Spuds, 660 F.2d at 18 (quoting Hau-dek, The Settlement and Dismissal of Stockholders’Action — Part II: The Settlement, 23 Sw.L.J. 765, 773 (1969)). Because class actions do not require the active participation of class members, class settlements pose a danger not present in traditional litigation: that representative plaintiffs and their lawyers will “endeavor to obtain a better settlement by sacrificing the claims of others at no cost to themselves.” Id. at 19. For this reason, “a court may not delegate to [class] counsel the performance of its [own] duty to protect the interests of absent class members.” Plummer v. Chemical Bank, 668 F.2d 654, 659 n. 4 (2d Cir.1982). Instead, in order to protect the rights of absent class members, the court must assume a far more active role than it typically plays in traditional litigation. A class action, thus, is less an individual lawsuit than “a quasi-administrative proceeding, conducted by the judge.” Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 809, 105 S.Ct. 2965, 2973, 86 L.Ed.2d 628 (1985).
This notwithstanding, Matsushita argues that the failure of the Epstein plaintiffs to opt out of the class is sufficient reason to bind them contractually to the release of their federal claims. In Matsushita’s view, the opportunity to opt out changes the essence of a class action settlement from an exercise of judicial power restrained by jurisdictional limits into an exercise of individual consent. This argument fails for two reasons. First, that the class members have a right to opt out does not diminish the extent to which a class action settlement is an exercise of judicial power. Regardless of whether class members are given opt-out rights, the court is still required to ensure that representation is adequate and that the settlement is fair to class members. See, e.g., Prezant v. Salton/Maxim Housewares, Inc., 636 A.2d 915 (Del.1994). The settlement of the class action is not an act of judicial mediation; it is an act of judicial power. Second, the “consent” purportedly given by class members in deciding not to opt out is hardly comparable to the consent given by individual plaintiffs in deciding to settle their own traditional lawsuits. Because opt-out rights are, as the Delaware Supreme Court has observed, “infrequently utilized and usually economically impracticable,” Prezant, 636 A.2d at 924, we would be blind to reality to think that any consent implied by class members in deciding not to opt out is comparable to the consent given by individual plaintiffs in settling their own lawsuits.
In sum, we reject Matsushita’s effort to equate class settlements with settlements of individual lawsuits. We believe that a state court cannot require passive class members to contractually release their exclusively federal claims in order to enjoy the benefits of a state class action, when the court has no jurisdictional power to dispose of those claims either directly or indirectly through the doctrine of issue preclusion.33
*668D. Conclusion
We respectfully decline to give full faith and credit to the Delaware judgment to the extent that it releases exclusively federal claims that depend upon different underlying facts than the state claims depend upon.34 Accordingly, we hold that the Delaware judgment does not preclude the Epstein class action.
V. Class Certification
We now turn to the question whether the district court abused its discretion in denying the Epstein plaintiffs’ motion for class certification. See Six (6) Mexican Workers v. Arizona Citrus Growers, 904 F.2d 1301, 1304 (9th Cir.1990) (class certification rulings reviewed for abuse of discretion). The district court gave no explanation for its ruling.
This case fits the requirements of both Rule 23(a) and Rule 23(b)(3) like a glove. The questions of Matsushita’s liability are the same for each of the 7,000 shareholders who tendered their MCA shares: (1) Did Matsushita violate section 14d-10 in acquiring Wasserman’s shares on terms that were different from the terms of the tender offer, and (2) Did Matsushita violate section 14d-10 in authorizing the post-tender offer payment of $21 million to Sheinberg? The claims of every tendering shareholder turn on identical facts and law—regardless of the identity or circumstances of the particular shareholder.
Under these circumstances, a class action would plainly be “superior to other available methods for the fair and efficient adjudication of the controversy,” because the “interest of members of the class in individually controlling the prosecution ... of separate actions,” is minimal, while the “desirability ... of concentrating the litigation of claims in the particular forum,” is compelling. Fed. R.Civ.P. 23(b)(3). Indeed, it is difficult to imagine a case where class certification would be more appropriate. Without it, thousands of identical complaints by former MCA shareholders would have to be filed— the very result the class action mechanism was designed to avoid.
