The stock of Microdot, Inc., was traded on the New York Stock Exchange when General Cable Corp. announced, on December 2, 1975, an intention to make a tender offer to holders of Microdot's shares. Microdot's stock had been trading for $11 3/4 per share; General Cable planned to offer $17; the price available on the Exchange almost immediately topped $17, reaching $18 3/8 on December 8, 1975. The price retreated toward $17 through December, however, as Microdot opposed the offer. On December 5 Microdot issued a press release (published as an advertisement in the Wall Street Journal on December 8) stating that the offer of $17 was unacceptable to Microdot's board of directors. It stated: "After careful consideration of the proposed offer and consulting with Goldman, Sachs & Co., our investment bankers, we have concluded that the General Cable offer is totally inadequate. We will employ all resources at our disposal to defeat the offer."

Other communications from Microdot advised the shareholders not to sell their stock, either on the market or to General Cable. Rudolph Eberstadt, Jr., the president and CEO of Microdot, bewailed the fate of medium-sized, high-tech companies in a world of takeovers, complaining that if hostile takeovers continued American industry would be less inventive. One of Microdot's ads, published in the Wall Street Journal on December 15, stated:

If no successful growth company of medium size is able to resist the raids of large outsiders, there will be no way for American stockholders to realize the potential these growth companies represent. Then there will be no future IBM, no Xerox, no Polaroid. They will be choked off and smothered as soon as they start to show their real growth potential.

Perhaps appreciating, however, that the "growth potential" in which the investors were interested was growth from $11 3/4 to a higher price, Microdot also decided to see whether more than $17 could be obtained. The search for a "White Knight" was on. As it turned out, other than trying to turn up a better bid and publishing peevish advertisements, Microdot did nothing to hinder General Cable's bid. It did not have many "resources at [its] disposal to defeat the offer."

Microdot authorized Goldman, Sachs to approach other firms that might be interested. Goldman, Sachs tried its hand with more than 100 firms; not a one showed interest through December 1975. Five were persuaded to meet with Microdot's management between December 12 and 20; all five told Microdot on the spot that they were not interested in further discussions. Northwest Industries, with which Goldman, Sachs spoke on December 12, was one of the early nay-sayers, declining to meet with Microdot's managers. Between December 20 and December 30, Goldman, Sachs could not even induce another firm to attend an exploratory meeting with Microdot's management. It is therefore no surprise that the price of the stock fell toward $17. Goldman's activity and its lack of success could not have been a secret on Wall Street; no one contacts 100 firms and escapes notice of arbitrageurs who invest millions in the stock of targets.

Goldman, Sachs kept trying, however, and after receiving more information Northwest let Microdot know that it had a lingering interest in discussing matters. Managers of the firms met on January 19. Ben W. Heineman, the president of Northwest, told Eberstadt on January 19 that Northwest was prepared to offer $21 per share in a tender offer but would not pursue the matter unless Microdot made an advance commitment to take Northwest's offer. Microdot's board made that commitment on January 24. On January 26 Northwest's board approved the offer, and the two firms made a public announcement on the same day. General Cable dropped out of the picture; Northwest acquired the stock for $21. Like many another marriage, this ended in divorce. Northwest later spun off Microdot, which is once more an independent firm traded on the New York Stock Exchange.

* We have a suit filed by Ann and Arnold Flamm as representatives of the class of investors who sold Microdot stock between December 5, 1975, and January 23, 1976. They maintain that Microdot's search for a White Knight was material information that had been withheld, making Microdot liable under Section 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. Sec. 78j(b), and the SEC's Rule 10b-5, 17 C.F.R. Sec. 240.10b-5. The district court certified the class, and the case was tried to a jury. The facts we have narrated are undisputed. The parties locked horns, however, about whether the search for the White Knight was material and whether Arnold Flamm sold his stock in reliance on a belief that Microdot was determined to remain independent. The parties and the district judge treated other members of the class as clones of Arnold Flamm, so that if he loses the whole class loses. We shall do likewise without suggesting that this is the proper approach.

Flamm, a veteran investor, sold his stock on December 29, 1975. He acknowledged at trial that he knew that White Knights frequently ride to the rescue of managers determined to resist the initial bid but unable to remain independent. He acknowledged as well that he appreciated the significance of the price of $18 3/8--professional investors were staking their own wealth behind a belief that a higher bid would come along. And although Flamm stated that he concluded after the December 15 advertisement that Microdot was determined to stay independent, he acknowledged that it was "quite possible" that he had held the stock for another two weeks on the off chance that a White Knight would gallop into view.

