On March 10 of this year Dynamics Corporation of America, which already owned 9.6 percent of the common stock of CTS Corporation, made a tender offer for another million shares. The offer if accepted would bring its stock holdings up to 27.5 percent of the company. On the same day, Dynamics filed this suit in the federal district court in Chicago; as later amended, the suit sought to enjoin the enforcement of Indiana's statute regulating takeovers, on the ground that the statute violates the supremacy and commerce clauses of the federal Constitution. When CTS "opted in" to a new Indiana statute on the subject, the Control Share Acquisition Chapter as it is called, Ind.Code Secs. 23-1-42-1 et seq., Dynamics further amended its complaint to challenge the new statute on the same grounds. A pendent count in the complaint sought to enjoin CTS from enforcing a recently adopted shareholders' rights plan ("poison pill"), on the ground that it violated the fiduciary obligations of CTS's management toward its shareholders. CTS counterclaimed against Dynamics, seeking an injunction against the tender offer on the grounds that it would result in a violation of section 8 of the Clayton Act, 15 U.S.C. Sec. 19 (interlocking directorates), and that it failed to disclose material information. There are some individual parties, but as they are not important to the legal issues we shall ignore them.

Both Dynamics and CTS moved for preliminary injunctions. After a month of frantic pretrial discovery a one-day evidentiary hearing was held before the district judge, who in a series of orders then ruled that the poison pill plan violated Indiana law, that the Indiana statute violated both the supremacy and commerce clauses of the federal Constitution, and that CTS was not entitled to a preliminary injunction. 637 F. Supp. 389 (N.D.Ill. 1986). She therefore granted the preliminary injunction requested by Dynamics, 637 F. Supp. 406 (N.D.Ill. 1986). CTS, joined by the Attorney General of Indiana, who has intervened in the case to defend his state's statute, appeals under 28 U.S.C. Sec. 1292(a)(1), which allows an immediate appeal from an order granting or denying a preliminary injunction. We accelerated our consideration of the appeal because Dynamics had only till April 24 to decide whether to buy the shares tendered in response to its offer. We heard argument on April 23 and later that day affirmed the district judge's orders, with a notation that an opinion explaining the grounds of our decision would follow.

The main issues we must address are the lawfulness of CTS's poison pill scheme, the district court's compliance with a federal statute requiring that a state's attorney general be notified that the constitutionality of a statute of his state is being challenged, the constitutionality of the new Indiana takeover statute under the supremacy clause and also under the commerce clause, the significance of the potential violation of the Clayton Act, and the adequacy of the disclosures made in the tender offer.

Before taking up these issues we shall comment briefly on the procedural posture of the case in this court, an appeal from orders granting and denying requests for preliminary injunctions. As emphasized in our recent opinions, in different but compatible formulations, the task for a district judge asked to grant a preliminary injunction is to compare the irreparable harm to the plaintiff if the injunction is denied, weighted by the likelihood that the denial would be erroneous because the plaintiff will prevail in the plenary trial, with the irreparable harm to the defendant if the injunction is granted, weighted by the likelihood that the grant would be erroneous because the defendant, not the plaintiff, will prevail in the trial. See Lawson Products, Inc. v. Avnet, Inc., 782 F.2d 1429, 1433-34 (7th Cir.1986); American Hospital Supply Corp. v. Hospital Products Ltd., 780 F.2d 589, 593 (7th Cir.1986); Roland Machinery Co. v. Dresser Industries, Inc., 749 F.2d 380, 387-88 (7th Cir.1984). So, for example, the greater the probability that the plaintiff will win the case in the end, the less irreparable harm he need show relative to the defendant in order to get the preliminary injunction. If both parties are likely to suffer the same amount of irreparable harm, so far as estimation is possible, then likelihood of success becomes decisive. See American Hospital Supply Corp. v. Hospital Products Ltd., supra, 780 F.2d at 598. That seems a reasonable description of the present case; the fact that both parties sought preliminary injunctions does not affect the analysis.

If the tender offer is blocked, Dynamics will lose an opportunity that may never recur. The present owners of shares in CTS who have tendered them to Dynamics may lose, too, but their loss is easily quantified--it is the difference between the price in the tender offer and the price to which CTS stock falls. The only problem is uncertainty as to how many shares Dynamics would actually have bought if its offer was oversubscribed, as it was. But, conversely, if the tender offer is not blocked, CTS's shareholders will be irrevocably harmed if, as CTS predicts, they are stampeded into selling their shares to Dynamics for a lower price than the shares would eventually command if the tender offer were defeated, or if shareholders who do not tender are coerced into surrendering their shares later at inadequate prices in a "back-end" deal engineered by a board of directors that Dynamics is expected to control if its tender offer succeeds. The harms are very difficult to quantify and it seems best to treat them as offsetting. Hence Dynamics is entitled to the relief it sought (including the denial of the preliminary injunction sought by CTS) if but only if it is more likely than CTS to prevail at a full trial, in the unlikely event that one is ever held.

When the only issue on appeal from an order granting or denying a request for a preliminary injunction is likelihood of success, the role of the appellate court is little different from that in an appeal from a final judgment. Review of the district judge's legal rulings is plenary, review of the judge's findings of fact limited to clear errors. Since the complex, particularistic, often intuitive process of weighing likelihood of success against the balance of irreparable harms is not involved when that balance is assumed to be even, the deference that we give the district judge's striking of that balance is not a factor in this appeal.

Against this background we first ask whether the district court was right to conclude that the adoption of the poison pill violated the fiduciary obligations of CTS's management to its shareholders. The parties agree both that the question is governed by the common law of Indiana and that Indiana takes its cues in matters of corporation law from the Delaware courts, which are more experienced in such matters since such a large fraction of major corporations is incorporated in Delaware and such a small fraction in Indiana.