Matsushita argues nonetheless that the claims of each MCA shareholder do not have enough in common to justify class certification because an “indispensable element of each plaintiffs claim is that he or she would have elected to take the preferred stock [given to Wasserman] if it was offered.” Brief of Matsushita at 52. Matsushita also argues that each individual shareholder’s per share damages will vary because whether “the preferred stock” that Wasserman received “could have constituted ‘higher’ consideration” than the cash all other shareholders received “would depend on the individual [tax] circumstances of each plaintiff.” Id. “Each plaintiff,” it claims, “would be subject to cross examination on these issues.” Id.
These arguments are meritless. In the first place, Matsushita cites neither law nor logic for the proposition that each shareholder is required to prove as an element of a Rule 14d-10 violation that she would have elected to take the preferred stock given to Wasserman had she been offered it. All a shareholder needs to prove to recover damages as a result of the Wasserman transaction is that she was not offered the consideration provided to Wasserman, and that the consideration Wasserman received was worth more than the consideration she received. Since it is undisputed that each shareholder (save Wasserman and perhaps Sheinberg) received $66 in cash and $5 worth of stock in WWOR-TV, each shareholder’s per share damages would be measured by the same yardstick: the extent to which the value of the preferred stock Wasserman received for each of his shares was greater than the $66 cash per share other shareholders received.35 The tax situation of each shareholder has *669nothing to do with it. Furthermore, “the amount of damages is invariably an individual question and does not defeat class action treatment.” Blackie v. Barrack, 524 F.2d 891, 905 (9th Cir.1975), cert. denied, 429 U.S. 816, 97 S.Ct. 57, 50 L.Ed.2d 75 (1976).
In arguing against class certification, Mat-sushita mentions only the Wasserman transaction. It says nothing about why a class should not be certified to litigate the claim that Matsushita made the $21 million payment to Sheinberg. If plaintiffs prevail on their claim that Matsushita violated Rule 14d-10 by paying Sheinberg a premium of $21 million for his 1,179,635 shares, the damages of each tendering shareholder would simply be $17.80 per share.
Finally, Matsushita argues that the named plaintiffs are unsuitable to serve as class representatives, citing the district court’s finding that plaintiff Lawrence Epstein prosecuted this action under “false pretenses” because he “deliberately concealed” from the court that he held his MCA stock in a tax-free IRA account. This finding, however, was first proposed by Matsushita in a motion filed well after the district court denied class certification, which means that it could not have informed in any way the district court’s denial of the Epstein plaintiffs’ motion. Moreover, we fail to see why the fact that Epstein held his shares in an IRA account should disqualify him from serving as a named plaintiff in this case. A shareholder’s individual tax circumstances, as we have stated, has no relevance to whether Matsushita violated Rule 14d-10 or how much in damages it would owe tendering MCA shareholders if it did.36
The record shows that the performance of the Epstein plaintiffs and their counsel in pursuing this litigation has been exemplary. We see no reason why they should not represent the class of MCA shareholders who tendered their shares. The district court’s order denying class certification is vacated, and the district court is instructed on remand to enter an order certifying the class.
VI. The Motion to Amend the Complaint
The Epstein plaintiffs argue that the district court abused its discretion in denying their motion to amend their complaint to allege that the tender offer documents released to shareholders were false and misleading, in violation of section 14(e) of the Williams Act.37 The district court failed to explain why it denied the motion. “In the absence of some statement of reasons or findings of fact showing bad faith or prejudice, we cannot [on the record before us] determine whether it was an abuse of discretion to deny” the motion for leave to amend the complaint. United States v. Webb, 655 F.2d 977, 980 (9th Cir.1981). Accordingly, we vacate the district court’s order denying the motion for leave to amend and remand to the district court for reconsideration of the motion.
VII. Conclusion
1. The district court’s order denying plaintiffs’ motion for partial summary judgment against Matsushita on the Wasserman claim is reversed, and the district court is instructed to grant such motion.
