Summary
finding that damages could not adequately compensate the plaintiffs in part because the injury to their reputations and goodwill were “extremely difficult to calculate”
Summary of this case from Sonic Indus. v. Olympic Cascade Drive Ins LLCOpinion
No. 85-2203.
Argued September 23, 1985.
Decided January 2, 1986.
John Powers Crowley, Cotsirilos Crowley, Ltd., Chicago, Ill., for defendants-appellants.
Irving B. Levinson, Rivkin, Radler, Dunne Bayh, Chicago, Ill., for plaintiffs-appellees.
Appeal from the United States District Court for the Northern District of Illinois.
Before POSNER, Circuit Judge, and SWYGERT and PELL, Senior Circuit Judges.
A supplier terminated a distributor, who sued for breach of contract and got a preliminary injunction. The supplier has appealed under 28 U.S.C. § 1292(a)(1). The appeal raises issues of procedure and contract law.
The supplier, Hospital Products (as we shall call the affiliated corporations that are the defendants), a small firm now undergoing reorganization in bankruptcy, is one of the world's two principal manufacturers of "reusable surgical stapling systems for internal surgical procedures" ("surgical stapling systems," for short). The terminated distributor, American Hospital Supply Corporation, the world's largest distributor of medical and surgical supplies, in 1982 became the exclusive distributor in the United States of Hospital Products' surgical stapling systems. The contract of distribution was for three years initially, but provided that it would be renewed automatically for successive one year periods (to a limit of ten years) unless American Hospital Supply notified Hospital Products at least 90 days before the three years were up (or any successive one-year period for which the contract had been renewed) that it wanted to terminate the contract; and this meant, by June 3, 1985.
On that day Hospital Products hand-delivered a letter to American Hospital Supply demanding to know whether it intended to renew the contract and reminding it that if it failed to respond by the end of the day this would mean that the contract had been renewed. American Hospital Supply responded the same day in a letter which pointed out that since it wasn't terminating, the contract was, indeed, renewed. But on the next day Hospital Products announced that it was going to treat the contract as having been terminated, and on June 7 it sent a telegram to American Hospital Supply's dealers informing them that effective June 3 American Hospital Supply was "no longer the authorized distributor of [Hospital Products'] stapling products."
American Hospital Supply forthwith brought this diversity breach of contract suit against Hospital Products and moved for a preliminary injunction, which was granted on July 8 after an evidentiary hearing. The injunction forbids Hospital Products to take any action in derogation of American Hospital Supply's contract rights so long as the injunction is in force (i.e., pending the outcome of the trial). It also requires Hospital Products to notify American Hospital Supply's dealers that American Hospital Supply is still Hospital Products' authorized distributor, and this has been done. Hospital Products counterclaimed, alleging breach of contract, fraud, and unfair competition.
Two months after the entry of the injunction, Hospital Products, which had been in parlous financial state even before this litigation began, filed a petition for bankruptcy under Chapter 11 of the Bankruptcy Act (reorganization). The bankruptcy court suspended the automatic stay of litigation against a bankrupt, see 11 U.S.C. § 362(a), (d); 2 Collier on Bankruptcy ¶¶ 362.04[1], 362.07 (15th ed., King ed., 1985), to permit this appeal. Hospital Products moved the bankruptcy court to disaffirm (i.e., cancel) the renewed contract that went into effect on September 1, 1985, upon the expiration of the original contract. Its ground was that the renewed contract was still executory when Hospital Products declared bankruptcy, and hence was subject to disaffirmance under 11 U.S.C. § 365. See In re Minges, 602 F.2d 38, 41 (2d Cir. 1979). The bankruptcy court has not yet acted on the motion. Hospital Products has not asked the district court to dissolve the preliminary injunction on the basis of events, notably the bankruptcy, since the injunction was entered.
A district judge asked to decide whether to grant or deny a preliminary injunction must choose the course of action that will minimize the costs of being mistaken. Because he is forced to act on an incomplete record, the danger of a mistake is substantial. And a mistake can be costly. If the judge grants the preliminary injunction to a plaintiff who it later turns out is not entitled to any judicial relief — whose legal rights have not been violated — the judge commits a mistake whose gravity is measured by the irreparable harm, if any, that the injunction causes to the defendant while it is in effect. If the judge denies the preliminary injunction to a plaintiff who it later turns out is entitled to judicial relief, the judge commits a mistake whose gravity is measured by the irreparable harm, if any, that the denial of the preliminary injunction does to the plaintiff.
These mistakes can be compared, and the one likely to be less costly can be selected, with the help of a simple formula: grant the preliminary injunction if but only if P X H[p] (1 — P) X H[d], or, in words, only if the harm to the plaintiff if the injunction is denied, multiplied by the probability that the denial would be an error (that the plaintiff, in other words, will win at trial), exceeds the harm to the defendant if the injunction is granted, multiplied by the probability that granting the injunction would be an error. That probability is simply one minus the probability that the plaintiff will win at trial; for if the plaintiff has, say, a 40 percent chance of winning, the defendant must have a 60 percent chance of winning (1.00 -.40 = .60). The left-hand side of the formula is simply the probability of an erroneous denial weighted by the cost of denial to the plaintiff, and the right-hand side simply the probability of an erroneous grant weighted by the cost of grant to the defendant.
