ARGUED APRIL 4, 1996
DECIDED MAY 23, 1996
Fay Clayton, Robinson, Curley Clayton, Chicago, IL, Charles Barnhill, Jr., Davis, Miner, Barnhill Galland, Madison, WI, David B. Kahn, Mark E. King, argued, Kahn Associates, Northfield, IL, for Plaintiffs-Appellants.
Susan Getzendanner, argued, Charles F. Smith, John K. Lyons, Skadden, Arps, Slate, Meagher Flom, Stephen P. Sawyer, United Airlines, Chicago, IL, for Defendant-Appellee.
Appeal from the United States District Court for the Northern District of Illinois, Eastern Division.
No. 90 C 999
Ruben Castillo, Judge.
Before POSNER, Chief Judge, and KANNE and EVANS, Circuit Judges.
This is a class action on behalf of persons who sold either common stock of Allegis Corporation (now UAL Corporation) or puts in common stock of Allegis between October 29 and December 8, 1987. The plaintiffs complain that Allegis defrauded them, in violation of Rule 10b-5 of the Securities and Exchange Commission, by concealing information that if revealed would have caused the price of the stock to rise during the complaint period. There are other charges but they need not be discussed separately. The district judge granted summary judgment for the defendant.
There is a threshold question: whether the sellers of puts, or other options, are (as the district judge held) within the class of investors protected by Rule 10b-5. If not, they should have been dismissed from the suit without consideration of the merits of their claim that they were defrauded along with the owners of the stock.
A put entitles the holder to sell a specified security by a specified delivery date at a specified price (the "strike price") to the seller of the put. The buyer hopes that the price of the security will fall between the date of the contract and the date of delivery so that he can buy it at a price lower than the strike price, the price at which the seller of the put is obligated to purchase the stock if it is tendered. The seller of the put, conversely, hopes that the price of the security will stay above the strike price, for then the buyer will not exercise his option and the seller will have a profit equal to the price at which he sold the put. If the market price falls below the strike price, the seller of the put will lose the difference between those two prices (minus the price of the put itself). See Richard A. Brealey Stewart C. Myers, Principles of Corporate Finance 486-88 (4th ed. 1991).
So sellers of puts during the complaint period (whether the original writers of the puts or sellers in the secondary market), when the price of Allegis stock is alleged to have been artificially depressed, should have been happy rather than sad when the price jumped back up, as distinct from sellers of the stock itself, who lost this profit if they sold their stock during the period of depressed prices, and writers of puts before the complaint period (clearly not members of the plaintiff class), who may have had to honor the puts during the complaint period, when the price of the stock was depressed. The plaintiffs seem not to understand that stock and put prices move inversely to each other. But the defendant makes nothing of this, so we shall swallow our doubts and assume that some of the options traders whom the complaint, perhaps inaccurately, calls sellers of puts were hurt.
Puts and other stock options are securities within the meaning of the Securities Exchange Act, 15 U.S.C. § 78c (a)(10); Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 750-51 (1975), but they are not necessarily (or in this case) securities issued by the corporation that issued the underlying securities, the subject of the option contracts. For the proposition that Rule 10b-5 does not make the issuer of the underlying securities responsible for the effects of a securities fraud on the owners of options on those securities, the defendant cites Laventhall v. General Dynamics Corp., 704 F.2d 407, 412-14 (8th Cir. 1983). The leading case in favor of issuer's liability is Deutschman v. Beneficial Corp., 841 F.2d 502, 506-07 (3d Cir. 1988); see 8 Louis Loss Joel Seligman, Securities Regulation 3601 and n. 360 (3d ed. 1991). The objection to liability is that the extent of harm to the option holders depends on factors beyond the control of the issuer of the underlying securities, including the number of options and the terms of the option contracts.