The whole issue of permissible defensive tactics in the face of a tender offer is immensely contentious, and it is no business of ours, whose duty on this branch of the appeal is only to predict how the Indiana courts would evaluate CTS's poison pill maneuver, to choose sides. There are two polar positions in the debate. One views hostile takeovers as a bad thing, on a variety of grounds such as that they make managers of companies that are potential targets of takeover bids worry too much about short-term financial results and that they promote absentee ownership and control. See, e.g., Scherer, Takeovers: Present and Future Dangers, Brookings Rev. (winter-spring 1986), at 15; Herman, Corporate Control, Corporate Power 100-01 (1981). Whether or not Dynamics ever merges CTS into it, the parties seem agreed that if the tender offer succeeds, Dynamics, as by far the largest shareholder of CTS, will probably be able to elect a majority of the board of directors. Dynamics is a New York corporation with headquarters in Connecticut, CTS an Indiana corporation with headquarters in Indiana. The record is not clear on where the firms' assets and employees are concentrated, and indeed reveals little about the companies except that CTS is a manufacturer of electronic and electromechanical components and Dynamics a diversified manufacturer of consumer and industrial products and that both are large companies whose stock is traded on the New York Stock Exchange.

The other pole is that all resistance to takeover attempts is bad. See, e.g., Easterbrook & Fischel, The Proper Role of a Target's Management in Responding to a Tender Offer, 94 Harv.L.Rev. 1161 (1981); Gilson, A Structural Approach to Corporations: The Case Against Defensive Tactics in Tender Offers, 33 Stan.L.Rev. 819 (1981); cf. SEC Office of Chief Economist, A Study on the Economics of Poison Pills, Fed.Sec.L.Rep. (CCH) p 83,971 (March 5, 1986). The market price of publicly traded stock impounds all available information about the value of the stock, and anyone who offers a higher price (Dynamics' tender offer price was $43, and when the offer was made CTS's stock was trading at $36) thereby offers an unequivocal benefit to the shareholders of the target firm, which management if it is really a fiduciary of the shareholders should embrace rather than oppose. In that way the market for corporate control will be kept fluid and corporate assets will be transferred, with a minimum of friction, to those who value them the most, as measured by the prices they offer. See also Ginsburg & Robinson, The Case Against Federal Intervention in the Market for Corporate Control, Brookings Rev. (winter-spring 1986), at 9.

It is a safe prediction that the Indiana courts would reject the polar views, as the Delaware courts have done. To allow management to use its control of the board of directors to frustrate all hostile takeovers would nullify an important protection for shareholders. The threat of hostile takeover plays a vital role in keeping management on its toes. CTS has been a troubled firm of late; a major acquisition (which Dynamics, long a major shareholder, opposed) soured, and was written off at a large loss. If CTS's management is allowed to insulate the company from any change of control to which management does not agree, the shareholders may be unable to realize the potential value of their investment. Maybe under different control, specifically Dynamics' control, CTS would be a more valuable company; then its shareholders would benefit from a takeover, hostile or otherwise. It is only human for CTS's officers and directors to doubt that a company which, if it takes control of CTS, will fire them can actually do a better job of running "their" company. But it is not their company; at least it is not supposed to be. It is supposed to be the shareholders' company, for it is they who are entitled to all the income that the company generates after paying off all contractually or otherwise obligated expenses. The officers and directors are the agents and fiduciaries of the shareholders and owe a duty of complete loyalty which is inconsistent with erecting insuperable barriers to hostile takeovers.

But it does not follow that loyalty requires passivity. See, e.g., Bebchuk, The Case for Facilitating Competing Tender Offers, 95 Harv.L.Rev. 1028 (1982). If someone stops you on the street and says, "Say, that's a beautiful watch you're wearing--I'll give you $250 for it," you won't necessarily agree to the sale even if the watch is worth only $100 to you (apart from any sale value it may have). You may want to see whether you can sell it for even more than $250, now that you have an inkling of what its market value may be. Likewise the first tender offer may not be the best. It is true that each shareholder can decide for himself whether he is likely to do better by holding out for a better offer, and if enough decide this way the offer will fail and maybe the offeror, or some other investor, will offer more. But many shareholders are passive investors. They know little about the companies in which they invest or about the market for corporate control. They want to be told whether they should sell their shares now or wait for a better offer to come along. Better yet--for it is difficult to absorb information about a subject you don't know much about--they want management to create a process that will maximize the value of their shares in a takeover situation. That may require the use of some defensive tactics.

One defensive tactic that is permitted and indeed required by federal law is, as we shall see, to have a cooling off period between the announcement and the consummation of a tender offer. Such a period gives the shareholders time to do some shopping around. This is not an unmixed blessing. It reduces the expected gain to the first maker of a tender offer, for now he may have to compete with other offerors and may have to raise his price--a possibility that will reduce the likelihood of a tender offer's being made in the first place. On the other hand, the waiting period creates an opportunity for other investors to make competing offers, and thus encourages an auction of the firm's assets with more than one bidder. Maybe the second effect dominates the first. To take another example, if a corporation offers its key managers "golden parachutes" (generous severance pay in the event they lose their jobs because of a takeover), this may make them resist takeovers less; and the benefits to shareholders may exceed the costs of the golden parachutes themselves as well as the effect of the parachutes in making the takeover more costly to the acquiring firm.

