delivered the opinion of the Court.

Section 14(a) of the Securities Exchange Act of 1934, 48 Stat. 895, 15 U. S. C. § 78n(a), authorizes the Securities and Exchange Commission (SEC) to adopt rules for the solicitation of proxies, and prohibits their violation.1 In J. I. Case Co. v. Borak, 377 U. S. 426 (1964), we first recognized an *1087implied private right of action for the breach of § 14(a) as implemented by SEC Rule 14a-9, which prohibits the solicitation of proxies by means of materially false or misleading statements.2

The questions before us are whether a statement couched in conclusory or qualitative terms purporting to explain directors’ reasons for recommending certain corporate action can be materially misleading within the meaning of Rule 14a-9, and whether causation of damages compensable under § 14(a) can be shown by a member of a class of minority shareholders whose votes are not required by law or corporate bylaw to authorize the corporate action subject to the proxy solicitation. We hold that knowingly false statements of reasons may be actionable even though conclusory in form, but that respondents have failed to demonstrate the equitable basis required to extend the § 14(a) private action to such shareholders when any indication of congressional intent to do so is lacking.

I

In December 1986, First American Bankshares, Inc. (FAB I), a bank holding company, began a “freeze-out” merger, in which the First American Bank of Virginia (Bank) eventually merged into Virginia Bankshares, Inc. (VBI), a *1088wholly owned subsidiary of FABI. VBI owned 85% of the Bank’s shares, the remaining 15% being in the hands of some 2,000 minority shareholders. FABI hired the investment banking firm of Keefe, Bruyette & Woods (KBW) to give an opinion on the appropriate price for shares of the minority holders, who would lose their interests in the Bank as a result of the merger. Based on market quotations and unverified information from FABI, KBW gave the Bank’s executive committee an opinion that $42 a share would be a fair price for the minority stock. The executive committee approved the merger proposal at that price, and the full board followed suit.

Although Virginia law required only that such a merger proposal be submitted to a vote at a shareholders’ meeting, and that the meeting be preceded by circulation of a statement of information to the shareholders, the directors nevertheless solicited proxies for voting on the proposal at the annual meeting set for April 21, 1987.3 In their solicitation, the directors urged the proposal’s adoption and stated they had approved the plan because of its opportunity for the minority shareholders to achieve a “high” value, which they elsewhere described as a “fair” price, for their stock.

Although most minority shareholders gave the proxies requested, respondent Sandberg did not, and after approval of the merger she sought damages in the United States District Court for the Eastern District of Virginia from VBI, FABI, and the directors of the Bank. She pleaded two counts, one for soliciting proxies in violation of § 14(a) and Rule 14a-9, and the other for breaching fiduciary duties owed to the minority shareholders under state law. Under the first count, Sandberg alleged, among other things, that the directors had not believed that the price offered was high or that the terms *1089of the merger were fair, but had recommended the merger only because they believed they had no alternative if they wished to remain on the board. At trial, Sandberg invoked language from this Court’s opinion in Mills v. Electric Auto-Lite Co., 396 U. S. 375, 385 (1970), to obtain an instruction that the jury could find for her without a showing of her own reliance on the alleged misstatements, so long as they were material and the proxy solicitation was an “essential link” in the merger process.

The jury’s verdicts were for Sandberg on both counts, after finding violations of Rule 14a-9 by all defendants and a breach of fiduciary duties by the Bank’s directors. The jury awarded Sandberg $18 a share, having found that she would have received $60 if her stock had been valued adequately.

While Sandberg’s case was pending, a separate action on similar allegations was brought against petitioners in the United States District Court for the District of Columbia by several other minority shareholders including respondent Weinstein, who, like Sandberg, had withheld his proxy. This case was transferred to the Eastern District of Virginia. After Sandberg’s action had been tried, the Weinstein respondents successfully pleaded collateral estoppel to get summary judgment on liability.

On appeal, the United States Court of Appeals for the Fourth Circuit affirmed the judgments, holding that certain statements in the proxy solicitation were materially misleading for purposes of the Rule, and that respondents could maintain their action even though their votes had not been needed to effectuate the merger. 891 F. 2d 1112 (1989).4 We granted certiorari because of the importance of the issues presented. 495 U. S. 903 (1990).

*1090I — I I — I

The Court of Appeals affirmed petitioners liability for two statements found to have been materially misleading in violation of § 14(a) of the Act, one of which was that “The Plan of Merger has been approved by the Board of Directors because it provides an opportunity for the Bank’s public shareholders to achieve a high value for their shares.” App. to Pet. for Cert. 53a. Petitioners argue that statements of opinion or belief incorporating indefinite and unverifiable expressions cannot be actionable as misstatements of material fact within the meaning of Rule 14a-9, and that such a declaration of opinion or belief should never be actionable when placed in a proxy solicitation incorporating statements of fact sufficient to enable readers to draw their own, independent conclusions.

A

We consider first the actionability per se of statements of reasons, opinion, or belief. Because such a statement by definition purports to express what is consciously on the speaker’s mind, we interpret the jury verdict as finding that the directors’ statements of belief and opinion were made with knowledge that the directors did not hold the beliefs or opinions expressed, and we confine our discussion to statements so made.5 That such statements may be materially significant raises no serious question. The meaning of the materiality requirement for liability under § 14(a) was discussed at some length in TSC Industries, Inc. v. Northway, Inc., 426 U. S. 438 (1976), where we held a fact to be material “if there is a substantial likelihood that a reasonable shareholder would consider it important in deciding how to vote.” Id., at 449. We think there is no room to deny that a statement of belief by corporate directors about a recommended course of action, or an explanation of their reasons for recommending *1091it, can take on just that importance. Shareholders know that directors usually have knowledge and expertness far exceeding the normal investor’s resources, and the directors’ perceived superiority is magnified even further by the common knowledge that state law customarily obliges them to exercise their judgment in the shareholders’ interest. Cf. Day v. Avery, 179 U. S. App. D. C. 63, 71, 548 F. 2d 1018, 1026 (1976) (action for misrepresentation). Naturally, then, the shareowner faced with a proxy request will think it important to know the directors’ beliefs about the course they recommend and their specific reasons for urging the stockholders to embrace it.