This formula, a procedural counterpart to Judge Learned Hand's famous negligence formula, see United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir. 1947); United States Fidelity Guaranty Co. v. Jadranska Slobodna Plovidba, 683 F.2d 1022, 1026 (7th Cir. 1982), is not offered as a new legal standard; it is intended not to force analysis into a quantitative straitjacket but to assist analysis by presenting succinctly the factors that the court must consider in making its decision and by articulating the relationship among the factors. It is actually just a distillation of the familiar four (sometimes five) factor test that courts use in deciding whether to grant a preliminary injunction. The court asks whether the plaintiff will be irreparably harmed if the preliminary injunction is denied (sometimes also whether the plaintiff has an adequate remedy at law), whether the harm to the plaintiff if the preliminary injunction is denied will exceed the harm to the defendant if it is granted, whether the plaintiff is reasonably likely to prevail at trial, and whether the public interest will be affected by granting or denying the injunction (i.e., whether third parties will be harmed — and these harms can then be added to H[p] or H[d] as the case may be). See, e.g., Palmer v. City of Chicago, 755 F.2d 560, 576 (7th Cir. 1985). The court undertakes these inquiries to help it figure out whether granting the injunction would be the error-minimizing course of action, which depends on the probability that the plaintiff is in the right and on the costs to the plaintiff, the defendant, or others of granting or denying the injunction. All this is explained at length in Roland Machinery Co. v. Dresser Industries, Inc., 749 F.2d 380, 382-88 (7th Cir. 1984), where a panel of this court applied the verbal counterpart to our algebraic formula, as did a different panel in Maxim's Ltd. v. Badonsky, 772 F.2d 388, 391 (7th Cir. 1985). See also Leubsdorf, The Standard for Preliminary Injunctions, 91 Harv.L.Rev. 525 (1978). The formula is new; the analysis it capsulizes is standard.
The formula does not depend on the legal basis of the plaintiff's claim, whether it is antitrust law ( Roland) or trademark law ( Badonsky) or, as here, the common law of contract, although the nature of the right asserted by the plaintiff may affect the weighting of the harms, see, e.g., Shondel v. McDermott, 775 F.2d 859, 866-67 (7th Cir. 1985). So may the nature of the permanent remedy to which the plaintiff would be entitled if he prevailed at trial. For example, prevailing parties in breach of contract cases normally are not awarded specific performance, that is, a mandatory injunction to perform. Since many breaches of contract are involuntary, implying that performance would be very costly, routinely ordering specific performance would create situations where the defendant was forced to bargain desperately to buy his way out of the injunction. The high bargaining costs that would result are a deadweight cost of equitable relief. To the extent that those costs attend a preliminary injunction, they are of course relevant to the decision whether to issue such an injunction. But the formula takes account of this; the case we have described would be one where the harm to the defendant from granting the injunction would be very great. Thus the fact that a plaintiff might have no hope of getting specific performance ordered at the conclusion of the trial need not prevent him from obtaining a preliminary injunction. Cf. Roland v. Dresser Industries, Inc., supra, 749 F.2d at 386. The premise of the preliminary injunction is that the remedy available at the end of trial will not make the plaintiff whole; and, in a sense, the more limited that remedy, the stronger the argument for a preliminary injunction — provided the remedy is not limited for reasons that would make a preliminary injunction equally inappropriate.
As explained in Roland, the scope of judicial review of a district judge's decision to grant or deny a preliminary injunction is limited. See 749 F.2d at 384-85, 388-91. The usual formulation is that the decision will be reversed only if it is found to be an "abuse of discretion." Unfortunately this phrase covers a family of review standards rather than a single standard, and a family whose members differ greatly in the actual stringency of review. See, e.g., id. at 388-91; Metlyn Realty Corp. v. Esmark, Inc., 763 F.2d 826, 831-32 (7th Cir. 1985); Friendly, Indiscretion About Discretion, 31 Emory L.J. 747 (1982). For example, when we review an order granting or denying a preliminary injunction, we do not do so with as much deference (virtually complete deference) as when reviewing a criminal sentence that is within legal limits but is challenged as too harsh; we do "not simply engage in a perfunctory rubber-stamping of the district court's decision." Olin Water Services v. Midland Research Laboratories, Inc., 774 F.2d 303, 307 n. 7 (8th Cir. 1985). But we do give that decision substantial deference, bearing in mind that the district judge had to act in haste, that he had to balance factors which, though they can be related in a neat formula, usually cannot be quantified, and that in dealing with the parties and their witnesses and counsel in the hectic atmosphere of a preliminary-injunction proceeding the judge may have developed a feel for the facts and the equities that remote appellate judges cannot obtain from a transcript. To reverse an order granting or denying a preliminary injunction, therefore, it is not enough that we think we would have acted differently in the district judge's shoes; we must have a strong conviction that he exceeded the permissible bounds of judgment. And although recognizing that the order granting such an injunction must set forth the judge's reasons, see Fed.R.Civ.P. 65(d), we also recognize that the haste with which the judge must act precludes as full a statement of reasons as could reasonably be demanded of a final decision.
We have now to apply these precepts, and we begin with the balance of harms. Hospital Products points out, irrelevantly as it seems to us, that American Hospital Supply did not prove more than a handful of lost sales as a result of the mailgram of June 7 which announced the termination of American Hospital Supply's distributorship to its dealers. The reason is simply that American Hospital Supply promptly asked for and promptly received a preliminary injunction to head off any losses by restoring its distributorship. The question is not whether there was an actual loss but whether there was an impending loss that the preliminary injunction prevented, how great it was, and whether it could have been made up by a judgment for damages after trial.