2. The district court’s order granting Matsushita’s motion for summary judgment is vacated.
3. The district court’s order dismissing plaintiffs’ claims against MCA, Wasserman, *670and Sheinberg for aiding and abetting is affirmed.
4.The district court’s orders denying the Epstein plaintiffs’ motions for class certification and for leave to amend their complaint are vacated and remanded.
. Tendering shareholders received $66 in cash and $5 worth of stock in WWOR-TV, a television station that was spun-off from MCA because of legal restrictions on foreign ownership of domestic broadcast stations.
. Wasserman also received for each of his shares the $5 worth of stock in WWOR-TV that all tendering shareholders received.
. Upon the death of Wasserman or his wife, the basis of all their community property, including the Matsushita preferred stock, would "step up” to its fair market value. See Internal Revenue Code §§ 1014, 2032.
. The consolidated appeals involve two sets of plaintiffs, each of whom filed separate complaints. Walter Minton ("Minton") brought this action in his individual capacity. Lawrence Epstein, John Linder, Jane Rockford, Maurice ¡Kar-lin, Ruth Karlin, Beth Karlin, and Bert Karlin ("Epstein plaintiffs”) brought suit both individually and on behalf of all MCA shareholders at the time of the tender offer.
. Plaintiffs also claim that the Wasserman and Sheinberg transactions violated SEC Rule 1 Ob-13, 17 C.F.R. § 240.10b-13 (1994). The district court dismissed this claim, pursuant to Fed. R.Civ.P. 12(b)(6), for want of a private right of action. On the facts of this case, any relief plaintiffs could obtain under Rule 10b-13 is also available under Rule 14d-10. See Field v. Trump, 850 F.2d 938, 946 n. 3 (2d Cir.1988), cert. denied, 489 U.S. 1012, 109 S.Ct. 1122, 103 L.Ed.2d 185 (1989). We therefore do not consider plaintiffs’ Rule 10b-l 3 claim.
. Plaintiffs did not seek summary judgment on their claim that the $21 million payment to Sheinberg also violated Rule 14d-10.
. Plaintiffs also argued in the district court and initially on appeal that MCA, Wasserman, and Sheinberg themselves should be held liable for aiding and abetting Matsushita's alleged violation of Rules 14d-10 and 10b-13. The district court dismissed this claim for failure to state a claim. Fed.R.Civ.P. 12(b)(6). In light of Central Bank v. First Interstate Bank, — U.S. -, 114 S.Ct. 1439, 128 L.Ed.2d 119 (1994), the Epstein plaintiffs now concede that the district court's judgment dismissing the aiding and abetting claims against MCA, Wasserman, and Sheinberg must be affirmed. Plaintiff Walter Minton, however, maintains that Wasserman and Sheinberg *649may still be held liable, notwithstanding Central Bank, for conspiring to violate Rules 14d-10 and 10b-13. We reject Minton’s attempt to avoid Central Bank's restrictions on private aiding and abetting actions by using the label “conspiracy." Nothing in the language of Rule 14d-10, which imposes restrictions on the activities of "bidder[s],” or in the language of Rule 10b-13, which speaks of “person[s] who make[] a cash tender offer,” suggests that liability may be imposed upon shareholders who are the beneficiaries of a bidder's violation of these rules. We therefore affirm the dismissal of plaintiffs’ claims against MCA, Wasserman and Sheinberg.
. In Robertson v. Dean Witter, 749 F.2d 530 (9th Cir.1984), we held that an agency rule contains a private right of action if Congress intended the statute under which it was promulgated to be privately enforceable. Id, at 536. While the parties agree that Rule 14d-10 was promulgated under sections 14(d)(6) and 14(d)(7) of the Williams Act, plaintiffs contend that the SEC also issued the Rule to implement section 14(e). Because we hold that Congress intended section 14(d)(7) to be privately enforceable, we need not address the relationship between section 14(e) and Rule 14d-10.