B

1

But, assuming materiality, the question remains whether statements of reasons, opinions, or beliefs are statements “with respect to . . . material fact[s]” so as to fall within the strictures of the Rule. Petitioners argue that we would invite wasteful litigation of amorphous issues outside the readily provable realm of fact if we were to recognize liability here on proof that the directors did not recommend the merger for the stated reason, and they cite the authority of Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723 (1975), in urging us to recognize sound policy grounds for placing such statements outside the scope of the Rule.

We agree that Blue Chip Stamps is instructive, as illustrating a line between what is and is not manageable in the litigation of facts, but do not read it as supporting petitioners’ position. The issue in Blue Chip Stamps was the scope of the class of plaintiffs entitled to seek relief under an implied private cause of action for violating § 10(b) of the Act, prohibiting manipulation and deception in the purchase or sale of certain securities, contrary to Commission rules. This Court held against expanding the class from actual buyers and sellers to include those who rely on deceptive sales practices by taking no action, either to sell what they own or *1092to buy what they do not. We observed that actual sellers and buyers who sue for compensation must identify a specific number of shares bought or sold in order to calculate and limit any ensuing recovery. Id., at 734. Recognizing liability to merely would-be investors, however, would have exposed the courts to litigation unconstrained by any such anchor in demonstrable fact, resting instead on a plaintiff’s “subjective hypothesis” about the number of shares he would have sold or purchased. Id., at 734-735. Hindsight’s natural temptation to hypothesize boldness would have magnified the risk of nuisance litigation, which would have been compounded both by the opportunity to prolong discovery and by the capacity of claims resting on undocumented personal assertion to resist any resolution short of settlement or trial. Such were the premises of policy, added to those of textual analysis and precedent, on which Blue Chip Stamps deflected the threat of vexatious litigation over “many rather hazy issues of historical fact the proof of which depended almost entirely on oral testimony.” Id., at 743.

Attacks on the truth of directors’ statements of reasons or belief, however, need carry no such threats. Such statements are factual in two senses: as statements that the directors do act for the reasons given or hold the belief stated and as statements about the subject matter of the reason or belief expressed. In neither sense does the proof or disproof of such statements implicate the concerns expressed in Blue Chip Stamps. The root of those concerns was a plaintiff’s capacity to manufacture claims of hypothetical action, unconstrained by independent evidence. Reasons for directors’ recommendations or statements of belief are, in contrast, characteristically matters of corporate record subject to documentation, to be supported or attacked by evidence of historical fact outside a plaintiff’s control. Such evidence would include not only corporate minutes and other statements of the directors themselves, but circumstantial evidence bearing on the facts that would reasonably underlie *1093the reasons claimed and the honesty of any statement that those reasons are the basis for a recommendation or other action, a point that becomes especially clear when the reasons or beliefs go to valuations in dollars and cents.

It is no answer to argue, as petitioners do, that the quoted statement on which liability was predicated did not express a reason in dollars and cents, but focused instead on the “indefinite and unverifiable” term, “high” value, much like the similar claim that the merger’s terms were “fair” to shareholders.6 The objection ignores the fact that such conclu-sory terms in a commercial context are reasonably understood to rest on a factual basis that justifies them as accurate, the absence of which renders them misleading. Provable facts either furnish good reasons to make a conclusory commercial judgment, or they count against it, and expressions of such judgments can be uttered with knowledge of truth or falsity just like more definite statements, arid defended or attacked through the orthodox evidentiary process that either substantiates their underlying justifications or tends to disprove their existence. In addressing the analogous issue in an action for misrepresentation, the court in Day v. Avery, 179 U. S. App. D. C. 63, 548 F. 2d 1018 (1976), *1094for example, held that a statement by the executive committee of a law firm that no partner would be any “worse off” solely because of an impending merger could be found to be a material misrepresentation. Id., at 70-72, 548 F. 2d, at 1025-1027. Cf. Vulcan Metals Co. v. Simmons Mfg. Co., 248 F. 853, 856 (CA2 1918) (L. Hand, J.) (“An opinion is a fact. . . . When the parties are so situated that the buyer may reasonably rely upon the expression of the seller’s opinion, it is no excuse to give a false one”); W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Law of Torts § 109, pp. 760-762 (5th ed. 1984). In this case, whether $42 was “high,” and the proposal “fair” to the minority shareholders, depended on whether provable facts about the Bank’s assets, and about actual and potential levels of operation, substantiated a value that was above, below, or more or less at the $42 figure, when assessed in accordance with recognized methods of valuation.

Respondents adduced evidence for just such facts in proving that the statement was misleading about its subject matter and a false expression of the directors’ reasons. Whereas the proxy statement described the $42 price as offering a premium above both book value and market price, the evidence indicated that a calculation of the book figure based on the appreciated value of the Bank’s real estate holdings eliminated any such premium. The evidence on the significance of market price showed that KBW had conceded that the market was closed, thin, and dominated by FABI, facts omitted from the statement. There was, indeed, evidence of a “going concern” value for the Bank in excess of $60 per share of common stock, another fact never disclosed. However conclusory the directors’ statement may have been, then, it was open to attack by garden-variety evidence, subject neither to a plaintiff’s control nor ready manufacture, and there was no undue risk of open-ended liability or uncontrollable litigation in allowing respondents the opportunity *1095for recovery on the allegation that it was misleading to call $42 “high.”

This analysis comports with the holding that marked our nearest prior approach to the issue faced here, in TSC Industries, 426 U. S., at 454-455. There, to be sure, we reversed summary judgment for a Borak plaintiff who had sued on a description of proposed compensation for minority shareholders as offering a “substantial premium over current market values.” But we held only that on the case’s undisputed facts the conclusory adjective “substantial” was not materially misleading as a necessary matter of law, and our remand for trial assumed that such a description could be both materially misleading within the meaning of Rule 14a-9 and actionable under § 14(a). See TSC Industries, supra, at 458-460, 463-464.