The district court told the jury that unless Flamm showed that the failure to disclose the search was material, and unless Flamm "justifiably relied" to his detriment on Microdot's statements and omissions, the class could not recover. The instruction stated:

In the case of misrepresentations it must be proved that [Flamm] in fact relied upon the false statements. In other words, if you find that [Flamm] would have engaged in the transaction anyway, and that the misrepresentation or omission had no effect upon [his] decision, then there was no reliance and there can be no recovery. Further, [Flamm] must prove that [his] reliance was justified; that [he] did not intentionally close [his] eyes and refuse to investigate, concerning the circumstances in disregard of a risk known to [him], or so obvious that [he] must be taken to have been aware of it, and so great as to make it highly probable that harm would follow.

In the case of omissions or non-disclosure of material facts, if such an omission is proved, then the element of reliance ... may be presumed. The law infers that [Flamm] would have relied upon facts which are shown to be material and intentionally withheld. The Defendants, however, may rebut this presumption if they are able to prove, by a preponderance of the evidence, that even if the material facts had been disclosed, [Flamm's] decision as to the transaction would not have been different from what it was.

Flamm directs a barrage against this instruction. The two-tier burden was beyond the jury's ability to comprehend, he maintains, because misleading statements (half-truths) and material omissions (the rest of the truth) really are the same thing. Confusion was compounded by the phrase "misrepresentation or omission" in the first paragraph, which assigns the burden to Flamm, while the second paragraph assigns the burden on omissions to defendants. The first paragraph also states that reliance must be "justified", which means both that the investor may not "refuse to investigate" and may not act "in disregard of a risk known ... or so obvious that [the investor] must be taken to have been aware of it". Flamm contends that both parts of this definition of "justified reliance" are wrong.

The district judge used as his model the Fifth Circuit's pattern jury instructions. These pattern jury instructions are based roughly on Dupuy v. Dupuy, 551 F.2d 1005 (5th Cir.1977), a case involving a family corporation rather than trading in public markets. We disagreed with the holding of Dupuy in Teamsters Local 282 Pension Trust Fund v. Angelos, 762 F.2d 522 (7th Cir.1985), decided almost eleven months before the trial of this case. Angelos reviews the extent to which "justified reliance" is an element of an action under Rule 10b-5 and holds both that investors do not have a "duty to investigate" and that "reliance" means only materiality and causation in conjunction. 762 F.2d at 527-30. See also Robert Charles Clark, Corporate Law Sec. 8.10.5 (1986) ("reliance" became anachronistic in securities law when the Supreme Court made liability under Rule 10b-5 depend on scienter). Defendants concede that the instruction is erroneous under Angelos to the extent it mentions a "duty to investigate" but insist that the error is harmless because "investigation" did not play a role in the case. They defend the "known risk" portion of the instruction, but not persuasively.

Under Angelos action in the teeth of a known risk should be evaluated as part of the inquiry into materiality, not reliance. 762 F.2d at 530. If an investor knows a particular fact demonstrating a risk, the failure to disclose related or corroborating facts is not material because it does not significantly affect the mix of information on which the investor acts. Knowledge of a risk does not prevent recovery if the additional information would have been a significant increment to the total available knowledge.

The instruction in this case not only treats lack of knowledge of a risk as an inquiry in addition to that of materiality but also neglects to distinguish among kinds of risk. Every investor knows of risk in the abstract; there is a risk that any statement will be a lie. If this sort of knowledge were enough to defeat "justified reliance", no thoughtful investor could recover. Any reader of the business pages of the newspapers knows that White Knights appear in a significant fraction of tender offers. This is not the same as knowledge that there was apt to be a bid by a White Knight for Microdot. The critical question was whether Microdot would be among those acquired by White Knights. Material facts are those that help evaluate Microdot's chances, firm-specific information that adds to the existing knowledge that there is a probability distribution of outcomes. The instruction here allowed the jury to decide for Microdot just because Flamm knew that some firms were bailed out by White Knights--even if the failure to disclose the probability that Microdot would be among them was a material omission.

We also have some doubts about the split allocation of the risk of nonpersuasion. Perhaps this is the right one in some ethereal sense. See Affiliated Ute Citizens v. United States, 406 U.S. 128, 153-54, 92 S. Ct. 1456, 1472, 31 L. Ed. 2d 741 (1972); Michaels v. Michaels, 767 F.2d 1185, 1199-1200 (7th Cir.1985). But the instruction here was garbled, and even in the best of circumstances an effort to create different burdens on functionally identical things must set jurors' heads reeling. The definition of a "misleading" statement (on which Flamm had the burden) is a half-truth, a statement true as far as it goes but deceptive because of the omission of some other truth; yet the instruction assigned the burden on omissions to defendants. It may be that a distinguishing feature of the legal mind is the ability to entertain contradictory propositions simultaneously; this is not a distinguishing feature of jurors' minds. The notion of a burden of persuasion is hard enough by itself for someone who may encounter the legal system once in a lifetime. Even Learned Hand had trouble with shadings among burdens. United States v. Feinberg, 140 F.2d 592, 594 (2d Cir.1944). Several courts accordingly have held or suggested that instructions of this character not be given in jury trials. E.g., Sharp v. Coopers & Lybrand, 649 F.2d 175, 188-89 (3d Cir.1981); Austin v. Loftsgaarden, 675 F.2d 168, 178 n. 21 (8th Cir.1982), reversed in part on other grounds after remand under the name Randall v. Loftsgaarden, --- U.S. ----, 106 S. Ct. 3143, 92 L. Ed. 2d 525 (1986); Cavalier Carpets, Inc. v. Caylor, 746 F.2d 749, 753-57 (11th Cir.1984); Grossman v. Waste Management, Inc., 589 F. Supp. 395, 410-12 (N.D.Ill.1984). The instruction should be avoided whenever possible in favor of a single burden--one that is not hard to come by in light of Angelos, which essentially removes reliance as an element independent of causation and materiality in a case under Rule 10b-5. See also Competitive Associates, Inc. v. Laventhal, Krekstein, Horwath & Horwath, 516 F.2d 811 (2d Cir.1975).