The division of authority is illusory. Laventhall was an insider-trading case. The decision against liability turned on the fact that a corporation has no fiduciary obligation to option traders, as it does to its own shareholders. The existence of fiduciary obligation was critical because insider trading is the abuse of a fiduciary position, Chiarella v. United States, 445 U.S. 222 (1980), rather than out-and-out fraud, that is, fraud accomplished by misrepresentations (or, what amounts to the same thing, misleading omissions). The duty not to make misrepresentations does not depend on the existence of a fiduciary relationship. Basic, Inc. v. Levinson, 485 U.S. 224, 240 n. 18 (1988); Elizabeth M. Sacksteder, Note, "Securities Regulation for a Changing Market: Option Trader Standing Under Rule 10b-5," 97 Yale L.J. 623, 641 (1988). If it did, very little fraud would be actionable. The garden-variety fraud in which the seller of a product misrepresents its qualities to the buyer would not be, because a seller is not his buyer's fiduciary.
So the case law provides no support for the defendant's position. And since fraud (whether it takes the form of a misrepresentation or of an abuse of a fiduciary position) is a deliberate wrong, the considerations that lead courts to limit the scope of liability for accidental wrongs to consequences that are in some sense "foreseeable" are not in play. Option traders play an important role in maintaining the efficiency of the securities markets, moreover, so they have a claim to protection against being defrauded.
Two arguments can be made against liability to options traders. First, since it is very difficult to see how they can be hurt by a securities fraud without the owner of the optioned securities being hurt, a more limited liability should be sufficient to deter most securities frauds. Second, it is very difficult for the corporation that issued the underlying securities to gauge its potential liability for a securities fraud if options traders are within the protected class. We have to worry about the danger of overdeterrence because, while securities fraud is nominally a species of deliberate wrongdoing, the doctrines of securities law and particularly the application of those doctrines to particular factual situations are so difficult, complex, and uncertain that there is a serious danger of erroneous impositions of liability.
But these two Rubicons were crossed when short sellers were held to have standing to sue under Rule 10b-5, Zlotnick v. TIE Communications, 836 F.2d 818, 821 (3d Cir. 1988); cf. United States v. Russo, 74 F.3d 1383, 1392 (2d Cir. 1996), a holding that we do not understand the defendant to question. The corporation has no control over the number of its shares that are shorted — the number may in fact exceed the total number of outstanding shares, Sullivan Long, Inc. v. Scattered Corp., 47 F.3d 857, 859 (7th Cir. 1995) — or over its potential liability, which is the difference between the price of the short sale and the (higher) market price when the short seller has to cover. The potential liability to short sellers is thus greater than that to either the buyers of puts, who can lose only the price they have paid for the put, or the sellers of puts, who can lose only the difference between the strike price and a lower market price (but not lower than zero of course), minus the price of the put. Only the writer of a call, who like a short seller agrees to sell stock to the buyer at the strike price, obliging the writer of the call to buy stock (unless he owns it already) at the market price — however high it goes — if necessary to honor the call, has as much exposure to loss as a short seller.
We don't, strictly speaking, have to resolve the issue of the right of buyers of puts, or other option traders, to sue under Rule 10b-5. We described it as a threshold issue but it is not jurisdictional. It has nothing to do with "standing" in the Article III sense; the buyers were harmed as a consequence of the defendant's conduct and would be helped if the relief they seek is granted. It is true that when Congress makes clear that a statute is not intended to confer rights on a particular class of persons, a suit under the statute by a member of that class does not engage the jurisdiction of the federal courts, Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U.S. 804, 814 (1986), so that the court must dismiss it even if neither party notices the problem, just as if there were no standing in the Article III sense. But if the scope of the statute is unclear, the question whether a particular class is protected by it becomes just another issue concerning the merits of the suit and therefore one that the court need not decide if, as we shall see is the case here, another issue is fully dispositive. This is especially true where, as is also true in this case, that other dispositive issue cannot be avoided. The plaintiffs include sellers of Allegis stock as well as option traders, and the rights of those sellers do not depend on whether option traders are entitled to participate in the suit.
Still, since the issue misnamed standing has been fully briefed and is fully ripe for decision, and its resolution does not appear to be open to substantial doubt, and the investment community is entitled to know where it stands on important issues of liability, we think it best not to be coy. We hold that option traders have standing to sue under Rule 10b-5.