2

Under § 14(a), then, a plaintiff is permitted to prove a specific statement of reason knowingly false or misleadingly incomplete, even when stated in conclusory terms. In reaching this conclusion we have considered statements of reasons of the sort exemplified here, which misstate the speaker’s reasons and also mislead about the stated subject matter (e. g., the value of the shares). A statement of belief may be open to objection only in the former respect, however, solely as a misstatement of the psychological fact of the speaker’s belief in what he says. In this case, for example, the Court of Appeals alluded to just such limited falsity in observing that “the jury was certainly justified in believing that the directors did not believe a merger at $42 per share was in the minority stockholders’ interest but, rather, that they voted as they did for other reasons, e. g., retaining their seats on the board.” 891 F. 2d, at 1121.

The question arises, then, whether disbelief, or undisclosed belief or motivation, standing alone, should be a sufficient basis to sustain an action under § 14(a), absent proof by the sort of objective evidence described above that the *1096statement also expressly or impliedly asserted something false or misleading about its subject matter. We think that proof of mere disbelief or belief undisclosed should not suffice for liability under § 14(a), and if nothing more had been required or proven in this case, we would reverse for that reason.

On the one hand, it would be rare to find a case with evidence solely of disbelief or undisclosed motivation without further proof that the statement was defective as to its subject matter. While we certainly would not hold a director’s naked admission of disbelief incompetent evidence of a proxy statement’s false or misleading character, such an unusual admission will not very often stand alone, and we do not substantially narrow the cause of action by requiring a plaintiff to demonstrate something false or misleading in what the statement expressly or impliedly declared about its subject.

On the other hand, to recognize liability on mere disbelief or undisclosed motive without any demonstration that the proxy statement was false or misleading about its subject would authorize § 14(a) litigation confined solely to what one skeptical court spoke of as the “impurities” of a director’s “unclean heart.” Stedman v. Storer, 308 F. Supp. 881, 887 (SDNY 1969) (dealing with § 10(b)). This, we think, would cross the line that Blue Chip Stamps sought to draw. While it is true that the liability, if recognized, would rest on an actual, not hypothetical, psychological fact, the temptation to rest an otherwise nonexistent § 14(a) action on psychological enquiry alone would threaten just the sort of strike suits and attrition by discovery that Blue Chip Stamps sought to discourage. We therefore hold disbelief or undisclosed motivation, standing alone, insufficient to satisfy the element of fact that must be established under § 14(a).

C

Petitioners’ fail-back position assumes the same relationship between a conclusory judgment and its underlying facts *1097that we described in Part II-B-1, supra. Thus, citing Radol v. Thomas, 534 F. Supp. 1302, 1315, 1316 (SD Ohio 1982), petitioners argue that even if conclusory statements of reason or belief can be actionable under § 14(a), we should confine liability to instances where the proxy material fails to disclose the offending statement’s factual basis. There would be no justification for holding the shareholders entitled to judicial relief, that is, when they were given evidence that a stated reason for a proxy recommendation was misleading and an opportunity to draw that conclusion themselves.

The answer to this argument rests on the difference between a merely misleading statement and one that is materially so. While a misleading statement will not always lose its deceptive edge simply by joinder with others that are true, the true statements may discredit the other one so obviously that the risk of real deception drops to nil. Since liability under § 14(a) must rest not only on deceptiveness but materiality as well (i. e., it has to be significant enough to be important to a reasonable investor deciding how to vote, see TSC Industries, 426 U. S., at 449), petitioners are on perfectly firm ground insofar as they argue that publishing accurate facts in a proxy statement can render a misleading proposition too unimportant to ground liability.

But not every mixture with the true will neutralize the deceptive. If it would take a financial analyst to spot the tension between the one and the other, whatever is misleading will remain materially so, and liability should follow. Gerstle v. Gamble-Skogmo, Inc., 478 F. 2d 1281, 1297 (CA2 1973) (“[I]t is not sufficient that overtones might have been picked up by the sensitive antennae of investment analysts’’). Cf. Milkovich v. Lorain Journal Co., 497 U. S. 1, 18-19 (1990) (a defamatory assessment of facts can be actionable even if the facts underlying the assessment are accurately presented). The point of a proxy statement, after all, should be to inform, not to challenge the reader’s critical wits. Only when the inconsistency would exhaust the misleading conclu*1098sion’s capacity to influence the reasonable shareholder would a § 14(a) action fail on the element of materiality.

Suffice it to say that the evidence invoked by petitioners in the instant case fell short of compelling the jury to find the facial materiality of the misleading statement neutralized. The directors claim, for example, to have made an explanatory disclosure of further reasons for their recommendation when they said they would keep their seats following the merger, but they failed to mention what at least one of them admitted in testimony, that they would have had no expectation of doing so without supporting the proposal, App. 281-282.7 And although the proxy statement did speak factually about the merger price in describing it as higher than share prices in recent sales, it failed even to mention the closed market dominated by FABI. None of these disclosures that the directors point to was, then, anything more than a half-truth, and the record shows that another fact statement they invoke was arguably even worse. The claim that the merger price exceeded book value was controverted, as we have seen already, by evidence of a higher book value than the directors conceded, reflecting appreciation in the Bank’s real estate portfolio. Finally, the solicitation omitted any mention of the Bank’s value as a going concern at more than $60 a share, as against the merger price of $42. There was, in sum, no more of a compelling case for the statement’s immateriality than for its accuracy.

*1099HH b-i h — I

The second issue before us, left open in Mills v. Electric Auto-Lite Co., 396 U. S., at 385, n. 7, is whether causation of damages compensable through the implied private right of action under § 14(a) can be demonstrated by a member of a class of minority shareholders whose votes are not required by law or corporate bylaw to authorize the transaction giving rise to the claim.8 J. I. Case Co. v. Borak, 377 U. S. 426 (1964), did not itself address the requisites of causation, as such, or define the class of plaintiffs eligible to sue under § 14(a). But its general holding, that a private cause of action was available to some shareholder class, acquired greater clarity with a more definite concept of causation in Mills, where we addressed the sufficiency of proof that misstatements in a proxy solicitation were responsible for damages claimed from the merger subject to complaint.