Defendants assert, nevertheless, that nothing in the jury instructions materially injured Flamm. The concatenation of errors would make a harmless error inquiry very difficult. We bypass that inquiry, however, in favor of an alternative ground of affirmance. See Massachusetts Mutual Life Insurance Co. v. Ludwig, 426 U.S. 479, 96 S. Ct. 2158, 48 L. Ed. 2d 784 (1976). Defendants moved for summary judgment on the ground that the search for a White Knight is immaterial as a matter of law. The district court denied this motion, but we conclude that the court should have granted it. This case should not have gone to the jury, and a remand based on errors in the jury instructions would be pointless.1

An omission is material when there is a "substantial likelihood that, under all the circumstances, the omitted fact would have assumed actual significance in the deliberations of the reasonable shareholder"--that is, when it "would have been viewed by the reasonable investor as having significantly altered the 'total mix' of information made available." TSC Industries, Inc. v. Northway, Inc., 426 U.S. 438, 449, 96 S. Ct. 2126, 2132, 48 L. Ed. 2d 757 (1976) (footnote omitted). TSC dealt with "materiality" under the proxy rules, but like every other court of appeals we have taken the definition in TSC as suitable for the term wherever it appears in securities law. E.g., Michaels, 767 F.2d at 1194.

Several courts of appeals have held that efforts by public corporations to arrange mergers are immaterial under this standard, as a matter of law, until the firms have agreed on the "price and structure" of the deal. Staffin v. Greenberg, 672 F.2d 1196, 1204-07 (3d Cir.1982); Reiss v. Pan American World Airways, Inc., 711 F.2d 11, 14 (2d Cir.1983); Greenfield v. Heublein, Inc., 742 F.2d 751, 756-58 (3d Cir.1984); cf. Levinson v. Basic Inc., 786 F.2d 741, 746 (6th Cir.1986) (reserving the issue), cert. granted, --- U.S. ----, 107 S. Ct. 1284, 93 L. Ed. 2d 143 (1987). See also Missouri Portland Cement Co. v. H.K. Porter Co., 535 F.2d 388, 398 (8th Cir.1976), and Susquehanna Corp. v. Pan American Sulphur Co., 423 F.2d 1075, 1084-86 (5th Cir.1970), which apply a similar rule to the bidder's disclosure obligations. Microdot and Northwest did not agree on the price and structure of a deal until January 24, 1976, at the earliest, the day after the period in which the last member of the class sold. On December 29, 1975, when Flamm sold, Microdot was forlornly looking for a rescuer; the most that can be said is that Microdot had authorized an investment bank to send distress signals. Staffin and Reiss treat such inquiries as immaterial.

From one perspective this conclusion is simply another cause for wonderment at the legal mind. Investors were looking at potential prices from $11.75 (if Microdot had defeated all bids) to $17 (if General Cable's bid had succeeded) to $21 (under Northwest's bid), and maybe more if a better bid were available. This is almost a 100% range. Only an addlepated investor would consider a 100% difference in price unimportant in deciding what to do. The range in this case is not unusual. Tender offers entail substantial premia compared with the prices shares carry before the bids--and afterward, should the offers be defeated. Managers who sell the firm to White Knights obtain gains in addition to those available from the initial bidders. The data may be found in Paul Asquith, Merger Bids, Uncertainty, and Stockholder Returns, 11 J. Financial Econ. 51, 62-72, 81 (1983); Michael Bradley, Anand Desai & E. Han Kim, The Rationale Behind Interfirm Tender Offers: Information or Synergy?, 11 J. Financial Econ. 183, 188-98 (1983); Frank H. Easterbrook & Gregg A. Jarrell,Do Targets Gain From Defeating Tender Offers?, 59 N.Y.U.L.Rev. 277, 289-91, 294-99 (1984); Gregg A. Jarrell, The Wealth Effects of Litigation by Targets: Do Interests Diverge in a Merge?, 28 J.L. & Econ. 151, 161-73 (1985); Wayne H. Mikkelson & Richard S. Ruback, An Empirical Investigation of the Interfirm Equity Investment Process, 14 J. Financial Econ. 523, 539-40 (1985).