Conceivably even the "poison pill" may, if it is not actually lethal, benefit shareholders of the target firm. This wonderfully vivid term refers to a family of shareholder rights agreements which, upon some triggering event such as the acquisition by a tender offeror of a certain percentage of the target corporation's common stock, entitle the remaining shareholders to receive additional shares of common stock (or other securities) at bargain prices. Suppose that the tender offeror makes an offer for 51 percent of the stock of the target firm, knowing that if the offer succeeds he can then force out the minority shareholders by voting his shares for a swap of the target's assets for cash (a cash merger). Although minority shareholders who are squeezed out in this fashion have a legal right to receive the fair value of their shares, that value may be less than either the tender offer price or the value that their shares would command in the market had there been no tender offer. A poison pill triggered by the acquisition of a majority stock interest gives the remaining shareholders more shares. This both improves their position if the tender offer succeeds and makes all shareholders less frantic to tender. Without the poison pill, shareholders will compete to tender their shares because if they do not they may miss out on an opportunity to sell their shares at a premium price and thus to escape the fate of the minority shareholder. But at the same time, by making a tender offer less certain to succeed and more costly to the offeror, the poison pill may reduce the number of tender offers and the price of each offer, thus hurting all shareholders ex ante.

The tradeoffs obviously are complex, so it is no surprise to find that the evidence on whether particular defensive tactics enhance or reduce shareholder welfare is mixed. See Jensen & Ruback, The Market for Corporate Control: The Scientific Evidence, 11 J.Financial Econ. 5, 29-40 (1983). An intriguing recent finding is that targets that resist tender offers yet are later acquired do better, at least in the short run, in maximizing shareholder wealth than targets that do not resist. The qualification is vital; if, as seems likely, defensive tactics reduce the number of tender offers, then shareholders may lose in the long run. Moreover, the study finds that if the target resists so stubbornly that it is not acquired later, its shareholders are made unequivocally worse off. See Jarrell, The Wealth Effects of Litigation by Targets: Do Interests Diverge in a Merge?, 28 J.Law & Econ. 151 (1985). But maybe, even with the risk that resistance will be too successful or that defensive tactics may eventually weaken the market for corporate control and hurt all shareholders, some resistance is the optimal strategy for managements perfectly loyal to their shareholders to follow. If so, the adoption of some defensive measures, perhaps even some poison pills, may be in the interest of all the corporation's shareholders, though we are not aware of any rigorous study which finds that poison pills help shareholders.

Personally we are rather skeptical about the arguments for defensive measures. They strike us as giving too little weight to the effect of "defensive" measures in rendering shareholders defenseless against their own managements. (The shareholders of CTS were not asked to approve the poison pill.) We are especially skeptical about the arguments used to defend poison pills. If the present case is representative, the poison pill seems (as we shall see) more a reflex device of a management determined to hold on to power at all costs than a considered measure for maximizing shareholder wealth. Unlike a fair price amendment, which requires the tender offeror to pay the same price to the nontendering as to the tendering shareholders in order to head off a stampede to tender that may reduce the price of the tender offer, the poison pill can (in this case did) substantially dilute the tender offeror's shares, thereby defeating the object of the offer, which is to take over the company in the hope of squeezing more profit out of its assets. Depending on its terms the poison pill may also make each shareholder think himself better off not tendering, hoping instead to get the goodies that nontendering shareholders receive by virtue of the poison pill. But of course if no one tenders, no tender offer can succeed.

So we have grave doubts about poison pills. But our personal views on a matter committed to the authority of the states are not terribly important; and given the complexity of the issue it is understandable why state courts would hesitate to condemn all defensive measures (even all poison pills) as breaches of fiduciary duty, on the basis of the present incomplete evidence of the actual effects of these measures.

Indeed, the Delaware courts have been quite emphatic that defensive measures in general and poison pills in particular are within the power of the board of directors of a target corporation. E.g., Revlon, Inc. v. MacAndrews & Forbes Holdings, Inc., 506 A.2d 173, 180 (Del.1986). But at the same time these courts have insisted that the measures be plausibly related to the goal of stockholder wealth maximization. See id. ("when a board implements anti-takeover measures, ... [the] potential for conflict [of interest] places upon the directors the burden of proving that they had reasonable grounds for believing there was a danger to corporate policy and effectiveness, a burden satisfied by a showing of good faith and reasonable investigation"); Moran v. Household Int'l, Inc., 500 A.2d 1346, 1356 (Del.1985) ("when the business judgment rule applies to adoption of a defensive mechanism, the initial burden [of proving that the directors acted on an informed basis, in good faith, and in the honest belief that the action taken was in the best interests of the company] will lie with the directors"); Unocal Corp. v. Mesa Petroleum Co., 493 A.2d 946, 954 (Del.1985) ("Because of the omnipresent specter that a board may be acting primarily in its own interests, rather than those of the corporation and its shareholders, there is an enhanced duty which calls for judicial examination at the threshold before the protections of the business judgment rule may be conferred"). The shifting of burdens adopted in these decisions was anticipated by Judge Cudahy's dissenting opinion in Panter v. Marshall Field & Co., 646 F.2d 271, 299-304 (7th Cir.1981). See also Hanson Trust PLC v. ML SCM Acquisition Inc., 781 F.2d 264, 273 (2d Cir.1986) (dictum).

Thus the Delaware courts have not, as CTS argues, written targets' management a blank check endorsed with "business judgment rule." This rule expresses a sensible policy of judicial noninterference with business decisions made in circumstances free from serious conflicts of interest between management, which makes the decisions, and the corporation's shareholders. Not only do businessmen know more about business than judges do, but competition in the product and labor markets and in the market for corporate control provides sufficient punishment for businessmen who commit more than their share of business mistakes. When however there is a serious conflict of interest, and in particular when management is making decisions that may thwart the operation of the market in corporate control, the judicial role (under Delaware law, and we assume Indiana law as well) is less deferential. When managers are busy erecting obstacles to the taking over of the corporation by an investor who is likely to fire them if the takeover attempt succeeds, they have a clear conflict of interest, and it is not cured by vesting the power of decision in a board of directors in which insiders are a minority (five of CTS's eight directors are outsiders). No one likes to be fired, whether he is just a director or also an officer. The so-called outsiders moreover are often friends of the insiders. And since they spend only part of their time on the affairs of the corporation, their knowledge of those affairs is much less than that of the insiders, to whom they are likely therefore to defer.