Although American Hospital Supply was able to replace Hospital Products' line of surgical stapling systems in the interval between the mailgram of June 7 and the entry of the preliminary injunction a month later, the mailgram, unless retracted as ordered by the injunction, might have impaired American Hospital Supply's goodwill, a factor emphasized in other cases where terminated dealers sought preliminary injunctions, see, e.g., Menominee Rubber Co. v. Gould, Inc., 657 F.2d 164, 167 (7th Cir. 1981). The suddenness of the termination and the urgent mode of announcement might have made the dealers think that American Hospital Supply must have engaged in unethical or unreasonable conduct. We do not put much weight on this point, however. It is speculative, and any harm may have been cured by the retraction, in which event the harm could not support the rest of the injunction.
But in addition, on June 7 American Hospital Supply was holding a large unsold inventory of Hospital Products' surgical stapling systems. To help Hospital Products overcome serious financial problems, American Hospital Supply had advanced it millions of dollars — part in loans, part by buying more of the product than it needed and keeping the excess in inventory. The mailgram jeopardized its investment because dealers might be reluctant to buy Hospital Products' goods from American Hospital Supply, not wanting to become enmeshed in a legal dispute between it and its supplier or perhaps even fearing that there might be some defect in the particular items being sold by American Hospital Supply. True, its investment would probably not be totally wiped out. American Hospital Supply might be able to sell the product to its dealers immediately at a sharp mark-down or with extra warranties or with promises of indemnity; or if it waited till it prevailed on the merits against Hospital Products, at no mark-down, in which event its loss would be just the interest and storage costs of keeping the product in inventory longer than expected. (If it didn't prevail, its losses would not be recoverable anyway.) No effort was made to quantify the loss to American Hospital Supply caused by the alleged breach of contract. But in the nature of things reliable estimation may have been infeasible; and since the estimates of the size of the unsold inventory range from $10 million to almost $30 million, probably the threatened loss was substantial.
What made the loss (whatever exactly it was in dollar terms) irreparable was not only or mainly the difficulty that American Hospital Supply might encounter down the road in quantifying its loss, although this difficulty has been stressed in other cases where a distributor sought a preliminary injunction in part to protect goodwill, see, e.g., Blackwelder Furniture Co. of Statesville, Inc. v. Seilig Mfg. Co., 550 F.2d 189, 197 (4th Cir. 1977); it was Hospital Products' insolvency, which was apparent when the district judge granted the preliminary injunction, even though the firm had not yet declared bankruptcy. The victim of a breach of contract, certainly one committed before the contract breaker declares bankruptcy, is just another general creditor, see In re Minges, supra, 602 F.2d at 41, and it is well known that general creditors fare poorly in most bankruptcy proceedings. This is true whether the proceeding ends in liquidation or in reorganization, though as a matter of fact most bankruptcies that start in reorganization nonetheless end in liquidation. Maybe, as Hospital Products argues, if only it could get the injunction lifted it would soon recover its solvency, even to the point of being able to pay any damages that American Hospital Supply is awarded in the trial of this case. But if Hospital Products really believes that, it ought to have gone to the district judge and asked him to dissolve the injunction, and it did not — an omission that makes us profoundly skeptical of Hospital Products' claim that the grant of the injunction precipitated its declaration of bankruptcy.
Although a defendant's insolvency is a standard ground for concluding that a plaintiff's harm if the preliminary injunction is denied will not be cured by an award of damages at the end of the trial, see Roland Machinery Co. v. Dresser Industries, Inc., supra, 749 F.2d at 386; Signode Corp. v. Weld-Loc Systems, Inc., 700 F.2d 1108, 1111 (7th Cir. 1983), we must ask whether the cases that accept it give enough weight to the policies of bankruptcy law. To use the defendant's insolvency as a reason for granting the plaintiff an injunction now rather than making him wait for damages till the end of the trial may seem to give the plaintiff a preference in the distribution of the defendant's assets and thus impose harms on third parties, the defendant's other creditors. The responsibility for assessing those harms, however, has been placed in the bankruptcy court (which in this case happens to be in Connecticut) rather than the court asked to grant a preliminary injunction. Ordinarily the bankruptcy court can, just by not granting a motion to lift the automatic stay of litigation against the bankrupt, prevent a creditor from seeking the "preference" represented by the grant of a preliminary injunction in the creditor's favor. That course was ruled out here by the fact that the injunction was granted before bankruptcy was declared; it was the bankrupt that moved to lift the automatic stay, so that it could appeal. The bankruptcy court can still protect the creditors, however, by disaffirming the renewed contract. We assume that disaffirmance (at least if upheld on appeal) would cause the district judge to dissolve the preliminary injunction, as it would show that American Hospital Supply had no legal basis for complaining about the termination of the contract after it was renewed but before it was even partially executed.
We conclude that there was a threat of irreparable harm to the plaintiff; and although the dollar amount of that harm is not known with any precision and we hesitate to call it great, it seems substantial. We must next consider the irreparable harm to the defendant from the injunction. The district judge found there would be none, because American Hospital Supply — which has billions of dollars in sales and earns substantial profits — will be good for any money judgment that Hospital Products may obtain on its counterclaim. This ground is not entirely satisfactory. There is a difference between the damages caused Hospital Products by the breach of contract or other (alleged) misconduct committed by American Hospital Supply before this suit began, damages which we may assume American Hospital Supply is good for, and damages caused Hospital Products by the preliminary injunction — that is, by an order fastening Hospital Products to American Hospital Supply for another year. Those are not the damages that Hospital Products seeks to recover on its counterclaim, and while there may be considerable overlap between the injunction damages and the counterclaim damages, we cannot say that the overlap is complete.