The judges who decided Staffin, Reiss, and similar cases did not deny that the difference between merger and no merger may be substantial. They drew instead on a different strand of the Supreme Court's reasoning in TSC, the recognition that "[s]ome information is of such dubious significance that insistence on its disclosure may accomplish more harm than good." 426 U.S. at 448, 96 S.Ct. at 2132. Our colleagues on other courts of appeals have suggested two reasons why the disclosure of ongoing negotiations may "accomplish more harm than good." One is that disclosure of ongoing negotiations may befuddle the investors, leading them to think the outcome more certain than it is. E.g., Staffin, 672 F.2d at 1206; Reiss, 711 F.2d at 14 ("Such negotiations are inherently fluid and the eventual outcome is shrouded in uncertainty. Disclosure may in fact be more misleading than secrecy so far as investment decisions are concerned."); Greenfield, 742 F.2d at 756. The other reason is that premature disclosure may frustrate the achievement of the firm's objective, destroying the source of the value sought to be disclosed. E.g., Staffin, 672 F.2d at 1206-07 ("disclosure of preliminary merger negotiations would, by and large, do more harm than good to shareholders", collecting sources of support; "If the announcement is withheld until an agreement in principle ... is reached, the greatest good for the greatest number results."); Greenfield, 742 F.2d at 757 ("considering the delicate nature of most merger discussions, [early disclosure] might seriously inhibit such acquisition ventures."). The cases also allude to a third rationale, the need to create a bright-line rule that will allow firms to plan corporate transactions with the assurance that they will not be condemned no matter which way they proceed on disclosure.

The first of these reasons, that disclosure may confuse investors rather than illuminate their choices, is weak. It assumes that investors are nitwits, unable to appreciate--even when told--that mergers are risky propositions up until the closing. Almost all corporate ventures, from building a new plant to angling for a merger partner, may go well or poorly, with a probability attached to each outcome. To attribute to investors a child-like simplicity, an inability to grasp the probabilistic significance of negotiations, implies that they should not be told about new plants, new products, new managers, or any of the other changes in the life of the corporation. These new events--things with potential for boom or bust--are exactly the news on which sophisticated investors make most decisions; "old" news, with settled value, already is reflected in the price of the stock and so is no news at all. Doubtless some unsophisticated investors think that negotiations for a merger are the same thing as a completed merger, but such babes in the woods are not apt to follow contested tender offers day by day. Disclosures to the market as a whole cannot be limited to what is fit for rubes. The Wall Street Journal is filled with rumors on which investors act; Rule 175, 17 C.F.R. Sec. 230.175, allows and even encourages the disclosure of "projections" about uncertain events; any Schedule 13D filed in a tender offer will disclose a range of plans and contingent choices. These documents regularly contain statements about potential mergers, accompanied by disclaimers and cautions about risks. We do not doubt that these schedules, drafted with more haste than the prospectus used to sell newly issued stock, are correspondingly more likely to be incomplete. But some information is almost always preferable to none. Investors, who appreciate the necessary omissions, can deal with risk. And the disclosure of a fact (such as the hiring of an investment banker to hunt up a White Knight) does not invite suit just because the search turns out poorly.

The effect of premature disclosure on the probability of merger is a much more substantial concern. Investors seek monetary returns, and few want disclosure for its own sake. To the extent investors' wealth depends on withholding information, all favor that course. So a firm that is working on a valuable invention may elect not to disclose the existence of the project, let alone daily progress. Much of the value of the invention may come from stealing a march on rivals, a value that would be dissipated if the firm turned valuable information over to its investors and therefore inescapably to rivals. An investor who sells the day before the public disclosure of the invention cannot complain that he missed out on an increase in the price of the firm's stock, any more than the investor who sells the day before a favorable quarterly report of earnings or an increase in dividends may demand to receive the increase. See State Teachers Retirement Board v. Fluor Corp., 654 F.2d 843, 850 (2d Cir.1981). The famous Texas Gulf Sulphur case furnishes another illustration. TGS found the world's biggest lode of nickel. To capitalize, it had to line up the mineral rights for the whole area. If TGS had released the assays immediately, other firms (or the owners of the surface interests) could have captured the rewards of TGS's search. TGS had to keep its find secret for a while, and investors who sold their stock while TGS was silent had no complaint. See SEC v. Texas Gulf Sulphur Co., 401 F.2d 833 (2d Cir.1968) (en banc). TGS did its investors a favor by saying as little as possible for as long as necessary. Once TGS started disclosing it had a duty not to lie, but that's another problem (which we discuss in Part III).