These problems have seemed serious enough to warrant a more searching judicial review of corporate decisions concerning defensive measures to takeovers than of decisions concerning ordinary business decisions. Such review is not without its costs. It makes directors overcautious, makes people reluctant to serve as directors, drives up directors' fees and officers' and directors' liability insurance rates, and leads boards of directors to adopt ponderous, court-like procedures. But the price is one the courts have been willing to pay.

From these general reflections we turn at last to the particulars of this case. Since the present management of CTS took over the company in 1981, a year after Dynamics first became a major shareholder, CTS's rate of return has declined substantially, in part because of a series of acquisitions to which Dynamics objected and which indeed turned out to be flops. Dynamics therefore coupled its March 10 announcement of the tender offer with a declaration that it would field a slate of candidates for the board of directors election scheduled for April 24 (since postponed). On the very same day (March 10, which was also the day this lawsuit was filed) CTS's management announced its opposition to Dynamics' actions--without having studied their business and financial implications or even having consulted CTS's outside directors. The next day CTS retained as its investment advisor Smith Barney under a contract whereby the advisor would receive a bonus if Dynamics lost the proxy fight. The following day the board met to discuss the matter. Without consulting the board, CTS's chairman wrote the shareholders the next day urging them not to vote for Dynamics' slate. On March 22 Smith Barney presented its poison pill proposal to the board, with an accompanying "fairness opinion" in which it opined that Dynamics' tender offer was unfair but did not opine on whether the $43 price in the offer was fair or unfair. The board did not discuss price either, but did at that very meeting unanimously adopt the poison pill.

The tender offer was not evaluated in a cool, dispassionate, and thorough fashion. We do not mean to suggest that the board was obliged to accord due process to Dynamics; we have no desire to judicialize board of directors meetings. But it is apparent that the insiders on the board, in particular the chairman, decided from the start to block the tender offer, before its ramifications for shareholder welfare were considered; judgment first, trial later, as the Queen of Hearts said in Alice in Wonderland. Smith Barney held itself out as a blocker, and would have lost its $75,000 bonus if it had advised the board that the tender offer was fair and if the end result of this advice had been Dynamics' wresting control of the board from the existing directors. How the fairness of the tender offer could be determined without any consideration of the fairness of the offer price is mystifying. CTS argues that not all the shareholders could get the tender offer price, because Dynamics was seeking tenders of only 17.9 percent of the shares. This is doubly misleading. Every fifth shareholder (.179 divided by .904--the other 9.6 percent of the shares being owned by Dynamics already--is .198) would get that price. More important, all the shareholders benefited ex ante. Upon announcement of the tender offer the price of CTS stock rose from below $36 to above $40 (a much bigger percentage leap than the NYSE Composite, S & P 500, or Dow Jones indexes took that day). Those shareholders who did not expect to be in the lucky one-fifth to sell to Dynamics at $43 could sell the day the offer was announced for $40. (Of course some stockholders would not have known about the movement in the price and some who did know would adhere to a buy-and-hold strategy, and so not sell.) It is worth noting that the price of CTS shares dropped substantially the day after the poison pill was announced, and rose again when the district judge's order invalidating the poison pill was released.

CTS argues that it did not need to investigate the tender offer in order to know it was bad, for there was a long history of bad blood between it and Dynamics. Dynamics had been a very restive, unhappy shareholder. There had been litigation between the companies. CTS's management had formed the judgment that the companies had incompatible philosophies, with CTS focusing on the long term and hence making substantial capital investments and Dynamics going for the quick buck. It knew a takeover of CTS by Dynamics would be bad. But given CTS's disappointing performance in recent years and the fact that Dynamics seemed to have been vindicated in its opposition to CTS's acquisitions by CTS's having written off the largest one as a loss, CTS's management could not be so confident that a takeover by Dynamics would reduce shareholder wealth. The friction between the companies required, if anything, more than the usual amount of care by CTS's board of directors in evaluating the proposal, to make sure that personal feelings would not be allowed to interfere with the board's fiduciary obligations.

All this might be of little moment if the particular poison pill which Smith Barney sold CTS's board were a plausible measure for maximizing shareholder wealth, for as we have said we have no desire to force boards of directors into a judicialized mode of proceeding and we recognize the time pressure under which the board was operating, with the stockholders' meeting scheduled for April 24 and the starting gun for selling stock to Dynamics in response to its tender offer due to go off only 28 days after the announcement of the offer. So let us look at the terms of the poison pill. As soon as one shareholder had 15 percent or more of CTS's stock, all the others would get the right to buy a securities package, consisting of stock and debenture, for 25 percent of the then market price of the package. The right would thus be triggered if the tender offer went through, and presumably would be exercised since the terms are so advantageous to the shareholder (illustrating our earlier point that poison pills discourage acceptance of tender offers). Assuming that all rights were exercised, the effect of the additional shares that would be issued to the existing shareholders would be to reduce Dynamics' holdings from 27.5 percent of CTS's common stock to 20.7 percent. Not only would this reduce Dynamics' voting power in the election for the board of directors but it would inflict a substantial capital loss on Dynamics. CTS would be worth no more just because, the "poison pill" having taken effect, the company had more shares outstanding than it did previously. Therefore 20.7 percent of the company would be worth less than the 27.5 percent Dynamics thought it was getting: $24 million less, assuming a $40 price for CTS shares.

In addition, the total amount of debentures to be issued to the shareholders as part of the poison pill would burden CTS with a new, long-term fixed debt of $80 million at a high interest rate (13 percent), further reducing the value of Dynamics' holdings. This large debt, for which CTS would not receive commensurate value in exchange, would reduce CTS's net profits. It could, indeed, imperil its financial health very seriously; for the taking on of such a large debt would entitle some existing creditors of CTS to treat CTS as having thereby defaulted on their loans to it, and these creditors would therefore be entitled to call the loans--and the whole house of cards might collapse. At the very least, the new debt would increase the volatility of CTS's earnings by increasing the fraction of fixed costs in the company's financial structure. This in turn would reduce the value of the stock quite apart from the effect of lower earnings, simply because most investors are risk averse. See, e.g., Lorie & Hamilton, The Stock Market: Theories and Evidence, chs. 11-12 (1973).