But we may not overlook the bond that a plaintiff who obtains a preliminary injunction is required to post (see Fed.R.Civ.P. 65(c)) and that this plaintiff did post, in an amount, $5 million, whose adequacy Hospital Products does not challenge. Again it can be argued that damages may be hard to prove. The injunction bond merely puts a ceiling on damages — they must still be proved. See Coyne-Delany Co. v. Capital Development Bd., 717 F.2d 385, 391-94 (7th Cir. 1983). But the irreparability that comes from the difficulty of proving damages is not of the same order as that which comes from the uncollectibility of a damage judgment. The award of damages in a case where injury is difficult to measure is as likely to overcompensate as to under compensate the injured party. When a court speaks of damages as being "irreparable" because they are difficult to measure it can mean only that confining the injured party to a remedy of damages creates a risk he may not like (because he is risk averse — the disposition that leads people to buy insurance), even though the upside risk is as large as the downside risk. That is, the plaintiff is as likely to do better than he expects as he is to do worse, but he would lose more utility by losing big than he would gain utility by winning equally big. So there is a harm, but probably not a big one. The specific harms to which the Second Circuit referred in Jack Kahn Music Co. v. Baldwin Piano Organ Co., 604 F.2d 755, 764 (2d Cir. 1979), in explaining why a supplier should not be forced by a preliminary injunction to continue a dealership, are not alleged here.
But what of the fact that an award of damages on the injunction bond, even if it made Hospital Products' creditors whole (and it might not), might not come in time to save the company itself? Many reorganizations in bankruptcy end in liquidation, and Hospital Products' reorganization may end there too, before the case is tried on the merits and therefore before the injunction bond can be enforced; and even if the company is successfully reorganized, the shareholders may be wiped out, and all the stock come into the hands of creditors. But the full financial losses to Hospital Products' shareholders from the bankruptcy are not the correct measure of the harm from granting the injunction. Bankruptcy is not, not intentionally anyway, a device for reducing wealth, but a device for distributing the impact of a business failure over various claimants to the bankrupt's assets. The costs that bankruptcy imposes, as distinct from the costs that the underlying failure imposes, are therefore merely the costs of administering the bankruptcy proceeding. Nevertheless these costs are not negligible. See, e.g., Ang, Chua McConnell, The Administrative Costs of Corporate Bankruptcy: A Note, 37 J.Fin. 219 (1982); Altman, A Further Empirical Investigation of the Bankruptcy Cost Question, 39 J.Fin. 1067 (1984). Although the expenses of administration are paid to someone — to lawyers, accountants, etc. — they are paid for time that would have value elsewhere; and that value is a deadweight loss from bankruptcy, as it cannot be recovered by a payment of damages to the bankrupt estate. A preliminary injunction that will or may precipitate a firm into bankruptcy is therefore a source of costs which ought to be considered in deciding whether to grant such an injunction.
A related and often more significant type of cost, but one whose implications for legal policy are more ambiguous, is the cost of the business failure itself. If the firm's assets would be worth less on the auction block than as part of a going concern, this might seem a powerful argument against granting a preliminary injunction that increased the risk of failure. It is not a complete answer to point out that the creditors will get the benefit of any injunction damages awarded the bankrupt; those damages may not offset the loss of going-concern value. But since business failure provides social as well as private benefits — provides, indeed, the essential discipline of the capitalist system — it seems questionable to give firms that conduct their affairs in an unduly risky or careless manner an advantage in litigation. A more mundane point is that Hospital Products should have asked for a bigger injunction bond if it thought the injunction might play the role of the proverbial nail for want of which the kingdom was lost.
We need penetrate no further into this maze; it is enough that the effect of a preliminary injunction in precipitating insolvency is a factor arguing against the grant of the injunction (how strongly we shall not have to decide). Reflection on this overlooked point might conceivably persuade the district judge to grant a motion to dissolve the preliminary injunction, should such a motion be made, as it can be at any time. See, e.g., Winterland Concessions Co. v. Trela, 735 F.2d 257, 259-60 (7th Cir. 1984). But it does not persuade us that the preliminary injunction was in error. It does not even persuade us that the district judge erred in concluding that the balance of harms inclined in favor of granting the preliminary injunction. Keep in mind that when the district judge acted, Hospital Products had not yet declared bankruptcy. The judge thought the injunction bond might tide the company over the crisis; he also pointed out (and this has great relevance to the issue of American Hospital Supply's chances of winning the case on the merits, of which more presently) that American Hospital Supply had loaned millions of dollars to Hospital Products to prevent the latter from going broke.
We attach no legal significance to the difference in the size of the parties. Although American Hospital Supply is a giant and Hospital Products a pygmy, we cannot think what legal difference that makes. We have taken an oath to do justice to rich and poor alike — and anyway one can't tell merely from the relative size of two corporations what the relative wealth of their shareholders is. Relative size is relevant where, as in Machlett Laboratories, Inc. v. Techny Industries, Inc., 665 F.2d 795, 798 (7th Cir. 1981); Semmes Motors, Inc. v. Ford Motor Co., 429 F.2d 1197, 1205 (2d Cir. 1970) (Friendly, J.), and Roso-Lino Beverage Distributors, Inc. v. Coca-Cola Bottling Co. of New York, Inc., 749 F.2d 124 (2d Cir. 1984) (per curiam), one party, being small and weak, faces bankruptcy if the preliminary injunction is granted or denied; for bankruptcy imposes social costs, as we have seen. See Sperry Int'l Trade, Inc. v. Government of Israel, 670 F.2d 8, 12 (2d Cir. 1982). But we have already considered the issue of bankruptcy risk.