Negotiations for a merger present a similar problem. Some potential acquirers may demand that negotiations proceed in secrecy. They may fear that premature disclosure may spark competition that will deprive them of part of the value of their effort, so that bids in a world of early disclosure will be lower than bids in a world of deferred disclosure. One specter facing any potential buyer is the winner's curse--the prospect that the high bidder wins the auction only because he alone has placed an unrealistically high value on the assets. Negotiations in the glare of publicity may lead putative buyers to think that whenever they locate a bargain they will be preempted by some other bidder who waits in the wings, but whenever they offer too much they will be left with their "prize". The heads-I-win-tails-you-lose quality of this process will lead either to a general reduction in offers (to assure that no offer is too high) or to a refusal to offer the best price without being assured of victory. Either way, silence during negotiations may be beneficial for investors. See also Note, Rule 10b-5 and the Duty to Disclose Merger Negotiations in Corporate Statements, 96 Yale L.J. 547, 554-56 (1987). Both the New York and the American Stock Exchanges therefore suggest that listed firms postpone announcements until definitive agreements have been reached. Staffin quotes the manual of the American Stock Exchange. See also Sec. 202.01 of the New York Stock Exchange's manual, reproduced as an appendix to this opinion.

Flamm may reply that silence is not beneficial to all investors, that there is a conflict of interest between investors who sell before the disclosure and investors who hold, thereby receiving the benefits later on. See Ronald J. Gilson, The Law and Finance of Corporate Acquisitions 977 (1986). Perhaps so, although the corporation is not required by the securities law to favor hair-trigger sellers over other investors. From a longer perspective, however, even this conflict disappears. Investors who wanted to prescribe their managers' behavior during merger discussions would favor a rule of silence until the discussions had reached agreement on price and structure. Such discussions may occur anytime during the life of the firm. Ex ante, each investor's chance of selling during that window is small. The chance of selling for "too little" is offset by an identical chance of buying at a bargain; every sale has a buyer and a seller. Over the long run, then, the prospect of selling for too little and buying a bargain are a wash, leaving only the prospect of receiving (or scaring away) beneficial opportunities to merge. All investors would prefer whichever approach maximized their anticipated wealth.2 The legal rule governing disclosure is like this hypothetical bargain among investors. It applies to all firms, to all investors be they buyers or sellers, at all times. In selecting a legal rule, a court must consider the effects on all investors in all firms, not just the effects on the plaintiff.

Even the unlucky investors, such as Flamm, who sell their stock in a particular firm too soon can take comfort in knowing that they do not lose the whole gain. To the extent the appearance of White Knights is predictable, the probability of a White Knight appearing in this contest will be reflected in the price of the stock. Most buyers during tender offer contests are arbitrageurs, professional investors who are exceptionally knowledgable. These professionals make money by taking risk--they take the risk that all bids will vanish (and the price fall here to $11.75) in exchange for the prospect of gain from the offer (here at $17) and the chance of a higher bid. When Flamm sold his stock, he passed to the arbitrageurs the risk that the price would fall to $11.75, or even to $17 (he received $17 3/8); the arbitrageurs did not take the risk off Flamm's shoulders for free but were compensated by the possibility (remote as of December 29!) of a higher price. Undoubtedly many arbitrageurs had learned that Goldman, Sachs was shopping for a deal. Their bids reflected the value of a potential deal, and Flamm received this value without knowing about the prospects himself. It is not right to reply that the arbitrageurs--"speculators", transient investors--are swiping gains that "belong" to the longer-term investors such as Flamm. Arbitrageurs must compete among themselves to buy stock. The more likely the gain from a later White Knight bid, the more any given arbitrageur is willing to pay for stock. To make a profit the arbitrageur must put his hands on the stock; to acquire the stock he must outbid other arbitrageurs, who have the same end in view; the competition ultimately passes back to Flamm and the other original investors the gains from the probability of a White Knight bid, as of December 29 (or any other date), less the premium for taking risk off Flamm's hands. Premature disclosure could have reduced the chances of an acquisition by a White Knight, and therefore reduced the bids made in the market for Flamm's stock.

So silence pending settlement of the price and structure of a deal is beneficial to most investors, most of the time. We do not think that the securities laws war against the best interests of investors. Rule 10b-5 is about fraud, after all, and it is not fraudulent to conduct business in a way that makes investors better off--that all investors prefer ex ante and that most prefer even ex post. Cf. Dirks v. SEC, 463 U.S. 646, 653-59, 103 S. Ct. 3255, 3260-63, 77 L. Ed. 2d 911 (1983) (liability depends on a "duty" to disclose, a duty defined in part to ensure the welfare of investors as a group).