Although a majority stockholder in Mills controlled just over half the corporation’s shares, a two-thirds vote was needed to approve the merger proposal. After proxies had been obtained, and the merger had carried, minority shareholders brought a Borak action. Mills, 396 U. S., at 379. The question arose whether the plaintiffs’ burden to demonstrate causation of their damages traceable to the § 14(a) violation required proof that the defect in the proxy solicitation had had “a decisive effect on the voting.” Id., at 385. The Mills Court avoided the evidentiary morass that would have *1100followed from requiring individualized proof that enough minority shareholders had relied upon the misstatements to swing the vote. Instead, it held that causation of damages by a material proxy misstatement could be established by showing that minority proxies necessary and sufficient to authorize the corporate acts had been given in accordance with the tenor of the solicitation, and the Court described such a causal relationship by calling the proxy solicitation an “essential link in the accomplishment of the transaction. ” Ibid. In the case before it, the Court found the solicitation essential, as contrasted with one addressed to a class of minority shareholders without votes required by law or bylaw to authorize the action proposed, and left it for another day to decide whether such a minority shareholder could demonstrate causation. Id., at 385, n. 7.

In this case, respondents address Mills’ open question by proffering two theories that the proxy solicitation addressed to them was an “essential link” under the Mills causation test.9 They argue, first, that a link existed and was essential simply because VBI and FAB I would have been unwilling to proceed with the merger without the approval manifested by the minority shareholders’ proxies, which would not have been obtained without the solicitation’s express mis*1101statements and misleading omissions. On this reasoning, the causal connection would depend on a desire to avoid bad shareholder or public relations, and the essential character of the causal link would stem not from the enforceable terms of the parties’ corporate relationship, but from one party’s apprehension of the ill will of the other.

In the alternative, respondents argue that the proxy statement was an essential link between the directors’ proposal and the merger because it was the means to satisfy a state statutory requirement of minority shareholder approval, as a condition for saving the merger from voidability resulting from a conflict of interest on the part of one of the Bank’s directors, Jack Beddow, who voted in favor of the merger while also serving as a director of FAB I. Brief for Respondents 43-44, 45-46. Under the terms of Va. Code Ann. § 13.1-691(A) (1989), minority approval after disclosure of the material facts about the transaction and the director’s interest was one of three avenues to insulate the merger from later attack for conflict, the two others being ratification by the Bank’s directors after like disclosure and proof that the merger was fair to the corporation. On this theory, causation would depend on the use of the proxy statement for the purpose of obtaining votes sufficient to bar a minority shareholder from commencing proceedings to declare the merger void.10

*1102Although respondents have proffered each of these theories as establishing a chain of causal connection in which the proxy statement is claimed to have been an “essential link,” neither theory presents the proxy solicitation as essential in the sense of Mills’ causal sequence, in which the solicitation links a directors’ proposal with the votes legally required to authorize the action proposed. As a consequence, each theory would, if adopted, extend the scope of Borak actions beyond the ambit of Mills and expand the class of plaintiffs entitled to bring Borak actions to include shareholders whose initial authorization of the transaction prompting the proxy solicitation is unnecessary.

Assessing the legitimacy of any such extension or expansion calls for the application of some fundamental principles governing recognition of a right of action implied by a federal statute, the first of which was not, in fact, the considered focus of the Borak opinion. The rule that has emerged in the years since Borak and Mills came down is that recognition of any private right of action for violating a federal statute must ultimately rest on congressional intent to provide a private remedy, Touche Ross & Co. v. Redington, 442 U. S. 560, 575 (1979). From this the corollary follows that the breadth of the right once recognized should not, as a general matter, grow beyond the scope congressionally intended.

This rule and corollary present respondents with a serious obstacle, for we can find no manifestation of intent to recognize a cause of action (or class of plaintiffs) as broad as respondents’ theory of causation would entail. At first blush, it might seem otherwise, for the Borak Court certainly did not ignore the matter of intent. Its opinion adverted to the statutory object of “protection of investors” as animating Congress’ intent to provide judicial relief where “necessary,” 377 U. S., at 432, and it quoted evidence for that intent from House and Senate Committee Reports, id., at 431-432. *1103Borak,’s probe of the congressional mind, however, never focused squarely on private rights of action, as distinct from the substantive objects of the legislation, and one Member of the Borak Court later characterized the “implication” of the private right of action as resting modestly on the Act’s “ ‘exclusively procedural provision’ affording access to a federal forum.” Bivens v. Six Unknown Fed. Narcotics Agents, 403 U. S. 388, 403, n. 4 (1971) (Harlan, J., concurring in judgment) (internal quotation marks omitted). See generally L. Loss, Fundamentals of Securities Regulation 929 (2d ed. 1988). See also Touche Ross, supra, at 568, 578. In fact, the importance of enquiring specifically into intent to authorize a private cause of action became clear only later, see Cort v. Ash, 422 U. S., at 78, and only later still, in Touche Ross, was this intent accorded primacy among the considerations that might be thought to bear on any decision to recognize a private remedy. There, in dealing with a claimed private right under § 17(a) of the Act, we explained that the “central inquiry remains whether Congress intended to create, either expressly or by implication, a private cause of action.” 442 U. S., at 575-576.

Looking to the Act’s text and legislative history mindful of this heightened concern reveals little that would help toward understanding the intended scope of any private right. According to the House Report, Congress meant to promote the “free exercise” of stockholders’ voting rights, H. R. Rep. No. 1383, 73d Cong., 2d Sess., 14 (1934), and protect “[f]air corporate suffrage,” id., at 13, from abuses exemplified by proxy solicitations that concealed what the Senate Report called the “real nature” of the issues to be settled by the subsequent votes, S. Rep. No. 792, 73d Cong., 2d Sess., 12 (1934). While it is true that these Reports, like the language of the Act itself, carry the clear message that Congress meant to protect investors from misinformation that rendered them unwitting agents of self-inflicted damage, it is just as true that Congress was reticent with indications of *1104how far this protection might depend on self-help by private action. The response to this reticence may be, of course, to claim that § 14(a) cannot be enforced effectively for the sake of its intended beneficiaries without their participation as private litigants. Borak, supra, at 432. But the force of this argument for inferred congressional intent depends on the degree of need perceived by Congress, and we would have trouble inferring any congressional urgency to depend on implied private actions to deter violations of § 14(a), when Congress expressly provided private rights of action in §§ 9(e), 16(b), and 18(a) of the same Act. See 15 U. S. C. §§78i(e), 78p(b), and 78r(a).11

The congressional silence that is thus a serious obstacle to the expansion of cognizable Borak causation is not, however, a necessarily insurmountable barrier. This is not the first effort in recent years to expand the scope of an action originally inferred from the Act without “conclusive guidance” from Congress, see Blue Chip Stamps v. Manor Drug Stores, 421 U. S., at 737, and we may look to that earlier case for the proper response to such a plea for expansion. There, we accepted the proposition that where a legal structure of private statutory rights has developed without clear indications of congressional intent, the contours of that structure need not be frozen absolutely when the result would be demonstrably inequitable to a class of would-be plaintiffs with claims comparable to those previously recognized. Faced in that case with such a claim for equality in rounding out the scope of an implied private statutory right of action, we looked to policy reasons for deciding where the outer limits of *1105the right should lie. We may do no less here, in the face of respondents’ pleas for a private remedy to place them on the same footing as shareholders with votes necessary for initial corporate action.