As a matter of fact it could be argued that the relative size of the companies, viewed separately from the issue of bankruptcy costs, strengthens the case for the preliminary injunction. American Hospital Supply might suffer an irrecoverable loss of many millions of dollars if the injunction is denied it; Hospital Products' small size puts a cap on its loss of $5 million if the injunction bond was computed correctly. But we shall not pursue this interesting issue. Even if there were no clear basis for differentiating between the irreparable harms to American Hospital Supply from denying and to Hospital Products from granting a preliminary injunction, in which event those harms would have to be treated as equal, Hospital Products' appeal would fail. If the harms to the plaintiff and the defendant of denying and granting the injunction, respectively, are equal, the injunction must be granted if the plaintiff has a better than 50 percent chance of winning the case, for then P in our preliminary-injunction formula must exceed 1 — P, and therefore P X H[p] must exceed (1 — P) X H[d] from the assumption that H[p] = H[d].
The district judge was persuaded that Hospital Products, not American Hospital Supply, had broken the contract, implying a very high P. He undoubtedly was correct if the contract was renewed on June 3 and in force the next day when Hospital Products announced that the contract was terminated. But we must consider as did he whether American Hospital Supply repudiated the contract before this announcement. That would make this a case of anticipatory breach of contract — American Hospital Supply indicates that it will not perform its obligations under the renewed contract, Hospital Products therefore treats the contract as terminated. Hospital Products points to a letter from American Hospital Supply which threatens to end financial assistance to Hospital Products and call its loans unless Hospital Products agrees to modify the contract in American Hospital Supply's favor. American Hospital Supply responds with much show of reason that it had advanced millions of dollars to Hospital Products beyond anything that it was contractually obligated to do, and therefore had every right to condition the making of new loans or the extension of existing ones on contract concessions. Hospital Products says that no loans were due and what the letter referred to as financial assistance was a euphemism for money due on goods sold and delivered — to which another round of replies asserts that if no loans were due, Hospital Products had nothing to worry about and that American Hospital Supply actually had bought far more surgical stapling products from Hospital Products than it needed or was contractually obligated to buy, so that what looked like payment on the contract really was financial assistance.
This legal badminton is inconclusive; but as nearly as we can determine, the able and experienced district judge who resolved the uncertainty in American Hospital Supply's favor was on solid ground in doing so. Although his discussion of the merits of the parties' contract dispute was not extensive, we remind the reader of our earlier point that a district judge forced to act in haste on an application for a preliminary injunction will not be reversed merely because the reasons he gives are not so full as could properly be demanded of an order granting a permanent injunction or otherwise resolving a full trial on the merits.
Hospital Products was a deeply troubled company to which American Hospital Supply provided financial aid essential in keeping it afloat — not out of altruism, of course, but because it liked Hospital Products' product. Hospital Products acknowledges that American Hospital Supply is its largest creditor. Naturally American Hospital Supply wanted to recoup its investment and it had every right to condition further assistance on whatever concessions it could extract. There is nothing unlawful about offering a benefit to a promisee in exchange for a modification of the contract; the problematic modifications are those not supported by consideration. Thus a buyer who owes money to a financially distressed seller for goods sold and delivered and withholds it in order to force a modification of the contract without fresh consideration is guilty of duress, and the modification will not be enforced. See, e.g., Selmer Co. v. Blakeslee-Midwest Co., 704 F.2d 924 (7th Cir. 1983). But there was no modification here. Whatever concessions American Hospital Supply may have wanted to extract from Hospital Products, on June 3 on Hospital Products' invitation it renewed the contract on its original terms.
If when doing so American Hospital Supply had indicated that it would not honor the terms of the contract but would force Hospital Products into a deeper and deeper hole by refusing to pay for goods bought and received, Hospital Products would be entitled to terminate the contract as a measure of self-help. If you commit a material breach of contract, the other party can walk away from the contract without liability, and can do so as soon as you announce your intentions even if the time for the performance that you have repudiated hasn't arrived. See, e.g., B C Electric, Inc. v. Pullman Bank Trust Co., 96 Ill. App.3d 321, 328, 51 Ill.Dec. 698, 703, 421 N.E.2d 206, 211 (1981); Farnsworth, Contracts § 8.22, at p. 637 (1982); cf. Casio, Inc. v. S.M. R. Co., 755 F.2d 528, 532 (7th Cir. 1985). On this issue the key evidence for Hospital Products is the two letters that American Hospital Supply sent it on June 3. One is the letter in which American Hospital Supply in response to Hospital Products' demand to indicate whether the contract would be renewed made clear that it was not exercising its right of cancellation and therefore the contract was renewed. But the letter also states an "understanding" of the contract by American Hospital Supply that Hospital Products asserts is a clear indication of an intention not to abide by its terms, specifically, not to buy as much as American Hospital Supply was obligated to do during the period of renewal. American Hospital Supply responds that it had bought all it was contractually obligated to buy during the renewal term already, in advance as it were, to help Hospital Products with its cash-flow problems. Whether this is true is a difficult question that depends not only on collating a mass of documents but on resolving a clash of testimony, which only the judge who heard the testimony can do.