. In its brief, Matsushita did not argue that Rule 14d-10 does not contain a private right of action. Instead, it argued solely that Rule 10b-13 is not privately enforceable. Plaintiffs argue that Mat-sushita in fact conceded that Rule 14d-10 contains a private cause of action by writing that "[i]n addition to SEC enforcement actions and Rule 14d-10 claims, stockholders are further protected by Rule 10b-5....” Brief of Matsushita at 47 (emphasis added). Matsushita, however, denies that it made such a concession. See Supplemental Brief of Defendants-Appellants at 11 n. 11. In a letter submitted to the court several months after oral argument, Matsushita stated that Rule 14d-10 does not contain a private right of action, citing for support Central Bank v. First Interstate Bank, - U.S. -, 114 S.Ct. 1439, 128 L.Ed.2d 119 (1994), and referring to portions of its brief that discuss only Rule 10b-13. Nonetheless, because an argument concerning whether Rule 14d-10 provides a private right of action appears in the Brief of Wasserman and Sheinberg ("Wasserman Brief"), we will give Matsushita the benefit of the doubt and treat its 28(j) letter as incorporating that portion of the Wasserman Brief by reference.
.Section 14(d)(6) provides that when more shares are tendered than a bidder is "willing to take up and pay for, the securities taken up shall be taken up as nearly as may be pro rata ... according to the number of securities deposited by each depositor.” 15 U.S.C. § 78n(d)(6).
Section 14(d)(7) provides:
Where any person varies the terms of a tender offer or request or invitation for tenders before the expiration thereof by increasing the consideration offered to holders of such securities, such person shall pay the increased consideration to each security holder whose securities are taken up and paid for pursuant to the tender offer or request or invitation for tenders whether or not such securities have been taken up by such person before the variation of the tender offer or request or invitation.
15 U.S.C. § 78n(d)(7).
. In fact, in WPPSS we overruled two prior Ninth Circuit cases because of, among other reasons, their failure "to apply the Supreme Court's analysis in Cort v. Ash." WPPSS, 823 F.2d at 1351.
. Thus section 14(d)(7) stands in sharp contrast with the statute at issue in Middlesex County Sewerage Auth. v. Nat'l Sea Clammers Ass'n, 453 U.S. 1, 101 S.Ct. 2615, 69 L.Ed.2d 435 (1981), where the Court held that the existence of “elaborate enforcement provisions," including an express citizens suit provision, precluded the inference of additional private causes of action. Id. at 14, 101 S.Ct. at 2623. In contrast, the Williams Act itself contained no remedial provisions whatsoever, despite the fact that its “sole purpose" was the "protection of investors who are confronted with a tender offer.” Pryor, 794 F.2d at 56 (quoting Piper v. Chris-Craft Indus., 430 U.S. 1, 35, 97 S.Ct. 926, 946, 51 L.Ed.2d 124 (1977)). Matsushita's reliance on Nat’l Sea Clammers in this case is to no avail.
. The cases Matsushita does cite simply hold that courts are reluctant to infer causes of action absent an expression of congressional intent. We agree that congressional intent is the touchstone of an implied private right of action, and that a showing of “the utility of the proposed private right,” by itself, is insufficient to justify inferring a private cause of action. See Wasser-man Brief at 53. At the same time, the Court has stated over and over for the last fifteen years, in the cases Matsushita cites, that "the failure of Congress expressly to consider a private remedy is not inevitably inconsistent with an intent on its part to make such a remedy available,” Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S. 11, 18, 100 S.Ct. 242, 246, 62 L.Ed.2d 146 (1979), and that "congressional intent can be inferred from the language of the statute, the statutory structure, or some other source....” Karahalios v. Nat’l Fed. of Fed. Employees, Local 1263, 489 U.S. 527, 532-33, 109 S.Ct. 1282, *6521286-87, 103 L.Ed.2d 539 (1989) (internal citation and quotations omitted).