This bring us to the main question presented by the appeal, which is whether the district judge erred in granting summary judgment for Allegis on the ground that there was insufficient evidence of fraud to create a jury issue. The story told by the plaintiffs begins in May of 1987, at which time Allegis, in addition to owning United Air Lines, its single biggest asset, also owned the Hilton and Westin hotel chains and the Hertz car-rental company. A group of investors known as Coniston Partners, which owned 13 percent of the common stock of Allegis, announced that they wanted to take over Allegis's board of directors in order to get the company to sell its nonairline properties, which were two-thirds of its total assets. Allegis quickly yielded to the demand for a bust up, announcing the following month that it would sell its nonairline assets and distribute the proceeds to the shareholders. There was still the question of how to make the distribution. One possibility was to distribute the proceeds to the shareholders in the form of a dividend. Another was a self-tender, in which Allegis would offer to buy a stated number of shares, using the proceeds of the sale of the nonairline assets to fund the purchase. Still another possibility (market repurchase), but one that received little consideration, was for Allegis to go into the market and buy up as many shares of its stock as the proceeds would enable it, and the rules of the Securities and Exchange Commission would permit it, to do. Some consideration may have been given to Allegis's issuing puts on its stock and using the proceeds of the sales of the assets to buy back stock at the price specified in the puts. Not considered at all was a spin off of the nonairline subsidiaries, which would have given Allegis's shareholders shares in the spun-off companies to add to their Allegis shares. See generally Brealey Myers, supra, at 373-74.
Coniston's demand for a bust up did not specify the mode of distributing the proceeds. When the question was first discussed by the board of directors, at meetings in June and July of 1987, only the first two possibilities that we have outlined, a dividend distribution and a self-tender, were discussed, and the latter only very briefly. In August, Coniston raised for the first time the question whether a dividend distribution would be a tax-efficient method of distribution for it. A cash dividend is taxable to the recipient as ordinary income. If instead the proceeds of the sales of the nonairline assets were used to buy up shares, as in a self-tender, the sellers of the shares, to the extent that they received more than they had paid for them, would be taxed on this profit at capital-gains rates, which at the time were lower. 26 U.S.C. § 1(h), (j) (1987).
The net difference in tax consequences between the two methods of distribution might be larger or smaller than the difference between ordinary-income and capital-gains tax rates, as a simplified example will show. Suppose that Corporation A has two wholly owned subsidiaries (and no other assets), one worth $50 per share, the other $40 per share, so the price per share is $90. The second subsidiary, however, would be worth more were it owned by someone else. So A decides to sell it, anticipating that the sale of the subsidiary will yield $50 per share, so that when the sale is announced the price of A's stock will rise to $100. If the sale of the subsidiary goes through and the proceeds — $50 per share — are distributed to the shareholders, the price of the stock will drop back to $50. Therefore a shareholder who does not want to pay income tax at ordinary-income rates on the $50 dividend has only to sell his stock, at $100, before the dividend is paid. He will have a gain measured by the difference between the price he paid for his stock and $100 but he will pay income tax on that gain at the capital-gains rather than the ordinary-income rate. This method of avoiding the higher tax rate on ordinary income is called "selling into the market."
If instead of declaring a dividend our hypothetical Corporation A makes a self-tender for half its shares (half because the company is shrinking to half its original size as measured by value) at their current market value of $100 per share, a shareholder who accepts the tender will pay capital gains as before. But a shareholder is not required to accept a tender, and in the event he decides not to he won't have to pay any tax at the time of the tender because he will not have realized his gain. Deferral of tax is a form of tax savings because of the time value of money, unless tax rates are expected to be much higher in future years.
The tax consequences of the self-tender route depend, however, on the demand for Corporation A's stock by investors who are indifferent as between ordinary income and capital gains, perhaps because they are exempt from income tax. If there are few of them, it may be difficult to sell into the market at a good price, since any purchaser from such a seller who is not tax exempt will have to pay income tax at ordinary-income rates on the dividend that he will receive as the owner of the stock on the day the dividend is paid. Today, however, the market is thick with tax-exempt investors, such as pension funds and university endowment funds, and they are fond of bluechip stocks such as Allegis. Conceivably, they would be willing to buy all the Allegis stock offered by shareholders wanting to sell into the market, and in that event the price of the stock would not be depressed by purchasers' aversion to obtaining ordinary income in lieu of capital gains.