The same is true with regard to American Hospital Supply's other letter of June 3, the one referring to financial assistance, wherein Hospital Products finds a veiled threat not to pay for goods already bought unless it agreed to modify the contract in American Hospital Supply's favor. The letter does not say this, however; and the district judge, who listened to several witnesses as well as reading the various documents submitted by the parties, concluded that the renewal was unconditional and that while the parties may have differed with regard to their understanding of the extent of American Hospital Supply's obligations under the contract, American Hospital Supply's refusal to accede to Hospital Products' unilateral understanding was not a repudiation of the contract. This finding is not clearly erroneous and the legal principle that underlies it — the principle that it is not a breach of contract to refuse to accept the other party's interpretation of the contract — is unexceptionable, see, e.g., Farwell Construction Co. v. Ticktin, 84 Ill.App.3d 791, 800-01, 39 Ill. Dec. 916, 923-24, 405 N.E.2d 1051, 1058-59 (1980). Since, moreover, the finding was based on all the evidence we can think of that the parties might present at the actual trial on the merits, the chances that the judge was wrong and that his preliminary findings will be reversed at trial are small, and the probability of the plaintiff's prevailing at trial correspondingly great. In thus commenting on the merits, however, as we must do to review the district court's action in granting the injunction, we do not mean to prejudge the outcome of the full trial, which may cast the facts in a different light from how they appeared in the preliminary injunction proceeding. This caveat applies to all the discussion of the merits of the case in this opinion.
Before concluding that the grant of the preliminary injunction should be affirmed, we must consider Hospital Products' argument that the injunction should have been denied because of American Hospital Supply's "unclean hands" in advertising Hospital Products' line of surgical stapling products as its own; and a related argument that the grant of the injunction disserved the public interest.
Although the contract forbade American Hospital Supply to sell products competitive with Hospital Products' line during the term of the contract, American Hospital Supply developed, but did not (before the mailgram of June 7) sell, its own line. Hospital Products argues that in order to pave the way for dealers to switch to American Hospital Supply's line as soon as it reached the market, American Hospital Supply passed off Hospital Products' line as its own in its advertising.
If merely developing a competing line of products were a violation of the contract, then American Hospital Supply's conduct in developing and by devious means promoting a competing line (if indeed American Hospital Supply engaged in false advertising, a hotly contested and unresolved issue) would be relevant to the question who broke the contract first, and hence to the likelihood of American Hospital Supply's prevailing at trial on its breach of contract claim. But the contract does not forbid development of competing products. Or if Hospital Products were accusing American Hospital Supply of misbranding Hospital Products' line and thereby impairing Hospital Products' trademarks, this would be highly relevant to the question who broke the contract first, since it is a most serious violation of the implied contractual duty of good faith for a distributor to misbrand his supplier's product, whether by selling someone else's product under the supplier's name or the supplier's product under someone else's name (including the distributor's). See discussion in Lippo v. Mobil Oil Corp., 776 F.2d 706, 722-25 (7th Cir. 1985) (dissenting opinion). But we do not understand Hospital Products to be arguing that. Its argument is that even if American Hospital Supply made out an ironclad case that Hospital Products alone violated the distribution contract, the doctrine of unclean hands forbade the grant of a preliminary injunction.
The doctrine applies to preliminary injunctions as to other equitable remedies, of course, but as we have been at pains to emphasize in two recent cases, the doctrine is not to be used as a loose cannon, depriving a plaintiff of an equitable remedy to which he is otherwise entitled merely because he is guilty of unrelated misconduct. See Shondel v. McDermott, supra, 775 F.2d at 867-69; Polk Bros., Inc. v. Forest City Enterprises, Inc., 776 F.2d 185, 193 (7th Cir. 1985). To deny the preliminary injunction because American Hospital Supply was using false advertising to supplant Hospital Products as a maker of surgical stapling systems, as Hospital Products urges, would be to deny it for a reason unrelated to Hospital Products' alleged breach of contract in terminating American Hospital Supply as a distributor after the contract was renewed. By renewing the contract American Hospital Supply disabled itself for another year from selling in competition with Hospital Products. Denying the injunction would merely have accelerated the entry of American Hospital Supply into making a line of products previously dominated by Hospital Products and one other producer. It must not be forgotten that the contract which American Hospital Supply wants to enforce and Hospital Products wants to walk away from is an exclusive-dealing contract, with American Hospital Supply the dealer and Hospital Products the supplier. So long as the contract is in effect, American Hospital Supply is barred from competing with Hospital Products. Thus we do not see how an injunction that keeps the contract in effect for another year, or perhaps longer (more on this shortly), can be resisted on the basis that American Hospital Supply is trying to compete unfairly with Hospital Products.
Last, Hospital Products argues that the injunction is against the public interest. All such an argument means, as we said earlier, is that the injunction has effects on nonparties, in this case the members of the consuming public, which is to say the hospitals that are the ultimate purchasers of surgical stapling systems. Of course these effects must be taken into account in deciding whether to issue the injunction; when as here they are urged as reasons against the grant of the injunction, they are part of H[d] in our formula, that is, they are part of the harm (to the defendant — but also to anyone else adversely affected) if the injunction is granted. If the injunction had been denied, Hospital Products would have sold its products directly to dealers in the United States, and American Hospital Supply presumably would have sold its newly developed line to dealers, too, thus increasing the number of competing producers from two to three. From this it can be argued that renewal of the contract was anti-competitive because it has postponed American Hospital Supply's entry into the market as a competitor of Hospital Products. If "reusable surgical stapling systems for internal surgical procedures" constitute an economically meaningful market — meaning that if the producers colluded expressly or tacitly, they could raise the price of the product significantly without experiencing such a drastic loss of sales (as customers switched to substitutes, or producers of other products switched to making this product) as to make the price increase unprofitable — Hospital Products might have a winning argument; it is easier for two firms to collude without being detected than for three to do so. But Hospital Products has not tried to show that its product constitutes a genuine market and hence that there is a real danger of collusion; perhaps it has not tried to do so because it would be pointing to itself as one of the two colluders.