We do not want to paint too bleak a picture. Unless the debentures so weighed down CTS with debt that they forced the company into bankruptcy, the shareholders--other than Dynamics, of course--would be getting something of value out of the transaction: the debentures. Their investment in CTS would be converted from an all-equity investment to a package consisting of a riskier equity investment and a debt investment. Indeed, putting aside the effect on Dynamics, the immediate effect of issuing the debentures would just be to raise CTS's debt-equity ratio; and if the ratio was too low before, the shareholders might be made better off. But there is no indication that it was too low before; and if a company's debt-equity ratio is too high, the risk of bankruptcy may become very great.

All this Sturm und Drang seems a high price to pay for fending off a change of corporate control that may, for all that appears, benefit the shareholders greatly, though it will be a humiliation to the present officers and directors. It is defended as necessary to protect minority shareholders from a disadvantageous "back-end" transaction. But even after Dynamics obtains 27.5 percent of CTS's stock, it will not be a majority shareholder, and will therefore not be able to squeeze out the remaining shareholders. To be able to do that it will have to buy up another 22.5-plus percent of the shares. CTS's poison pill is thus to be administered prematurely. If the rationale is to protect minority shareholders, it should be triggered by a transaction that creates a majority shareholder or that attempts to squeeze out the minority shareholders, and it should give the minority the same price per share as the majority--not a higher price calculated to kill off the tender, indeed to kill off any tender. CTS's shareholder rights agreement is triggered much earlier, and at a higher price. It effectively precludes a hostile takeover, and thus allows management to take the shareholders hostage. To buy CTS, you must buy out its management.

CTS argues that once Dynamics controls the board of directors and hence the proxy machinery, it will be able to gull the remaining shareholders into selling their shares to it for too low a price. Setting aside the legal remedies against abuse of the proxy machinery, CTS's argument if correct underscores the importance of not impeding tender offers too much, since the premise of the argument is that management cannot be trusted to protect the interests of the shareholders.

We conclude that the poison pill was properly enjoined, and move on to the cluster of issues concerning the distinct takeover obstacle erected by the Indiana control-share acquisition statute, which if valid would thwart Dynamics' tender offer. The Attorney General of Indiana argues that the district judge violated 28 U.S.C. Sec. 2403(b), which provides that in a federal court action in which the constitutionality of a state statute "affecting the public interest is drawn in question, the court shall certify such fact to the attorney general of the State, and shall permit the State to intervene for presentation of evidence, if evidence is otherwise admissible in the case, and for argument on the question of constitutionality." On April 9 the district judge, who had not yet certified to the state's attorney general that the constitutionality of the Indiana takeover statute was being challenged, held that the statute violated the supremacy clause, because in conflict with the Williams Act. There is a question whether section 2403(b) is meant to apply to cases where the only constitutional challenge is under the supremacy clause, that is, where the state statute is contended to be in conflict with a federal statute. The argument against requiring certification in such a case is that preemption may leave the state statute in force in most of its domain, as in our recent decision in National Metalcrafters v. McNeil, 784 F.2d 817, 828-29 (7th Cir.1986). Whatever the actual force of the argument, the State of Indiana appears to concede the point, for it makes no complaint about the fact that the challenge based on the supremacy clause was not certified to it. On April 16, however, the district judge issued another order, this one holding that the takeover statute also violated the commerce clause; there is no dispute that certification is required in such a case. She did certify--but on April 16, and with oral argument in this court scheduled for April 23 we gave the attorney general only till April 19 to file a brief in this court. He argues that this was not enough time and that at a minimum the April 16 order should be vacated.

Section 2403(b) does not say when the certification must be issued. The suit was not filed till March 10, the constitutional issues did not surface till the complaint was amended on March 31, and ordinarily a delay of less than three weeks in certifying the existence of a constitutional challenge would not be undue. But this was not an ordinary case; and to issue the certification simultaneously with the decision does not comport with the spirit of a statute designed to give the attorney general a fair opportunity to argue and if necessary present evidence (only argument is at issue here) to save a state statute. Argument after decision--argument that asks the district court to reconsider its decision or us to reverse it--is an inferior substitute for argument before the district court's decision, which is what the statute obviously envisages though it does not require it in so many words.

Some cases, illustrated by Bridges v. Phillips Petroleum Co., 733 F.2d 1153, 1156 n. 7 (5th Cir.1984) (per curiam), seem to regard belated certification (e.g., at the appellate level) as adequate compliance nonetheless. A closely related line of cases in the Second Circuit deems it adequate compliance to allow the state to file a brief on appeal without worrying about any formalities of certification. See, e.g., Doyle v. Suffolk County, 786 F.2d 523, 526 (2d Cir.1986). Still other cases, illustrated by Merrill v. Town of Addison, 763 F.2d 80, 82-83 (2d Cir.1985), and by our own decision in Kealey Pharmacy & Home Care Services, Inc. v. Walgreen Co., 761 F.2d 345, 350 n. 8 (7th Cir.1985), emphasize that failure to comply with section 2403(b) does not deprive the district court of jurisdiction, so that the district court's decision need not be vacated just because certification was belated. In none of the cases we have found, however, did the appellate court hold the challenged state statute unconstitutional, so the issue of the district court's failure to comply with section 2403(b) was not momentous.