A

Blue Chip Stamps set an example worth recalling as a preface to specific policy analysis of the consequences of recognizing respondents’ first theory, that a desire to avoid minority shareholders’ ill will should suffice to justify recognizing the requisite causality of a proxy statement needed to garner that minority support. It will be recalled that in Blue Chip Stamps we raised concerns about the practical consequences of allowing recovery, under § 10(b) of the Act and Rule 10b-5, on evidence of what a merely hypothetical buyer or seller might have done on a set of facts that never occurred, and foresaw that any such expanded liability would turn on “hazy” issues inviting self-serving testimony, strike suits, and protracted discovery, with little chance of reasonable resolution by pretrial process. Id., at 742-743. These were good reasons to deny recognition to such claims in the absence of any apparent contrary congressional intent.

The same threats of speculative claims and procedural intractability are inherent in respondents’ theory of causation linked through the directors’ desire for a cosmetic vote. Causation would turn on inferences about what the corporate directors would have thought and done without the minority shareholder approval unneeded to authorize action. A subsequently dissatisfied minority shareholder would have virtual license to allege that managerial timidity would have doomed corporate action but for the ostensible approval induced by a misleading statement, and opposing claims of hypothetical diffidence and hypothetical boldness on the part of directors would probably provide enough depositions in the usual case to preclude any judicial resolution short of the credibility judgments that can only come after trial. Reliable evidence would seldom exist. Directors would under*1106stand the prudence of making a few statements about plans to proceed even without minority endorsement, and discovery would be a quest for recollections of oral conversations at odds with the official pronouncements, in hopes of finding support for ex post facto guesses about how much heat the directors would have stood in the absence of minority approval. The issues would be hazy, their litigation protracted, and their resolution unreliable. Given a choice, we would reject any theory of causation that raised such prospects, and we reject this one.12

B

The theory of causal necessity derived from the requirements of Virginia law dealing with postmerger ratification seeks to identify the essential character of the proxy solicitation from its function in obtaining the minority approval that would preclude a minority suit attacking the merger. Since the link is said to be a step in the process of barring a class of shareholders from resort to a state remedy otherwise available, this theory of causation rests upon the proposition of policy that § 14(a) should provide a federal remedy whenever a false or misleading proxy statement results in the loss under state law of a shareholder plaintiff’s state remedy for *1107the enforcement of a state right. Respondents agree with the suggestions of counsel for the SEC and FDIC that causation be recognized, for example, when a minority shareholder has been induced by a misleading proxy statement to forfeit a state-law right to an appraisal remedy by voting to approve a transaction, cf. Swanson v. American Consumers Industries, Inc., 475 F. 2d 516, 520-521 (CA7 1973), or when such a shareholder has been deterred from obtaining an order enjoining a damaging transaction by a proxy solicitation that misrepresents the facts on which an injunction could properly have been issued. Cf. Healey v. Catalyst Recovery of Pennsylvania, Inc., 616 F. 2d 641, 647-648 (CA3 1980); Alabama Farm Bureau Mutual Casualty Co. v. American Fidelity Life Ins. Co., 606 F. 2d 602, 614 (CA5 1979), cert. denied, 449 U. S. 820 (1980). Respondents claim that in this case a predicate for recognizing just such a causal link exists in Va. Code Ann. § 13.1-691(A)(2) (1989), which sets the conditions under which the merger may be insulated from suit by a minority shareholder seeking to void it on account of Beddow’s conflict.

This case does not, however, require us to decide whether § 14(a) provides a cause of action for lost state remedies, since there is no indication in the law or facts before us that the proxy solicitation resulted in any such loss. The contrary appears to be the case. Assuming the soundness of respondents’ characterization of the proxy statement as materially misleading, the very terms of the Virginia statute indicate that a favorable minority vote induced by the solicitation would not suffice to render the merger invulnerable to later attack on the ground of the conflict. The statute bars a shareholder from seeking to avoid a transaction tainted by a director’s conflict if, inter alia, the minority shareholders ratified the transaction following disclosure of the material facts of the transaction and the conflict. Va. Code Ann. *1108§ 13.1-691(A)(2) (1989). Assuming that the material facts about the merger and Beddow’s interests were not accurately disclosed, the minority votes were inadequate to ratify the merger under state law, and there was no loss of state remedy to connect the proxy solicitation with harm to minority shareholders irredressable. under state law.13 Nor is there a claim here that the statement misled respondents into entertaining a false belief that they had no chance to upset the merger until the time for bringing suit had run out.14

> HH

The judgment of the Court of Appeals is reversed.

It is so ordered.

concurring in part and concurring in the judgment.

I

As I understand the Court’s opinion, the statement “In the opinion of the Directors, this is a high value for the shares” *1109would produce liability if in fact it was not a high value and the directors knew that. It would not produce liability if in fact it was not a high value but the directors honestly believed otherwise. The statement “The directors voted to accept the proposal because they believe it offers a high value” would not produce liability if in fact the directors’ genuine motive was quite different — except that it would produce liability if the proposal in fact did not offer a high value and the directors knew that.

I agree with all of this. However, not every sentence that has the word “opinion” in it, or that refers to motivation for directors’ actions, leads us into this psychic thicket. Sometimes such a sentence actually represents facts as facts rather than opinions — and in that event no more need be done than apply the normal rules for § 14(a) liability. I think that is the situation here. In my view, the statement at issue in this case is most fairly read as affirming separately both the fact of the directors’ opinion and the accuracy of the facts upon which the opinion was assertedly based. It reads as follows:

“The Plan of Merger has been approved by the Board of Directors because it provides an opportunity for the Bank’s public shareholders to achieve a high value for their shares.” App. to Pet. for Cert. 53a.