. Moreover, as the Third Circuit stated hi Polaroid, the Court has been sensitive to the legal context in which Congress legislated. See Franklin v. Gwinnett County Pub. Sch., 503 U.S. 60, 71-73, 112 S.Ct. 1028, 1036, 117 L.Ed.2d 208 (1992) (applying "same contextual approach used to justify an implied right of action" to determine scope of available remedies); Thompson v. Thompson, 484 U.S. 174, 180, 108 S.Ct. 513, 516-17, 98 L.Ed.2d 512 (1988) (examining context of legislation "with an eye toward determining Congress' perception of the law that it was shaping or reshaping"); Merrill Lynch, Pierce, Fenner & Smith v. Curran, 456 U.S. 353, 381, 102 S.Ct. 1825, 1840-41, 72 L.Ed.2d 182 (1982) (taking into account the “contemporary legal context" to determine whether Congress thought it was creating implied cause of action); Cannon v. University of Chicago, 441 U.S. 677, 698-99, 99 S.Ct. 1946, 1958-59, 60 L.Ed.2d 560 (1979) (same); id. at 718, 99 S.Ct. at 1968-69 (Rehnquist, J., concurring) (approving implication of private right of action under Title IX solely because, at time of passage, the Court’s implied right of action jurisprudence gave Congress reason to think that it need not decide private right of action question itself).
. Matsushita also argues that the Supreme Court's recent decision in Central Bank v. First Interstate Bank, - U.S. -, 114 S.Ct. 1439, 128 L.Ed.2d 119 (1994), requires federal courts to get out of the business of inferring private causes of action from statutes. Central Bank, however, is inapposite; it held simply that a private right of action is not available for aiding and abetting actions not authorized by statute.
. Rule 14d-10 provides in relevant part:
(a) No bidder shall make a tender offer unless:
(1) The tender offer is open to all security holders of the class of securities subject to the tender offer; and
(2) The consideration paid to any security holder pursuant to the tender offer is the highest consideration paid to any other security holder during such tender offer....
(c) Paragraph (a)(2) of this section shall not prohibit the offer of more than one type of consideration in a tender offer, Provided, That:
(1) Security holders are afforded equal right to elect among each of the types of consideration offered; and
(2) The highest consideration of each type paid to any security holder is paid to any other security holder receiving that type of consideration.
The SEC promulgated Rule 14d-10 in 1986, in part to codify its view that § 14(d)(7) of the Williams Act Amendments to the Securities Exchange Act of 1934 contains an all-holders and a best-price requirement. See SEC Release No. 34-22198, 1985 WL 61507, 1985 SEC LEXIS 1175 at *2 (July 1, 1985) (stating that under Commission's interpretation of § 14(d)(7) "(i) a tender offer must be extended to all holders of the class of securities which is the subject of the tender offer (the ‘all-holders requirement'); and (ii) all such holders must be paid the highest consideration offered under the tender offer (the 'best price rule')”).
. Rule 14d-10(c)(l) provides that "Paragraph (a)(2) of this section shall not prohibit the offer of more than one type of consideration in a tender offer, Provided, That: (1) Security holders are afforded equal right to elect among each of the types of consideration offered[.]" At oral argument, counsel for Wasserman and Sheinberg argued for the first time and without authority that section (c)(1) cannot be violated. Counsel contended that rather than an independent prohibition, section (c)(1) is a "savings clause" designed to create an exception to section (a)(2) of the Rule.
We cannot agree. When different types of consideration are offered to different holders of the same class of security, it may be difficult for plaintiffs to prove that one package was more valuable than the other, as section (a)(2) requires. Section (c)(1) obviates the need for plaintiffs to do so. It is a prophylactic rule that entitles a court to presume, if a tender offer fails to make the type of consideration offered to some security holders available to all, that the selectively-offered consideration was "higher consideration” for purposes of section (a)(2). This reading of section (c)(1) is supported by the SEC’s comments in releasing Rule 14d-10, to which we are obliged to give substantial deference. See Stinson v. United States, - U.S. -, -, 113 S.Ct. 1913, 1919, 123 L.Ed.2d 598 (1993) (agency’s interpretation of own regulations must be given "controlling weight unless it is plainly erroneous or inconsistent with the regulation”); Leslie Salt Co. v. United States, 896 F.2d 354, 357 (9th Cir.1990), (agency’s interpretation of its own regulations is entitled to deference), cert. denied, 498 U.S. 1126, 111 S.Ct. 1089, 112 L.Ed.2d 1194 (1991). In its release, the SEC stated, under the heading "Alternative Types of Consideration," that "when more than one type of consideration is offered ... [s]ecurity holders ... must be afforded the right to elect among all types of consideration offered.” SEC Release No. 34-23421, 1986 WL 71340, 1986 SEC LEXIS 1179 at *28 (July 11, 1986).