Another way to take Hospital Products' argument, however, is that the injunction weakens competition by keeping Hospital Products, not American Hospital Supply, out of the market. But if only because of American Hospital Supply's interest in liquidating its large unsold inventory of Hospital Products' goods, it is hardly likely that the injunction will in fact keep Hospital Products out of the market in any sense relevant to the concerns of antitrust policy. With all the discord between the parties, it is still the case that American Hospital Supply wants to sell off its inventory of Hospital Products' goods, necessarily in competition with those of the other producer. Even when that inventory is completely sold off, American Hospital Supply will be contractually obligated to buy from Hospital Products until the contract lapses; and what it buys, economic logic compels it to sell — especially since, so long as the contract continues to be renewed, American Hospital Supply, being forbidden to sell competing products, will have no choice but to promote Hospital Products' line as vigorously as it can. All this will keep Hospital Products in the market.
We comment briefly on the term of the preliminary injunction. Since federal litigation is sometimes protracted and since the contract entitles American Hospital Supply to renew the contract annually for six more years after this year, there is a theoretical possibility that the injunction will remain in effect for many years. Over such a period the balance of harms and impact on the public interest may of course change. We trust that the district judge will bear this in mind in considering any requests to modify the injunction.
Finally, we note that the parties and the district court may wish to consider the possibility that this lawsuit might be transferred to the district court in Connecticut in which the bankruptcy proceeding is pending, assuming the requirements of 28 U.S.C. § 1404(a) can be met. Cf. Lank v. Federal Ins. Co., 309 F. Supp. 349, 353 (D.Del. 1970). Still another possibility might be the transfer of the bankruptcy proceeding to Chicago. See 28 U.S.C. § 1412. Either form of transfer might make it easier to ensure that this lawsuit is not handled in a way that would unfairly prejudice the interests of Hospital Products' other creditors. We mean to imply no view on whether transfer (in either direction) might be possible or desirable, but it is something for the parties and the district judge to consider in their quest to do equity in this matter.
AFFIRMED.
MEMORANDUM ORDER
After careful consideration of the papers filed by the parties and the fine arguments made in open court yesterday morning, the Court grants plaintiffs' motion for a temporary restraining order ("TRO").
The Seventh Circuit recently recast the equitable standards for emergency injunctive relief. To obtain relief the plaintiffs must show that (1) they have no adequate remedy at law and will suffer irreparable harm if the relief is not granted; (2) the irreparable harm they would suffer outweighs the irreparable harm defendant would suffer if the injunction is not granted; (3) they have some likelihood of success on the merits; and (4) the desired injunction would not frustrate the "public interest." See Roland Machinery Co. v. Dresser Industries, 749 F.2d 380, 386-88 (7th Cir. 1984); United States v. Board of Education of Chicago, [ 610 F. Supp. 702, 705,] No. 80 C 5124, slip op. at 4-5 (N.D.Ill. June 4, 1985) (Aspen, J.).
Plaintiffs have fairly easily met three of the four requirements. First, we are convinced that they have no adequate remedy at law and would suffer irreparable harm absent a TRO. Defendants' mailgram of June 11 to plaintiffs' customers and their subsequent actions have cut plaintiff out of the "staples" market altogether, apparently in breach of the exclusive distributorship agreement. Damages cannot adequately compensate plaintiffs. Defendants are clearly on the brink of insolvency, if not insolvent already. See Roland, 749 F.2d at 386. This fact distinguishes the cases defendants rely upon, such as Lafayette Beverage Distributors v. Anheuser-Busch Co., 545 F. Supp. 1137, 1151 (N.D.Ill. 1982). Moreover, the large losses caused by defendants' devastation of plaintiffs' market are not easily calculated and neither is the loss to the plaintiffs' good will. Roland, 749 F.2d at 386.
Second, plaintiffs have some likelihood of success on the merits. At this point the parties are accusing each other of breaches, and it is unclear who will prevail. But the pending motion focuses only on defendants' conduct since June 7, 1985, and plaintiffs have at minimum "some likelihood" of showing that this conduct was unlawful. Even though on June 3, 1985, plaintiffs told defendants that they intended to renew the distribution agreement (which defendants had asked them to confirm), defendants unilaterally terminated the agreement. On June 11, 1985, they sent a mailgram of plaintiffs' customers, telling them that plaintiffs were no longer authorized to distribute defendants' staples. Defendants are now marketing their products at much lower prices, presumably to raise quick cash for their emergency. This conduct appears to breach the distribution agreement, see Agreement ¶ 8, and to have caused plaintiffs' irreparable harm.
Third, the "public interest" appears only indirectly related to this suit and does not work significantly for or against either side. Thus, it does not counsel against issuance of a TRO.
Defendants' most significant argument is that the balance of harms weighs heavily in their favor. In approaching the "balance" issue, we must apply a sliding scale. "The more likely the plaintiff is to win, the less heavily need the balance of harms weigh in his favor; the less likely he is to win, the more need it weigh in his favor." Roland. 749 F.2d at 387. We think plaintiffs have a substantial chance of showing that defendants' conduct since June 3, 1985, has been unlawful. Thus, the balance need not weigh strongly in their favor.