In any event section 2403(b) does not prescribe a sanction for its violation; that is left to the courts. And in this case it seems to us the proper sanction (assuming the statute really was violated) is no sanction. The violation was inadvertent, reflecting the very rapid course of the litigation in the district court. The prejudice to the state seems to be nil, for it has yet even to suggest what arguments it might make that counsel for CTS has not made, and it does not want to present any evidence. And the state's apparent lack of interest in challenging the district judge's conclusion that the statute violates the supremacy clause is inconsistent with its passionate interest in challenging her conclusion that the statute also violates the commerce clause. As we said earlier, a finding of preemption may merely curtail the application of the statute--but not in this case. The district judge held that the only provision of the statute that is in issue is invalid in its entirety under the supremacy clause. This consequence the state contemplates with apparent equanimity; why should it insist on timely certification of a challenge that has the same effect--namely, lethal?

Two other threshold challenges to the district court's constitutional rulings need not detain us for long. The state argues that venue was improper in the Northern District of Illinois; the case should have been brought in Indiana. But objections to venue can be waived, and were waived here by CTS when it failed to follow up its initial objection to laying venue in Illinois. CTS argues that the district court should have abstained in favor of the state courts of Indiana, which might have given a saving construction to the Indiana statute. See Railroad Comm'n v. Pullman Co., 312 U.S. 496, 61 S. Ct. 643, 85 L. Ed. 971 (1941); Waldron v. McAtee, 723 F.2d 1348 (7th Cir.1983). Lack of time is one objection to this course, but the decisive objection is that although there is an issue of state law in this case that affects the constitutional question--whether the Indiana statute is limited to cases where the target corporation is incorporated in Indiana--we have resolved that issue in favor of the state, for we agree that the statute is so limited.

We come then to the merits of the constitutional issues. The issue under the supremacy clause is, as we have said, whether Indiana's control-share acquisition statute is preempted by the Williams Act, 15 U.S.C. Secs. 78m(d)-(e), 78n(d)-(f). The Indiana statute allows a corporation that has more than 100 shareholders and is incorporated in Indiana, and that has its principal place of business, principal office, or substantial assets there, and ten percent of whose shareholders, or 10,000 of whose shareholders, or owners of ten percent of whose shares, reside there, to elect to be covered by the "control share acquisition" provisions of the act; CTS elected. A control share acquisition is defined as an acquisition that along with any previous acquisitions gives the acquirer at least 20 percent of the voting stock of the covered firm. If the acquirer files a statement containing specified information about his intentions and financial capacity (similar to the information required by the Williams Act), management has 50 days within which to hold a special shareholders' meeting to consider whether the acquirer shall have voting rights. A decision in favor of awarding voting rights requires a majority both of all shares and of all "disinterested" shares, that is to say, shares of all shareholders except the acquirer and the officers and inside directors of the corporation. Without these two majorities, the acquirer's shares remain nonvoting shares. If the acquirer does not request a special meeting the question of voting rights will be taken up at the next regularly scheduled stockholders' meeting. And if no acquiring-person's statement is filed the corporation can redeem the acquirer's shares "at the fair value thereof pursuant to the procedures adopted by the corporation."

Cleverly drafted though the statute is to skirt judicial holdings that forbid states to delay tender offers beyond the period required by the Williams Act, the cleverness is fairly transparent. The effect of the Indiana statute is both to impose a 50-day delay on tender offers at the option of the target firm and to make it far more difficult for tender offers to succeed even if delay is not an impediment. The offeror dare not accept the tendered shares till the stockholders' meeting is held, since if he loses the vote on voting rights he will end up with nonvoting shares and will not be able to control the corporation--the main purpose of most tender offers. So he must hold the tender offer open for 50 days, rather than the 28 days required (on average) by the SEC's regulations under the Williams Act. See 17 C.F.R. Sec. 240.14e-1(a) (20 business days). And he can have no great confidence in being able to win the vote on voting rights, since he cannot vote his own shares. Neither can the officers and inside directors, but their aggregate shareholdings will often be small. Not filing an acquiring-person statement is an unattractive option for the tender offeror; it exposes him to a forced redemption of his shares at a price determined under procedures chosen by a hostile management.

In Edgar v. MITE Corp., 457 U.S. 624, 102 S. Ct. 2629, 73 L. Ed. 2d 269 (1982), which affirmed a decision by this court, three Supreme Court Justices opined that Illinois' takeover statute violated the supremacy clause, as we had held, see MITE Corp. v. Dixon, 633 F.2d 486, 490-99 (7th Cir.1980). The Illinois statute, so far as relevant here, required an offeror who tendered for more than 5 percent of a corporation's stock to issue a statement of intentions 20 days before the offer itself could be made. This was in addition to the post-offer waiting period required by the Williams Act. The Illinois statute also required the Secretary of State of Illinois to hold a hearing on the substantive fairness of the offer during the initial 20-day period if requested by a majority of the target's outside directors, or by Illinois owners of 10 percent of the target's stock; and empowered him to disallow the offer if he found that it was substantively unfair.

Most courts have agreed that the Williams Act strikes a balance between target management and tender offeror that the states may not upset. See, e.g., Martin-Marietta Corp. v. Bendix Corp., 690 F.2d 558, 565-66 (6th Cir.1982); National City Lines, Inc. v. LLC Corp., 687 F.2d 1122, 1128-33 (8th Cir.1982). Even Agency Rent-A-Car, Inc. v. Connolly, 686 F.2d 1029, 1035-39 (1st Cir.1982), while skeptical of the approach, concedes some preemptive force to the Williams Act--probably enough to preempt the Indiana statute. See id. at 1038-39. And CTS does not ask us to reexamine our analysis in MITE.