We agree, too, with the conclusion of the other circuits that the benefits of certainty supply additional support for the price-and-structure rule. If disclosure must occur at an earlier date, how much earlier? That would be fertile ground for disputation. No matter how soon the firm announced the negotiations, investors could say that it should have done so a little sooner. The pressure to advance the date of disclosure by "just a little" (at a time) would erode the benefits of deferral. No rule other than a clear one would have staying power. The time at which information should be disclosed ought to be readily ascertainable. The price-and-structure rule will leave some questions in doubt, but it is better than any alternative.

Although we adopt the holding of Staffin and Reiss that public corporations need not disclose ongoing negotiations until they have produced agreement on the price and structure of the deal, that does not produce immediate victory for the defendants. There is, at least potentially, a further question whether firms may deny the existence of ongoing negotiations or shade the truth.

This question has produced a conflict among the circuits. Greenfield holds (over a dissent by the author of Staffin ) that a firm may deny knowledge of new corporate developments even while in the final stages of negotiation. Levinson explicitly rejects this part of Greenfield 's holding, 786 F.2d at 748, while reserving judgment on the price-and-structure rule as applied to silent corporations. Flamm relies on Levinson. This is a difficult question--one on which our court in Michaels initially took the side of Greenfield but then amended the opinion to become neutral. See 767 F.2d at 1195-96.

The conclusion of Levinson is supported by the principle that he who speaks must tell the truth about important matters. The firm may be silent, leaving investors to take their chances, but may not lie; the lie may have greater adverse effect on the value of the stock than the premature truth. The conclusion of Greenfield is supported by the fact that unless the firm is entitled to conceal the negotiations, a demand by the Exchange to confirm or deny a rumor may flush out the truth no matter what the firm says, even though the firm is entitled to be silent and most investors would want it to be. Suppose a firm is engaged in negotiations that are best kept quiet, and the Exchange asks whether new developments account for activity in its stock. If the firm says yes and says why, the cat is out of the bag; if the firm says no, it faces liability for fraud; if the firm says "no comment" that is the same thing as saying "yes" because investors will deduce the truth. No corporation follows the CIA's policy of saying "no comment" to every inquiry; every firm regularly confirms or denies rumors, as the securities laws and the stock exchanges' rules require. The exchanges' rules require a response, not a refusal to respond, to inquiries. When a firm suddenly says "no comment", the inquisitor will realize that his suspicions have a foundation--yet the response may sow confusion all the same. If by hypothesis silence is the best course for investors, then it may be necessary to condone evasive answers, as the Third Circuit did in Greenfield, to put pursuers off the scent for a time. See also Gilson at 978-79; Note, 96 Yale L.J. at 558-64.

As Flamm describes the case, we must choose between Greenfield and Levinson. Microdot opposed General Cable's offer, issuing one press release after another. It told investors to hold their stock because $17 was inadequate and pledged to resist General Cable's bid by all available means. It cried wolf about the effect of tender offers on aggressive innovators. Flamm insists that he deduced from these statements that Microdot wanted to remain independent at any cost, even the return of the stock's price to $11.75. If the statements amount to a declaration of independence, then failure to disclose that Goldman, Sachs was beating the bushes for higher bids was an omission that made the statements materially misleading, just like the "no corporate developments" statements in Greenfield and Levinson.

Flamm offers an implausible reading of Microdot's statements, however. The test of materiality is objective, TSC, 426 U.S. at 445, 96 S.Ct. at 2130: Michaels, 767 F.2d at 1196, and no objectively reasonable reader of these comments would take them as denials of any interest in other bids, no matter how beneficial those bids were to Microdot's investors. The firm's management claimed to be representing the interests of investors, rather than to be pursuing a social philosophy that small is good no matter the cost to stockholders. None of the statements denied that the firm was searching for higher bids; the implication of the statement that General Cable's bid is "inadequate" is the contrary. Levinson, in contrast, dealt with five bald denials that the firm was attempting to arrange a merger. None of the materials Microdot issued states any view on the desirability of White Knights, for the economy in general or Microdot in particular. The only concrete advice Microdot gave its investors was not to sell, because $17 was too low. As things turned out, that was wise counsel. Flamm is hardly in a position to contend that because he rejected this advice he has been deceived.

Ours is therefore a case of silence on the subject of the omitted information. For all practical purposes, Microdot was mum on its strategy for defeating General Cable's bid. Defensive strategies include seeking White Knights as well as scorched earth and other destructive maneuvers. Microdot never announced a strategy involving moats, fences, or any tactic that would rebuff all bidders; it never announced antipathy to White Knights. At the right price, any corporation is for sale--even a corporation managed by people who say that tender offers should be illegal. A target of a tender offer is not just for sale; it is For Sale. If Flamm drew a different inference, he was unreasonable, and Flamm's errors are not grounds on which to force other investors to pay damages (which is the effect of awarding damages against Microdot). We therefore affirm the judgment without choosing between Greenfield and Levinson.