. Under Rule 14d-2, a tender offer begins for purposes of section 14(d)(7) not from the moment of the offer's announcement (as under Rule 1 Ob-13), but at 12:01 a.m. on the earlier of the date the tender offer is publicly announced or sent to securities holders. See Rule 14d-2(a); Rule 14d-2(b)(2)(ii); SEC Release No. 34-22198, July 1, 1985, 1985 WL 61507, 1985 SEC LEXIS at *10, 13. Matsushita claims that this provision of Rule 14d-2 does not apply to claims asserted under Rule 14d-10, as opposed to under section 14(d)(7), but they strain to do so. Rule 14d-10 was promulgated to codify the all-holders and best-price requirements in § 14(d)(7), and the SEC stated when proposing the Rule that the § 14(d)(7) time frame would carry over to the Rule. Id.
. If an ostensibly private agreement that pays a particular shareholder a sum greater than the tender price is deemed a part of the tender offer, the tender offer does not end, by definition, until that agreement is performed. Performance of the agreement, including payment of the shareholder, will therefore take place "during the tender offer,” in violation of Rule 14d — 10(a)(2). Similarly, if one shareholder is offered a different type of consideration than other shareholders in a transaction that is determined to be a part of the tender offer, the bidder has offered more than one type of consideration "in a tender offer" without affording each shareholder the right to choose the package she prefers, in violation of Rule 14d — 10(c)(1).
. Plaintiffs alleged in their complaint that Was-serman received consideration of greater value per share than other shareholders. Matsushita did not dispute this allegation in its motion for summary judgment. Moreover, in opposing Matsushita’s summary judgment motion, plaintiffs stated, on the basis of the calculations contained in Matsushita’s expert witness’ affidavit and on the fact that dividend payments are not tax-deductible, that Wasserman's preferred stock may have cost Matsushita more than $66 per share in cash. Plaintiffs also point to their own expert's answers to interrogatories that state that Wasserman's preferred stock was in fact worth more than $66 per share. These answers to interrogatories are not in the record because Matsushita did not controvert plaintiffs’ allegation that Wasserman’s preferred stock was worth more that $66 per MCA share he exchanged. On this record, we believe that plaintiffs have raised a genuine issue of fact as to whether Matsushita gave Wasserman consideration worth more than $66 per share.
Matsushita also asserts that to recover damages, plaintiffs must prove that they would have elected the consideration package given solely to Wasserman. The plain language of Rule 14d-10 is to the contrary. It requires plaintiffs to prove, as a prerequisite to recovering damages, simply that the consideration paid to Wasserman was "higher” than that paid to other shareholders. We express no view at this time as to the value, if any, that may be attributed to tax benefits Was-serman may have received.
. The only authority Matsushita cites for this argument is Kramer v. Time Warner, 937 F.2d 767 (2d Cir.1991). Kramer, however, is inappo-site. That case involved Time Incorporated's two-step acquisition of Warner Communications: first a tender offer, then a statutory merger. Time acquired 51% of Warner's common stock pursuant to the tender offer and the remaining 49% pursuant to the second-step statutory merger.