Ordinarily when Congress passes a statute punishing some supposedly unfair or unjust practice such as monopolization or misrepresentation, the states are free to add on their own penalties. See id. at 1037. If, therefore, the Williams Act is, as a critical literature forcefully argues, see, e.g., Fischel, Efficient Capital Market Theory, the Market for Corporate Control, and the Regulation of Cash Tender Offers, 57 Tex.L.Rev. 1 (1978), an anti-takeover statute, expressing a view, however benighted, that hostile takeovers are bad, one would be hard pressed to argue that the Act forbids the states to pass fiercer anti-takeover statutes. See Note, Securities Law and the Constitution: State Tender Offer Statutes Reconsidered, 88 Yale L.J. 510, 512-25 (1979). If intent can be inferred from effect, the characterization of the Williams Act as an anti-takeover statute is reinforced. See Jarrell & Bradley,The Economic Effects of Federal and State Regulations of Cash Tender Offers, 23 J.Law & Econ. 371 (1980). However, both we and the Supreme Court plurality in MITE pointed out that as the bill that ultimately became the Williams Act wended its way through the labyrinthine processes of Congress, support for hostile takeovers emerged and the original thrust of the bill was somewhat blunted. The importance of the tender offer in disciplining corporate management and shifting corporate assets into the hands of those who can manage them best was remarked, and the legislators' concern narrowed down to making sure that shareholders of target corporations would have enough information to make intelligent decisions. The bill that emerged essentially just requires the filing of a public statement of intentions and giving the shareholders enough time to digest the information in the statement as well as such counterinformation as management may care to supply. Although the ultimate balance that was struck may well incline--too far for some tastes--in favor of target management, that is not inconsistent with an inference that Congress would not have wanted the states to tip the balance any further.

Of course it is a big leap from saying that the Williams Act does not itself exhibit much hostility to tender offers to saying that it implicitly forbids states to adopt more hostile regulations, but this leap was taken by the Supreme Court plurality and us in MITE and by every court to consider the question since. The reasoning is that Congress in the Williams Act (as in the federal labor laws) struck a delicate balance between the contending factions in the takeover controversy and wanted its balance to mean something and not be undone by the states. There is some support for this view in the legislative history, notably in the statement in the House Report that "The bill avoids tipping the balance of regulation either in favor of management or in favor of the person making the takeover bid." H.R.Rep. No. 1711, 90th Cong., 2d Sess. 4 (1968) U.S.Code Cong. & Admin.News 1968, p. 2811. Admittedly, if this sentence is read with emphasis on the first two words, it ceases to express much if any policy with regard to the permissible scope of state regulation. But whatever doubts of the Williams' Act preemptive intent we might entertain as an original matter are stilled by the weight of precedent. See, besides the other cases we have cited, Schreiber v. Burlington Northern, Inc., --- U.S. ----, 105 S. Ct. 2458, 2463, 86 L. Ed. 2d 1 (1985), and Piper v. Chris-Craft Industries, Inc., 430 U.S. 1, 29, 97 S. Ct. 926, 943, 51 L. Ed. 2d 124 (1977).

If the general approach is correct, its application to this case is straightforward. The Indiana statute upsets the balance struck by the Williams Act. Whether it does so more than the Illinois statute struck down in MITE is hard to say. The statutes are incommensurable. The Illinois statute both imposed delay and put the acquirer at the mercy of the Illinois Secretary of State; the Indiana statute imposes slightly greater delay but puts the acquirer at the tenderer mercies of the "disinterested" shareholders. If we had to guess we would guess that the Indiana statute is less inimical to the tender offer, but that is unimportant. The Indiana statute is a lethal dose; the fact that the Illinois statute may have been two or three lethal doses has no practical significance. Very few tender offers could run the gauntlet that Indiana has set up. In any event, if the Williams Act is to be taken as a congressional determination that a month (roughly) is enough time to force a tender offer to be kept open, 50 days is too much; and 50 days is the minimum under the Indiana act if the target corporation so chooses.

The question of the Indiana act's validity under the commerce clause may seem doubly academic: if the statute is unenforceable by reason of the supremacy clause, it hardly matters, at least for this case, whether it is also unenforceable by reason of the commerce clause; and the commerce clause provides an independent bar to state action only when Congress has not spoken. The second point will bear elaboration. All that the commerce clause says is that Congress is empowered to regulate interstate and foreign commerce. Art. I, Sec. 8, cl. 3. Although the clause has long been interpreted to authorize the courts to strike down state regulations of interstate commerce that conflict with the clause's presumed purpose of making the nation a common market, provided that Congress has not spoken, see, e.g., Cooley v. Board of Wardens, 53 U.S. (12 How.) 299, 319 (1852), one might think that once Congress had spoken the "dormant" commerce clause would fall out and the only judicial function would be to enforce the congressional enactment. Congress has spoken to the interstate traffic in corporate control by passing the Williams Act. If the Act is intended to limit state anti-takeover statutes, that is the authoritative expression of congressional desire, and likewise if the Act is not intended to limit them.

But there is a third possibility: that the Act is intended not to affect their legality one way or the other. In MITE, and in the cases that follow it, the courts have separated the supremacy and commerce clause issues--have assumed that the commerce clause retains an independent force despite the enactment of the Williams Act. We shall do the same, if only because of lingering doubt that the Act really was intended to limit state anti-takeover statutes; there is no indication that it was intended to insulate such statutes from complaints that they unduly burden interstate commerce.