Lest we seem to decide by silence, we disclaim any implication that if Microdot's omissions were material, the class would have been entitled to damages. The class asked for the difference between the price ($17 to $18 3/8) that prevailed between December 5 and January 23 and the price ($21) offered by Northwest. Flamm did not offer to show that the market price was influenced by the non-disclosure at any time during the offer. If the price was uninfluenced, it must have reflected in some measure the prospect of a higher bid. Flamm's theory of damages implies that the class is entitled to keep what it got for surrendering the opportunity of a higher bid and still receive the full price of that bid--all the while avoiding risk that the price will return to the original level.

The usual measure of damages in a case under Rule 10b-5 is the difference between what the stock fetched and what it would have been worth had all of the information been disclosed. Affiliated Ute Citizens, 406 U.S. at 154-55, 92 S.Ct. at 1472-73; Alna Capital Associates v. Wagner, 758 F.2d 562, 566 (11th Cir.1985). (Recovery sometimes depends on the defendant's profit, but Microdot made none by deferring disclosure.) Flamm has not tried to show that the difference was other than zero, implying no recovery. See Harris Trust and Savings Bank v. Ellis, 810 F.2d 700, 706-07 (7th Cir.1987); Mills v. Electric Auto-Lite Co., 552 F.2d 1239, 1247-49 (7th Cir.1977).

To put this a little differently, this court concluded in Mills, 552 F.2d at 1247, that the market for the stock of widely traded firms efficiently impounds publicly available information about that firm. See also Metlyn Realty Corp. v. Esmark, Inc., 763 F.2d 826, 835 (7th Cir.1985). This implies the adoption of the "fraud on the market" approach to liability under Rule 10b-5. See Angelos, 762 F.2d at 529; Levinson, 786 F.2d at 749-51; Peil v. Speiser, 806 F.2d 1154, 1160-63 (3d Cir.1986); Clark at Sec. 8.10.5; Daniel R. Fischel, Use of Modern Finance Theory in Securities Fraud Cases Involving Actively Traded Securities, 38 Bus.Law. 1, 9-16 (1982); Note, The Fraud-on-the-Market Theory, 95 Harv.L.Rev. 1143 (1982). If the plaintiff establishes that a lie, misleading statement, or omission has affected the price of the stock, he may recover without establishing that he knew of or relied on the delict; his recovery is the difference between the actual price and the price the market would have reached if traders had been fully informed. The fraud-on-the-market theory is an important ingredient of class actions in securities cases, too, for otherwise individual differences in knowledge and reliance would make the class unmanageable. See Levinson, 786 F.2d at 750; Hal S. Scott, The Impact of Class Actions on Rule 10b-5, 38 U.Chi.L.Rev. 337, 345-56 (1971).

The fraud-on-the-market approach relieves the plaintiff of the need to show that he relied on or even read the misleading or incomplete disclosures; he receives a measure of damages based on the premise that actual versus "right" price is the appropriate comparison. The logic of this approach, however, implies that for widely traded securities only fraud on the market will establish entitlement to relief. Fraud-on-the-plaintiff won't do--not when the market price itself was unaffected and therefore "right". Flamm did not try to show that the price he received had been diminished by the omission of which he complains. If we take the fraud-on-the-market approach to its logical limit, that is dispositive against him. The competing position is that fraud on the market is just one way among many to show reliance and damages; a plaintiff who demonstrates particular reliance still may recover. Both the cases and respected scholars have vacillated between these perspectives. We need not and do not decide today whether only fraud on the market allows recovery in a case of this sort, and we hope that the issue will not be passed by in silence by future litigants.

202.01 Internal Handling of Confidential Corporate Matters

Unusual market activity or a substantial price change has on occasion occurred in a company's securities shortly before the announcement of an important corporate action or development. Such incidents are extremely embarrassing and damaging to both the company and the Exchange since the public may quickly conclude that someone acted on the basis of inside information.

Negotiations leading to mergers and acquisitions, stock splits, the making of arrangements preparatory to an exchange or tender offer, changes in dividend rates or earnings, calls for redemption, and new contracts, products, or discoveries are the type of developments where the risk of untimely and inadvertent disclosure of corporate plans are most likely to occur. Frequently, these matters require extensive discussion and study by corporate officials before final decisions can be made. Accordingly, extreme care must be used in order to keep the information on a confidential basis.

Where it is possible to confine formal or informal discussions to a small group of the top management of the company or companies involved, and their individual confidential advisors where adequate security can be maintained, premature public announcement may properly be avoided. In this regard, the market action of a company's securities should be closely watched at a time when consideration is being given to important corporate matters. If unusual market activity should arise, the company should be prepared to make an immediate public announcement of the matter.