The Second Circuit held that Rule 14d-10 was not violated when Warner paid its top executives different consideration than ordinary shareholders received in the second-step statutory merger. The court reasoned that a statutory merger following a successful tender offer for 51% of a target's shares is not a tender offer for purposes of the Williams Act. Id. at 779. Distinguishing Field v. Trump, 850 F.2d at 938 (holding that claim exists under Williams Act when offeror withdraws initial tender offer on the same evening he privately secures dissident shareholders' stock only to then renew tender offer next morning), the Second Circuit held: "The claim that the Williams Act applies to second-step statutory mergers is ... meritless.” Id. The instant case is distinguishable from Kramer for the same reason Field v. Trump is. As in Field v. Trump, no statutory merger, only a tender offer, is involved here.
. The settlement agreement permitted state class members to opt out. Although the Epstein plaintiffs did not opt out, 19 shareholders, including plaintiff-appellant Walter Minton, chose to do so. It is undisputed that shareholders who opted out may continue to pursue their federal claims regardless of the preclusive effect of the Delaware settlement on the Epstein class action.
. Class counsel's request of over $691,000 in fees was reduced by the Vice Chancellor to $250,000, which in turn reduced the amount available for class members to $1,750,000.
. Notice describing the settlement was sent to all class members, which included the Epstein plaintiffs. None of the Epstein plaintiffs appeared in the Delaware court to object to the terms of the settlement.
. The Delaware Court of Chancery determined that it was jurisdictionally competent to enter a judgment settling the class action. As a general rule, we are not at liberty to entertain a collateral attack on a state court's determination of its own subject matter jurisdiction. See Durfee v. Duke, 375 U.S. at 111, 84 S.Ct. at 245. As the Supreme Court went on to say in Durfee, however, "the general rule of finality of jurisdictional determinations is not without exceptions. Doctrines of federal pre-emption or sovereign immunity may in some contexts be controlling.” Id. at 114, 84 S.Ct. at 246. The Durfee Court concluded that when "the policy underlying the doctrine of res judicata is outweighed by the policy against permitting the court to act beyond its own jurisdiction,” collateral attack on a state *662court’s exercise of jurisdiction is warranted. Id. at 114 n. 12, 84 S.Ct. at 247 n. 12.
In this case the policy of finality of judgments is plainly outweighed by the federal interest in reviewing a state court's exercise of judicial power over claims which Congress has vested exclusively in federal courts. Section 14(d)(7) of the Williams Act, granting exclusive jurisdiction to the federal courts, embodies beyond cavil, "a highly important federal policy calling for unified and comprehensive proceedings in federal court.” Wright & Miller § 4428, at 285. See Murphy v. Gallagher, 761 F.2d 878, 885 (2d Cir.1985). When Congress has expressly denied state courts jurisdiction over an important area of federal law, and a state court nonetheless exercises judicial power over it, the potential for disrupting the distribution of governmental power is great. In such a context, the doctrine of federal pre-emption would control, and collateral attack on the state court’s exercise of jurisdiction would be appropriate. See Durfee, 375 U.S. at 114, 84 S.Ct. at 246-47.
. Matsushita also relies on one case, Abramson v. Pennwood Investment Corp., 392 F.2d 759 (2d Cir.1968), which did not involve a class action. We deal with this case in the section addressing Matsushita's contract bar argument. See infra note 32.
. See infra part IV.B.2.
. The "arising out of the same transaction" test is, of course, used in claim preclusion doctrine. See Restatement (Second) of Judgments § 24 (1981). However, Matsushita, as it must, expressly disclaims reliance on claim preclusion. See Clark v. Watchie, 513 F.2d 994, 997 (9th Cir.1975) (state court judgment can have no claim preclusive effect on exclusively federal securities claims).
. Wasserman, Sheinberg, and MCA were also named as defendants and charged with aiding and abetting Matsushita. See supra note 7. Because the Epstein plaintiffs concede that the district court's dismissal of this claim was appropriate, we do not address this claim further.