Had it read “because in their estimation it provides an opportunity, etc.,” it would have set forth nothing but an opinion. As written, however, it asserts both that the board of directors acted for a particular reason and that that reason is correct. This interpretation is made clear by what immediately follows: “The price to be paid is about 30% higher than the [last traded price immediately before announcement of the proposal] .... [T]he $42 per share that will be paid to public holders of the common stock represents a premium of approximately 26% over the book value .... [T]he bank earned $24,767,000 in the year ended December 31, 1986 . . . .” Id., at 53a-54a. These are all facts that sup*1110port — and that are obviously introduced for the purpose of supporting — the factual truth of the “because” clause, i. e., that the proposal gives shareholders a “high value.”

If the present case were to proceed, therefore, I think the normal § 14(a) principles governing misrepresentation of fact would apply.

II

I recognize that the Court’s disallowance (in Part II-B-2) of an action for misrepresentation of belief is entirely contrary to the modern law of torts, as authorities cited by the Court make plain. See Vulcan Metals Co. v. Simmons Mfg. Co., 248 F. 858, 856 (CA2 1918); W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Law of Torts § 109 (5th ed. 1984), cited ante, at 1094. I have no problem with departing from modern tort law in this regard, because I think the federal cause of action at issue here was never enacted by Congress, see Thompson v. Thompson, 484 U. S. 174, 190-192 (1988) (Scalia, J., concurring in judgment), and hence the more narrow we make it (within the bounds of rationality) the more faithful we are to our task.

* * *

I concur in the judgment of the Court, and join all of its opinion except Part II.

with whom Justice Marshall joins, concurring in part and dissenting in part.

While I agree in substance with Parts I and II of the Court’s opinion, I do not agree with the,reasoning in Part III.

In Mills v. Electric Auto-Lite Co., 396 U. S. 375 (1970), the Court held that a finding that the terms of a merger were fair could not constitute a defense by the corporation to a shareholder action alleging that the merger had been accomplished by using a misleading proxy statement. The fairness of the transaction was, according to Mills, a matter to be considered at the remedy stage of the litigation.

*1111On the question of the causal connection between the proxy solicitation and the harm to the plaintiff shareholders, the Court had this to say:

“There is no need to supplement this requirement, as did the Court of Appeals, with a requirement of proof of whether the defect actually had a decisive effect on the voting. Where there has been a finding of materiality, a shareholder has made a sufficient showing of causal relationship between the violation and the injury for which he seeks redress if, as here, he proves that the proxy solicitation itself, rather than the particular defect in the solicitation materials, was an essential link in the accomplishment of the transaction. This objective test will avoid the impracticalities of determining how many votes were affected, and, by resolving doubts in favor of those the statute is designed to protect, will effectuate the congressional policy of ensuring that the shareholders are able to make an informed choice when they are consulted on corporate transactions. Cf. Union Pac. R. Co. v. Chicago & N. W. R. Co., 226 F. Supp. 400, 411 (D. C. N. D. Ill. 1964); 2 L. Loss, Securities Regulation 962 n. 411 (2d ed. 1961); 5 id., at 2929-2930 (Supp. 1969).” Id., at 384-385.

Justice Harlan writing for the Court then appended this footnote:

“We need not decide in this case whether causation could be shown where the management controls a sufficient number of shares to approve the transaction without any votes from the minority. Even in that situation, if the management finds it necessary for legal or practical reasons to solicit proxies from minority shareholders, at least one court has held that the proxy solicitation might be sufficiently related to the merger to satisfy the causation requirement, see *1112Laurenzano v. Einbender, 264 F. Supp. 356 (D. C. E. D. N. Y. 1966) . . . .” Id., at 385, n. 7.

The case before us today involves a merger that has been found by a jury to be unfair, not fair. The interest in providing a remedy to the injured minority shareholders therefore is stronger, not weaker, than in Mills. The interest in avoiding speculative controversy about the actual importance of the proxy solicitation is the same as in Mills. Moreover, as in Mills, these matters can be taken into account at the remedy stage in appropriate cases. Accordingly, I do not believe that it constitutes an unwarranted extension of the rationale of Mills to conclude that because management found it necessary — whether for “legal or practical reasons” — to solicit proxies from minority shareholders to obtain their approval of the merger, that solicitation “was an essential link in the accomplishment of the transaction.” Id., at 385, and n. 7. In my opinion, shareholders may bring an action for damages under § 14(a) of the Securities Exchange Act of 1934, 48 Stat. 895, 15 U. S. C. § 78n(a), whenever materially false or misleading statements are made in proxy statements. That the solicitation of proxies is not required by law or by the bylaws of a corporation does not authorize corporate officers, once they have decided for whatever reason to solicit proxies, to avoid the constraints of the statute. I would therefore affirm the judgment of the Court of Appeals.

with whom Justice Marshall, Justice Blackmun, and Justice Stevens join, concurring in part and dissenting in part.

I am in general agreement with Parts I and II of the majority opinion, but do not agree with the views expressed in Part III regarding the proof of causation required to establish a violation of § 14(a). With respect, I dissent from Part III of the Court’s opinion.

*1113HH

Review of the jury’s finding on causation is complicated because the distinction between reliance and causation was not addressed in explicit terms in the earlier stages of this litigation. Petitioners, in effect, though, recognized the distinction when they accepted the District Court’s essential link instruction as to reliance but not as to causation. So I agree with the Court that the issue has been preserved for our review here.*

*1114The Court of Appeals considered the essential link presumption in rejecting petitioners’ argument that Sandberg must show reliance by demonstrating that she read the proxy and then voted in favor of the proposal or took some other specific action in reliance upon it. In the Court of Appeals, the parties did not brief, nor did the panel address, the possibility that nonvoting causation theories would suffice to allow for recovery.

Before this Court petitioners do not argue that Sandberg must demonstrate reliance on her part or on the part of other shareholders. The matter of causation, however, must be addressed.