At some point it usually becomes necessary to involve other persons to conduct preliminary studies or assist in other preparations for contemplated transactions, e.g., business appraisals, tentative financing arrangements, attitude of large outside holders, availability of major blocks of stock, engineering studies and market analyses and surveys. Experience has shown that maintaining security at this point is virtually impossible. Accordingly, fairness requires that the company make an immediate public announcement as soon as disclosures relating to such important matters are made to outsiders.

The extent of the disclosures will depend upon the stage of discussions, studies, or negotiations. So far as possible, public statements should be definite as to price, ratio, timing and/or any other pertinent information necessary to permit a reasonable evaluation of the matter. As a minimum, they should include those disclosures made to outsiders. Where an initial announcement cannot be specific or complete, it will need to be supplemented from time to time as more definitive or different terms are discussed or determined.

Corporate employees, as well as directors and officers, should be regularly reminded as a matter of policy that they must not disclose confidential information they may receive in the course of their duties and must not attempt to take advantage of such information themselves.

In view of the importance of this matter and the potential difficulties involved, the Exchange suggests that a periodic review be made by each company of the matter in which confidential information is being handled within its own organization. A reminder notice of the company's policy to those in sensitive areas might also be helpful.

A sound corporate disclosure policy is essential to the maintenance of a fair and orderly securities market. It should minimize the occasions where the Exchange finds it necessary to temporarily halt trading in a security due to information leaks or rumors in connection with significant corporate transactions.

While the procedures are directed primarily at situations involving two or more companies, they are equally applicable to major corporate developments involving a single company.

CUDAHY, Circuit Judge, concurring in the judgment and concurring in part:

I admire the clear and well-reasoned majority opinion which leads us through the cases and the economic theory required to come to grips with the problem before us. I have, however, important reservations and differences in perspective which require me to write separately--but not at great length.

First, I think the case could be decided, reaching the same result as the majority, on the issues of the instructions and the admission of evidence as the parties presented them. There were errors in the instructions but, taken in context, I am reasonably confident they were harmless. The instructions erred primarily in foisting some sort of erroneous burden of investigation on the plaintiffs. But the evidence does not present any real issue of investigation, and this error seems harmless. The majority does not mention the alleged errors of admission and exclusion of evidence, but here I would also find no reversible error.

Second, I am willing, though not eager, to go beyond the issues as presented by the parties to agree that the defendants win as a matter of law. The majority feels that this approach requires the adoption of a bright-line test (agreement on price and structure) and the specific approval of Staffin v. Greenberg, 672 F.2d 1196 (3d Cir.1982) and Reiss v. Pan American World Airways, Inc., 711 F.2d 11 (2d Cir.1983).1 I understand the majority's preference for a bright-line rule. If the secrecy of merger negotiations is in most investors' ultimate best interest (a plausible premise though not necessarily one carved in stone), there is certainly an argument for keeping them quiet until agreement in principle is reached. (And, of course, here we are apparently considering only the welfare of investors.) My problems with the majority's conclusion are primarily procedural--not necessarily substantive. We have heard no argument on the important issues which the majority boldly decides based on its own--no doubt well-informed--notions of how the corporate world turns. The Securities and Exchange Commission might have given us its views, had it not thought this a case about instructions and evidence.2 I do not think it is necessary--or wise--to be so bold. I would merely hold that on the specific facts of this case there has been no nondisclosure or misrepresentation from which a reasonable jury could conclude that the securities laws had been violated. The evidence, as the majority has reviewed it, supports this conclusion, and that is all we are required to decide. To go on to lay down general rules applicable to all failure-to-disclose cases--even when the rules have been the subject of considerable litigation and academic debate--is simply not called for.

Third, I believe at several points in the majority opinion the moral underpinnings of the law are at risk in the sweep of the economic analysis. For example, in its generally approving discussion of Greenfield v. Heublein, Inc., 742 F.2d 751 (3d Cir.1984), the majority speaks kindly of lying in the interest of maintaining the secrecy of merger negotiations. Such an analysis may be economically rational and may tend to maximize investors' wealth but it does seem to raise a few old-fashioned questions of corporate morality. Without pursuing the matter here, I do not think it is the place of the courts to undermine business morality even in the name of shareholder wealth maximization.

Fourth, I think the majority opinion is interesting and generally correct in most of its observations about fraud-on-the-market theory. Although their activities are not directly relevant, chartists and tape-readers find all they need to know about securities in their price (and sometimes in their volume of trading). True "technicians" eschew balance sheets, news, rumors and information of any kind outside the price as a distraction from the real, true and complete information discounted in the price attained in an efficient market. Perhaps, this is the right approach to security analysis (although I am certainly not suggesting that this is what the majority has in mind). But the securities laws were not written by market technicians and they put an emphasis on telling the truth and the whole truth about securities. Again I believe this is a moral dimension which may or may not contribute to wealth maximization in the context of an efficient market. It is important that we not lose sight of the moral underpinnings of the law in our concern for its economic consequences.