. The Delaware plaintiffs also alleged that MCA and Matsushita "subjected MCA to potential liability under Rules 14d-7, 14d-10, 10b-13, and 13e-3, ... thereby causing MCA to expend and waste corporate assets....” Amended Complaint at 11. There is, to state the obvious, no cause of action in Delaware for violating the federal securities laws, and there would thus be no issue preclusion if these claims were litigated on the merits in Delaware. Indeed, the Delaware Vice Chancellor did not address the settlement value of these claims in his opinion. The fact that the lawyers for the Delaware plaintiffs inserted such a claim into their amended complaint (not to mention that they did so the day after the Epstein action was filed) does little more than make readily apparent the extent to which the state suit was structured with an eye toward the preclusive effect it might have on the federal claims pending in the Central District of California.
. In rejecting the first proposed Delaware settlement, the Chancery Court observed that a shareholder is only prohibited from negotiating a preferential deal for himself if he uses inside information to do so. In re MCA, Inc. Shareholders Litig., 598 A.2d at 694. Indeed, as Wasserman and Sheinberg rightly observe, under Delaware law, "Wasserman would have been fully entitled to have held out for a premium in return for agreeing to sell his own MCA shares to Mat-sushita." Brief of Wasserman and Sheinberg at 14; see also Citron v. Steego Corp., 1988 WL 94738, at *8, 1988 Del. Ch. LEXIS 119, at *21 (Del.Ch. Sept. 9, 1988) (Allen, C.) ("Our law does not regard a control premium as a corporate asset that must be shared among all shareholders"). Rule 14d-10, in contrast, does prohibit a shareholder from receiving more, or even different, consideration in the course of a tender offer. Unlike Delaware law, it does prohibit the payment of control premiums in the course of a tender offer.. Indeed, Rule 14d-10 was promulgated precisely to prohibit practices that Delaware permits. See Polaroid Corp. v. Disney, 862 F.2d at 992.
. Furthermore, class actions are different from derivative actions, such as that at issue in Abramson v. Penwood Inv. Corp., 392 F.2d 759 (2d Cir.1968). Matsushita relies on Abramson as authority for the proposition that a class action settlement may release exclusively federal claims regardless of the limited subject matter jurisdiction of the forum court. However, Abramson was not a class action. Rather, it involved a shareholder derivative action that included the release of exclusively federal claims owned by the real party in interest, the corporation. This case is inapposite because of the essential difference between a shareholder derivative action and a class action. As Judge Friendly pointed out in a later case, there is a
fundamental difference between derivative suits and class actions. The plaintiff in a derivative suit is suing on behalf of a corporation. *667In a derivative suit settlement ... the legal entity whose unpleaded claims are being released — the corporation — is a party to the settlement. The corporation has the power to release its claims whether asserted in the complaint or not. Although there are other considerations that require a court to scrutinize carefully the scope of the releases, the court may be confident that the parties have the power to make them.
National Super Spuds, 660 F.2d at 18-19 (emphasis added).
. Whether the release of federal securities claims in this case is a valid contract defense to the Epstein class action is a question of federal law. See, e.g., Newton v. Rumery, 480 U.S. 386, 392. 107 S.Ct. 1187, 1191-92, 94 L.Ed.2d 405 *668(1987). A contract releasing federal statutory rights need not be enforced by a federal court if "the interest in its enforcement is outweighed in the circumstances by a public policy harmed by enforcement of the agreement.” Id.
. Again, we do not decide whether a state court may release exclusively federal claims in a class action settlement judgment when those claims depend on the same facts as state law claims before the court.
. Recall that Wasserman also received for each of his shares the $5 worth of stock in WWOR-TV.
. In addition, the record lends no support to the notion that Epstein misbehaved in pursuing this litigation. The district court’s finding that he prosecuted this action under false pretenses is clearly erroneous. There is no evidence that Epstein intentionally hid the fact that he kept his shares in an IRA account. Matsushita deposed Epstein twice, but failed to ask him how he held his MCA shares. Further, prior to his second deposition, Epstein produced transaction slips which indicated that he held his MCA shares in an IRA account.
. Plaintiff Walter Minton argues on appeal that the district court abused its discretion in denying his motion to amend his complaint to add a section 14(d)(7) claim. Because there is no material difference between a section 14(d)(7) and a Rule 14d-10 claim, the question whether the district court abused its discretion in denying Minton's motion for leave to amend is moot.