Taxes are not the only consideration in the choice between methods of distribution. An advantage of the self-tender is that a large shareholder need not reduce his influence over corporate management by selling any of his shares. An advantage of the dividend approach is that it gets the proceeds of the sale of the assets into the hands of the shareholders in the most direct way possible, without any intervening stock transaction and hence with minimum transaction (and also regulatory-compliance) costs.
Despite Coniston's preference for a self-tender, the next time the board discussed the method of distributing the proceeds of the contemplated sale of the nonairline assets, late in September, the only method discussed was the dividend. A few weeks later, on October 19, 1987, the stock market had its biggest crash since 1929. Allegis's stock plunged nearly 30 percent. Nevertheless, ten days later the board issued a press release saying that the proceeds of the sales of the subsidiaries would be distributed to the shareholders in the form of dividends. The following month the board began having second thoughts, advised by its investment banker that the stock market crash had made self-tender an attractive alternative to cash dividends because with the price of the stock depressed many shareholders who would want to convert ordinary income to capital gains by selling into the market in advance of the payment of the dividend would be reluctant to do so, thinking it a bad time to sell. Believers in the efficient-markets hypothesis of finance theory think no time can be known in advance to be a better or a worse time to buy or sell stock, but they are still a minority of investors.
On December 9, 1987, the board announced in a press release that it would consider both alternatives — dividend and self-tender, rather than just the first. And on January 28, 1988, after Coniston delivered an "ultimatum" to the board demanding self-tender, the board decided upon that method of distribution.
The plaintiffs argue that Allegis knew all along that self-tender would be the method of distribution, because that is what Coniston wanted; the only reason for the announcement of the dividend approach on October 29 was to put pressure on Coniston to relinquish its other demands, such as for control of the board. Because, the plaintiffs continue, self-tender was far more advantageous to most shareholders than the dividend approach, the announcement of October 29 caused share prices to fall, thus hurting shareholders who sold between then and December 8, when news that the company was at last considering a self-tender got about, ending the false impression that the October announcement had created.
There is no direct evidence either that Coniston before December 8 was insistent on a self-tender or that Allegis knew this and knew also that it must yield to Coniston's demand. The plaintiffs argue that the dividend method of distributing the proceeds of the sale of Allegis's non-airline assets was so patently irrational that Allegis must have been lying when it said that this was the method it was going to use. They support the argument with evidence that most recent bust ups have been implemented through self-tenders rather than dividend distributions. But they do not counter Allegis's evidence that the dividend approach has advantages that might have commended it to the board. A self-tender would require the filing of an offering statement with the SEC; a dividend would not. A self-tender would require each shareholder to make a decision (whether to accept the tender or not), and small shareholders might be in a quandary as to what to do, whereas the dividend would be automatic. Tax-indifferent shareholders, such as pension funds, might prefer the dividend because it would reduce transaction costs; they would not have to sell into the market, in order to obtain a tax advantage.
Against these advantages must be set the disadvantage to taxpaying shareholders of having to pay income tax at ordinary-income rather than capital-gains rates, or, if they sold into the market, to pay capital-gains taxes sooner rather than later; but the plaintiffs refuse to acknowledge the possibility that selling into the market might reduce the tax disadvantages of the dividend route substantially, even though not all the way to zero. When they say that Coniston would have incurred a $100 million tax liability if the proceeds of the sales of the nonairline assets had been distributed as dividends, they are ignoring the possibility of Coniston's selling into the market. True, Coniston would not want to do this if it wanted to retain its influence over Allegis, which depended on the size of its shareholdings. So far as appears, however, the main reason Coniston had wanted to gain control of Allegis was to get it to divest itself of its nonairline subsidiaries, which would be the result whether the company took the dividend or the self-tender path.