II

A

The severe limits the Court places upon possible proof of nonvoting causation in a § 14(a) private action are justified neither by our precedents nor by any case in the courts of appeals. These limits are said to flow from a shift in our approach to implied causes of action that has occurred since we recognized the § 14(a) implied private action in J. I. Case Co. v. Borak, 377 U. S. 426 (1964). Ante, at 1102-1105.

I acknowledge that we should exercise caution in creating implied private rights of action and that we must respect the primacy of congressional intent in that inquiry. See ante, at 1102. Where an implied cause of action is well accepted by our own cases and has become an established part of the securities laws, however, we should enforce it as a meaningful remedy unless we are to eliminate it altogether. As the *1115Court phrases it, we must consider the causation question in light of the underlying “policy reasons for deciding where the outer limits of the right should lie.” Ante, at 1104-1105; see Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723, 737 (1975).

According to the Court, acceptance of nonvoting causation theories would “extend the scope of Borak actions beyond the ambit of Mills.” Ante, at 1102. But Mills v. Electric Auto-Lite Co., 396 U. S. 375 (1970), did not purport to limit the scope of Borak actions, and as footnote 7 of Mills indicates, some courts have applied nonvoting causation theories to Borak actions for at least the past 25 years. See also L. Loss, Fundamentals of Securities Regulation 948, n. 81 (2d ed. 1988).

To the extent the Court’s analysis considers the purposes underlying § 14(a), it does so with the avowed aim to limit the cause of action and with undue emphasis upon fears of “speculative claims and procedural intractability.” Ante, at 1105. The result is a sort of guerrilla warfare to restrict a well-established implied right of action. If the analysis adopted by the Court today is any guide, Congress and those charged with enforcement of the securities laws stand forewarned that unresolved questions concerning the scope of those causes of action are likely to be answered by the Court in favor of defendants.

B

The Court seems to assume, based upon the footnote in Mills reserving the question, that Sandberg bears a special burden to demonstrate causation because the public shareholders held only 15 percent of the stock of First American Bank of Virginia (Bank). Justice Stevens is right to reject this theory. Here, First American Bankshares, Inc. (FABI), and Virginia Bankshares, Inc. (VBI), retained the option to back out of the transaction if dissatisfied with the reaction of the minority shareholders, or if concerned that the merger would result in liability for violation of duties to the minority shareholders. The merger agreement was con*1116ditioned upon approval by two-thirds of the shareholders, App. 463, and VBI could have voted its shares against the merger if it so decided. To this extent, the Court’s distinction between cases where the “minority” shareholders could have voted down the transaction and those where causation must be proved by nonvoting theories is suspect. Minority shareholders are identified only by a post hoc inquiry. The real question ought to be whether an injury was shown by the effect the nondisclosure had on the entire merger process, including the period before votes are cast.

The Court’s distinction presumes that a majority shareholder will vote in favor of management’s proposal even if proxy disclosure suggests that the transaction is unfair to minority shareholders or that the board of directors or majority shareholder is in breach of fiduciary duties to the minority. If the majority shareholder votes against the transaction in order to comply with its state-law duties, or out of fear of liability, or upon concluding that the transaction will injure the reputation of the business, this ought not to be characterized as nonvoting causation. Of course, when the majority shareholder dominates the voting process, as was the case here, it may prefer to avoid the embarrassment of voting against its own proposal and so may cancel the meeting of shareholders at which the vote was to have been taken. For practical purposes, the result is the same: Because of full disclosure the transaction does not go forward and the resulting injury to minority shareholders is avoided. The Court’s distinction between voting and nonvoting causation does not create clear legal categories.

III

Our decision in Mills v. Electric Auto-Lite Co., supra, at 385, rested upon the impracticality of attempting to determine the extent of reliance by thousands of shareholders on alleged misrepresentations or omissions. A misstatement or an omission in a proxy statement does not violate § 14(a) un*1117less “there is a substantial likelihood that a reasonable shareholder would consider it important in deciding how to vote.” TSC Industries, Inc. v. Northway, Inc., 426 U. S. 438, 449 (1976). If minority shareholders hold sufficient votes to defeat a management proposal and if the misstatement or omission is likely to be considered important in deciding how to vote, then there exists a likely causal link between the proxy violation and the enactment of the proposal; and one can justify recovery by minority shareholders for damages resulting from enactment of management’s proposal.

If, for sake of argument, we accept a distinction between voting and nonvoting causation, we must determine whether the Mills essential link theory applies where a majority shareholder holds sufficient votes to force adoption of a proposal. The merit of the essential link formulation is that it rests upon the likelihood of causation and eliminates the difficulty of proof. Even where a minority lacks votes to defeat a proposal, both these factors weigh in favor of finding causation so long as the solicitation of proxies is an essential link in the transaction.

A

The Court argues that a nonvoting causation theory would “turn on ‘hazy’ issues inviting self-serving testimony, strike suits, and protracted discovery, with little chance of reasonable resolution by pretrial process.” Ante, at 1105 (citing Blue Chip Stamps, 421 U. S., at 742-743). The Court’s description does not fit this case and is not a sound objection in any event. Any causation inquiry under § 14(a) requires a court to consider a hypothetical universe in which adequate disclosure is made. Indeed, the analysis is inevitable in almost any suit when we are invited to compare what was with what ought to have been. The causation inquiry is not intractable. On balance, I am convinced that the likelihood that causation exists supports elimination of any requirement that the plaintiff prove the material misstatement or omission caused the transaction to go forward when it otherwise would *1118have been halted or voted down. This is the usual rule under Mills, and the difficulties of proving or disproving causation are, if anything, greater where the minority lacks sufficient votes to defeat the proposal. A presumption will assist courts in managing a circumstance in which direct proof is rendered difficult. See Basic Inc. v. Levinson, 485 U. S. 224, 245 (1988) (discussing presumptions in securities law).

B

There is no authority whatsoever for limiting § 14(a) to protecting those minority shareholders whose numerical strength could permit them to vote down a proposal. One of §14(a)’s “chief purposes is The protection of investors.”’ J. I. Case Co. v. Borak, 377 U. S., at 432. Those who lack the strength to vote down a proposal have all the more need of disclosure. The voting process involves not only casting ballots but also the formulation and withdrawal of proposals, the minority’s right to block a vote through court action or the threat of adverse consequences, or the negotiation of an increase in price. The proxy rules support this deliberative process. These practicalities can result in causation sufficient to support recovery.