The commerce clause has been interpreted to limit the power of the states to impose burdens on people living in other states. The limitation is not absolute; for example, a state may apply its health and safety regulations to imported goods even though the cost of compliance will be borne in part by people in other states, the suppliers of the goods. It is all a matter of balancing the benefit to the state's residents against the burden to out-of-staters. In this case, as in Edgar v. MITE Corp., supra, 457 U.S. 640-46, 102 S. Ct. 2639, where a majority of the Supreme Court held that Illinois' anti-takeover statute violated the commerce clause, the balance inclines heavily against the out-of-staters; "the burden imposed on [interstate] commerce is clearly excessive in relation to the putative local benefits." Pike v. Bruce Church, Inc., 397 U.S. 137, 142, 90 S. Ct. 844, 847, 25 L. Ed. 2d 174 (1970); see Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 471, 101 S. Ct. 715, 728, 66 L. Ed. 2d 659 (1981), and with specific reference to state statutes regulating takeovers, Mesa Petroleum Co. v. Cities Service Co., 715 F.2d 1425, 1429-31 (10th Cir.1983), and cases cited there, and Levmore, Interstate Exploitation and Judicial Intervention, 69 Va.L.Rev. 563, 619-24 (1983). We have not been told the geographical distribution of Dynamics' and CTS's shareholders but we know that CTS need have only a small fraction of its shareholders resident in Indiana to take advantage of the Indiana statute and we can assume that the vast majority of both its shareholders and Dynamics' shareholders are nonresidents. The statute gravely impairs Dynamics' ability to do business with any of CTS's shareholders. Indiana has no interest in protecting residents of Connecticut from being stampeded to tender their shares to Dynamics at $43. See Edgar v. MITE Corp., supra, 457 U.S. at 642-43, 102 S. Ct. 2640 -41. Unlike a state's blue sky law the Indiana statute is calculated to impede transactions between residents of other states. For the sake of trivial or even negative benefits to its residents Indiana is depriving nonresidents of the valued opportunity to accept tender offers from other nonresidents.

Whether an appreciable number of Indiana shareholders or other Indianans will benefit at all from the statute may be doubted; the only beneficiaries in this case may be the officers and directors of CTS, some of whom may not even be Indianans. No evidence has been presented that a takeover by Dynamics might reduce the value of CTS or lead to a shift of assets or employment from Indiana--and if it did lead to such a shift, this might further condemn, rather than save, the statute. The commerce clause does not allow states to prevent corporations from moving assets and employees to other states. But whether or not an anti-takeover statute is vulnerable to challenge under the commerce clause if it impedes mobility of corporate assets, it is highly vulnerable if it impedes the important commerce in corporate control. Even if a corporation's tangible assets are immovable, the efficiency with which they are employed and the proportions in which the earnings they generate are divided between management and shareholders depends on the market for corporate control--an interstate, indeed international, market that the State of Indiana is not authorized to opt out of, as in effect it has done in this statute. See id. at 643, 102 S.Ct. at 2641.

L.P. Acquisition Co. v. Tyson, 772 F.2d 201, 205-07 (6th Cir.1985), is distinguishable. Because the Williams Act was inapplicable to the tender offer there, the disclosure required by the state statute conferred greater benefits on local residents than the disclosure required by the Indiana statute confers, so the balance described in Pike v. Bruce Church, Inc. inclined more in favor of protection of local interests. In Cardiff Acquisitions, Inc. v. Hatch, 751 F.2d 906, 909-12 (8th Cir.1984), the required disclosure was designed to give state residents information particularly pertinent to the impact that the takeover would have in the state itself, and the court was persuaded that the effect in discouraging takeovers through delay would be slight--as it is not here. Indiana has erected a barrier at once formidable and arbitrary to tender offers whose principal effects if they succeed will be felt outside Indiana.

The Indiana statute is not saved by the "internal affairs" doctrine. That principle of conflict of laws is designed to make sure that the law of only one state shall govern the internal affairs of a corporation or other association. See Edgar v. MITE Corp., supra, 457 U.S. at 645, 102 S. Ct. at 2642; Restatement (Second) of Conflict of Laws Sec. 302 (1971). In this case the state is indeed Indiana; but it does not follow that any law which Indiana adopts that affects the internal affairs of Indiana corporations is thereby insulated from review under the commerce clause. We may assume without having to decide that Indiana has a broad latitude in regulating those affairs, even when the consequence may be to make it harder to take over an Indiana corporation; an example is a law that by requiring cumulative voting for the board of directors makes it difficult to oust the entire existing board at one fell swoop. But in this case the effect on the interstate market in securities and corporate control is direct, intended, and substantial; it is not merely the incidental effect of a general regulation of internal corporate governance. The law in question is an explicit regulation of tender offers; that the mode of regulation involves jiggering with voting rights cannot take it outside the scope of judicial review under the commerce clause. Any other conclusion would invite facile evasions of the clause.

The next issue is whether the judge should have enjoined the tender offer on the ground that if it succeeded, Dynamics and CTS would be in violation of section 8 of the Clayton Act. Section 8 prohibits the director of one company from serving as the director of another if the elimination of competition between the companies, as by a merger between them, would violate any of the antitrust laws. The record of this case contains no persuasive evidence that Dynamics and CTS are in competition, but that hardly matters; the important point is that there is no reason to believe that any director of Dynamics will agree to serve as a director of CTS if by doing so he would violate section 8, which is addressed to the director and not the company. Should it wrest control of the board of CTS from the present management and should the firms be found to be in competition, Dynamics will have no difficulty in finding directors for CTS's board who are not also directors of Dynamics.

The last issue is whether the tender offer should have been enjoined because of a failure to disclose material facts. As the district judge correctly found, most of these facts were not material. One was. The tender-offer materials do not disclose Dynamics' intention to oust the present management of CTS if the tender offer succeeds and enables Dynamics to elect a new board of directors. The district judge held that this omission was cured by the distribution to all shareholders, two weeks before the end of the tender-offer period, of Dynamics' proxy materials, in which it urges the shareholders to elect Dynamics' slate of directors. CTS argues that this was not good enough, because some shareholders may not have paid attention to the proxy solicitation. But such shareholders might have paid equally little attention to the same disclosure in the tender offer. In truth Dynamics' desire to oust the present board was broadcast loudly and widely.

Even if exclusion from the tender offer could not be cured by inclusion in the proxy materials--an arcane question of federal securities law on which CTS's briefs cast little light--it would not follow that enjoining the tender offer was a proper remedy. The fashioning of remedies is largely within the discretion of the district judge and that discretion was not abused in this case.

AFFIRMED.