No rational jury could infer from a comparison in the abstract of the two methods of distributing the proceeds of the contemplated sales that Allegis was lying when it announced that it was going to use the dividend method. It thus becomes critical to determine whether there is any evidence besides the comparison. There is evidence that Coniston had always preferred the self-tender route, rather than coming gradually to a firm decision issuing in the ultimatum of January 1988. For in its self-tender offer, made the following month, Allegis acknowledged that back in the summer and fall of the previous year Coniston had "supported the making of a cash tender offer for a significant portion of the Shares rather than the payment of a dividend." There is also evidence that Coniston preferred to have Allegis write puts on its stock, a method of distribution which would operate like a self-tender, and that Coniston expected Allegis eventually to accede to Coniston's preferred method of distribution whatever that should turn out to be. But there is no evidence that Coniston's preference for a self-tender or the equivalent was so intense that Allegis's board of directors must have been fooling, or bluffing, when it opted for the dividend route. Coniston knew how to make its demands known to Allegis. It demanded a bust up, and got it, and in January 1988 it demanded a self-tender, and got that too. The absence of any evidence of a demand for the self-tender before January 1988 — the sort of evidence that was held to preclude summary judgment in Kronfeld v. Trans World Airlines, Inc., 832 F.2d 726, 729, 737 (2d Cir. 1987) — is a telling indication that until then Coniston was willing to go along with the dividend approach, which was the board's preference.
The plaintiffs try to turn this pregnant silence to their advantage by arguing that it suggests a cover-up by Allegis and Coniston (who are now friends). But a case cannot be built on the absence of evidence. One reads that it is an occupational failing of counterintelligence services that "those who believe in the enemy in our midst have minds insulated against evidence; its absence proves the extreme skill of the enemy." A.W. Brian Simpson, In the Highest Degree Odious: Detention Without Trial in Wartime Britain 108 (1992). That won't do as a theory of Rule 10b-5 liability. The natural interpretation of the absence of evidence that Coniston was demanding self-tender from the start is simply that not until sometime after October 29 did Coniston decide to make an issue of the method of distribution. Although it was a very powerful shareholder of Allegis (more precisely an alliance of shareholders) — the defendant's lawyer herself, at argument, described Coniston as an 800-pound gorilla — it did not control Allegis and even a very powerful shareholder does not insist on getting his way every time. And even if, despite its lack of control, Coniston was at all times powerful enough to make Allegis dance to its tune, perhaps because it could make a credible threat to wage a successful proxy fight for control of the board, the plaintiffs have failed to show that the additional tax liability faced by Coniston if the dividend route were followed was so large that it is inconceivable that Coniston would have permitted Allegis to follow that route, and that Allegis knew this on October 29.
To prove fraud by purely circumstantial evidence is difficult, and the plaintiffs have not been able to surmount the difficulty. Although Allegis may have known on October 29 that the dividend proposal was a complete nonstarter, there is no direct evidence of this and the plaintiffs' effort to prove it indirectly founders on its inability to show how great a tax disadvantage Allegis faced if the proposal was adopted.
We can imagine other kinds of indirect evidence of fraud. Had Coniston gone ballistic when the dividend proposal was announced, this would be some evidence, alternative to an actual calculation of its incremental tax liability under the proposal, that the tax disadvantage to it was so great as to make the proposal obviously nonviable, hence nonserious and merely a gambit in its ongoing struggle with the board of directors. But there is no evidence that Coniston protested before the end of the complaint period. The stock market, moreover, believed that the dividend proposal was viable. Otherwise the price of the stock would not have risen when it became known that Allegis would probably switch to a self-tender — the switch would have been anticipated and so discounted in the price of the stock. If it is obvious, as the plaintiffs believe, that the self-tender route would be the one eventually followed, why was the market surprised when Allegis disclosed that it was considering that route? The market is not infallible, but a case of fraud cannot be founded entirely on conjecture.
The plaintiffs argue, finally, that the October 29 announcement was misleading because Allegis's lawyers and investment bankers were still studying the tax consequences of the dividend approach. All the record shows, however, is, in the plaintiffs' own word, "sporadic" study of the issue both before and after the announcement. Even after the board decided on October 29 to go the dividend route, it could be expected to continue studying the tax issue, since the announcement might cause squawks among shareholders facing a heavy tax liability that a self-tender might reduce. Eventually it did cause a loud squawk — from Coniston. There is no evidence of an earlier squawk.
The district judge was right to grant summary judgment for the defendant.