The facts in the case before us prove this point. Sandberg argues that had all the material facts been disclosed, FABI or the Bank likely would have withdrawn or revised the merger proposal. The evidence in the record, and more that might be available upon remand, see infra, at 1120, meets any reasonable requirement of specific and nonspeculative proof.

FABI wanted a “friendly transaction” with a price viewed as “so high that any reasonable shareholder will accept it.” App. 99. Management expressed concern that the transaction result in “no loss of support for the bank out in the community, which was important.” Id., at 109. Although FABI had the votes to push through any proposal, it wanted a favorable response from the minority shareholders. Id., at 192. Because of the “human element involved in a transac*1119tion of this nature,” FABI attempted to “show those minority shareholders that [it was] being fair.” Id., at 347.

The theory that FABI would not have pursued the transaction if full disclosure had been provided and the shareholders had realized the inadequacy of the price is supported not only by the trial testimony but also by notes of the meeting of the Bank’s board, which approved the merger. The inquiry into causation can proceed not by “opposing claims of hypothetical diffidence and hypothetical boldness,” ante, at 1105, but through an examination of evidence of the same type the Court finds acceptable in its determination that directors’ statements of reasons can lead to liability. Discussion at the board meeting focused upon matters such as “how to keep PR afloat” and “how to prevent adverse reac[tion]/perception,” App. 454, demonstrating the directors’ concern that an unpopular merger proposal could injure the Bank.

Only a year or so before the Virginia merger, FABI had failed in an almost identical transaction, an attempt to freeze out the minority shareholders of its Maryland subsidiary. FABI retained Keefe, Bruyette & Woods (KBW) for that transaction as well, and KBW had given an opinion that FABI’s price was fair. The subsidiary’s board of directors then retained its own adviser and concluded that the price offered by FABI was inadequate. Id., at 297, 319. The Maryland transaction failed when the directors of the Maryland bank refused to proceed; and this was despite the minority’s inability to outvote FABI if it had pressed on with the deal.

In the Virginia transaction, FABI agaip decided to retain KBW. Beddow, who sat on the boards of both FABI and the Bank, discouraged the Bank from hiring its own financial adviser, out of fear that the Maryland experience would be repeated if the Bank received independent advice. Directors of the Bank testified they would not have voted to approve the transaction if the price had been demonstrated unfair to the minority. Further, approval by the Bank’s *1120board of directors was facilitated by FABI’s representation that the transaction also would be approved by the minority shareholders.

These facts alone suffice to support a finding of causation, but here Sandberg might have had yet more evidence to link the nondisclosure with completion of the merger. FAB I executive Robert Altman and Bank Chairman Drewer met on the day before the shareholders meeting when the vote was taken. Notes produced by petitioners suggested that Drewer, who had received some shareholder objections to the $42 price, considered postponing the meeting and obtaining independent advice on valuation. Altman persuaded him to go forward without any of these cautionary measures. This information, which was produced in the course of discovery, was kept from the jury on grounds of privilege. Sand-berg attacked the privilege ruling on five grounds in the Court of Appeals. In light of its ruling in favor of Sandberg, however, the panel had no occasion to consider the admissibility of this evidence.

Though I would not require a shareholder to present such evidence of causation, this case itself demonstrates that nonvoting causation theories are quite plausible where the misstatement or omission is material and the damages sustained by minority shareholders is serious. As Professor Loss summarized the holdings of a “substantial number of cases,” even if the minority cannot alone vote down a transaction,

“minority stockholders will be in a better position to protect their interests with full disclosure and ... an unfavorable minority vote might influence the majority to modify or reconsider the transaction in question. In [Schlick v. Penn-Dixie Cement Corp., 507 F. 2d 374, 384 (CA2 1974),] where the stockholders had no appraisal rights under state law because the stock was listed on the New York Stock Exchange, the court advanced two additional considerations: (1) the market would be informed; and (2) even ‘a rapacious controlling manage*1121ment’ might modify the terms of a merger because it would not want to ‘hang its dirty linen out on the line and thereby expose itself to suit or Securities Commission or other action — in terms of reputation and future takeovers.’” Fundamentals of Securities Regulation, at 948 (footnote omitted).

I conclude that causation is more than plausible; it is likely, even where the public shareholders cannot vote down management’s proposal. Causation is established where the proxy statement is an essential link in completing the transaction, even if the minority lacks sufficient votes to defeat a proposal of management.

IV

The majority avoids the question whether a plaintiff may prove causation by demonstrating that the misrepresentation or omission deprived her of a state-law remedy. I do not think the question difficult, as the whole point of federal proxy rules is to support state-law principles of corporate governance. Nor do I think that the Court can avoid this issue if it orders judgment for petitioners. The majority asserts that respondents show no loss of a state-law remedy, because if “the material facts about the merger and Beddow’s interests were not accurately disclosed, the minority votes were inadequate to ratify the merger under state law.” Ante, at 1108. This theory requires us to conclude that the Virginia statute governing director conflicts of interest, Va. Code Ann. § 13.1 — 691(A)(2) (1989), incorporates the same definition of materiality as the federal proxy rules. I find no support for that proposition. If the definitions are not the same, then Sandberg may have lost her state-law remedy. For all we know, disclosure to the minority shareholders that the price is $42 per share may satisfy Virginia’s requirement. If that is the case, then approval by the minority without full disclosure may have deprived Sandberg of the ability to void the merger.

*1122In all events, the theory that the merger would have been voidable absent minority shareholder approval is far more speculative than the theory that FABI and the Bank would have called off the transaction. Even so, this possibility would support a remand, as the lower courts have yet to consider the question. We are not well positioned as an institution to provide a definitive resolution to state-law questions of this kind. Here again, the difficulty of knowing what would have happened in the hypothetical universe of full disclosure suggests that we should “resolv[e] doubts in favor of those the statute is designed to protect” in order to “effectuate the congressional policy of ensuring that the shareholders are able to make an informed choice when they are consulted on corporate transactions.” Mills, 396 U. S., at 385.

I would affirm the judgment of the Court of Appeals.