Still, the harm to defendants will be significant. Plaintiffs are huge, their business with defendants adding up to only about one percent of their total. In contrast, defendants do ninety-nine percent of their business with plaintiffs. A TRO might drive defendants out of business. We have sympathy for defendants' predicament, as expressed in their affidavit, but nevertheless believe that it does not counsel against issuing a TRO. First, under defendants' logic, they can destroy plaintiffs' market without fear of injunctive relief simply because they can show that enjoining their unlawful conduct could put them out of business. We cannot allow defendants to use their insolvency to so shield their allegedly unlawful conduct. Cf. Horn Abbot v. Sarsaparilla Ltd., 601 F. Supp. 360, 369-70 (N.D.Ill. 1984) (Aspen, J.). Second, it appears likely that defendants, in large part, owe their present existence to plaintiffs who have loaned millions to defendants to prevent bankruptcy. Third, we think that the bond we set below could well stave off disaster until a hearing on a preliminary injunction is soon held. Finally, to the extent, if at all, plaintiffs have unlawfully harmed defendants, they have more than enough resources to compensate defendants in damages.
For the foregoing reasons, a TRO should issue, the terms of which are set forth below. Under Fed.R.Civ.P. 65(c), the Court may set as bond a sum it deems proper to make defendants whole in case they turn out to have been unlawfully enjoined. According to defendants' affidavit, plaintiffs have ordered and either failed to pay for or cancelled about $5 million worth of goods. We will set a bond in that amount, which should more than adequately protect defendants' interests pending a hearing on a preliminary injunction. Defendants shall file a memorandum opposing plaintiffs' preliminary injunction motion on or before June 26, 1985. Plaintiffs shall reply by July 1, 1985. A preliminary injunction hearing will be held July 2, 1985, at 11:00 a.m. It is so ordered. This schedule as well as the TRO can be extended, of course, if the parties so agree or if the parties resume their negotiations and reach a partial or complete settlement.
In sum, defendants are temporarily restrained from:
(1) selling products in the United States in violation of plaintiffs' exclusive rights under the Distribution Agreement;
(2) taking further action in advising plaintiffs' customers or anyone else that plaintiffs are not the exclusive authorized distributor of defendants' products;
(3) from appointing any additional distributors to distribute their products in the United States; and
(4) from taking any action in derogation of plaintiffs' rights under the Distribution Agreement.
This TRO shall remain in effect for twenty days, that is, until July 9, 1985, unless the parties agree to extend it longer. See Horn Abbot, 601 F. Supp. at 370 n. 12 (where TRO is issued after notice and hearing, court may impose it for twenty days pending prompt hearing on preliminary injunction within that time). Plaintiffs shall post a bond of $5 million by Friday, June 21, 1985. It is so ordered.
/s/ Marvin E. Aspen United States District Judge
DATED 6/19/85
The court today continues what it began in Roland Machinery v. Dresser Industries, 749 F.2d 380 (7th Cir. 1984): a wholesale revision of the law of preliminary injunctions. In affirming the district court's grant of a preliminary injunction, the court goes through the motions of evaluating the decision below according to the well-settled precepts of the law of injunctions, but affirms, in fact, on the basis of its independent evaluation of the case. In so doing this court, as in Roland, again transgresses the limits of its appellate authority.
I
Hospital Products Ltd. ("HPL") is an Australian corporation engaged in the manufacture of surgical stapling products. HPL's primary products are reusable stainless steel surgical staplers and disposable cartridges containing stainless steel staples. The stapling system apparently permits surgeons to close wounds in less time and with less blood loss than conventional sutures. HPL stapling systems are sold in the United States by its wholly-owned subsidiary Surgeons Choice, Inc. ("SCI"), a Delaware corporation.
American Hospital Supply ("AHS"), an Illinois corporation, is the world's largest distributor of medical supplies. American V. Mueller ("AVM") is a division of AHS. On July 27, 1982 AVM entered into an exclusive distribution agreement with HPL and SCI. The agreement provided, inter alia, that AVM would make guaranteed purchases of HPL products and required that AVM provide SCI with non-cancellable purchase orders ninety days in advance of delivery. Under the agreement AVM was obligated to purchase $12.8 million worth of HPL products in 1985.
Throughout the period of the agreement HPL was in need of fresh infusions of capital. Not surprisingly, AHS quickly became HPL's largest creditor. On December 16, 1982 AHS agreed to invest $6 million in HPL through the purchase of a convertible debenture. The debenture was due in 1988. On May 25, 1983 AHS loaned an additional $3.8 million to HPL. The parties disagree as to the repayment date of the loan. The relationship between HPL and AHS was not an ordinary supplier/distributor relationship. Because of the financial commitments AHS had made to HPL the American company exercised an extraordinary degree of control over HPL operations.
It is fair to say that the relationship between HPL and AHS throughout the period of the distribution agreement was not always harmonious. However, the events leading to the present appeal began on February 28, 1985 when HPL advised AHS that it was in a position to distribute more than $1 million to creditors. AHS exercised its rights under its loan agreements to prevent HPL from making the distributions. In March 1985 AHS advised HPL to take no further action with respect to a planned public offering until the parties' meeting in April.
Relations between the two companies continued to deteriorate and on April 30, 1985 AHS told HPL that it intended to acquire HPL through a "scheme of arrangement" or put HPL into receivership. The scheme of arrangement was a takeover plan proposed by AHS under which HPL creditors would receive 34% of their debts and shareholders would receive 9% of the market value of their shares. AHS cancelled $7.3 million in outstanding orders, refused to pay $3.4 million for products already shipped, announced its intention to return $3.1 million of products already purchased, and demanded repayment of the $3.8 million loan. On April 30, 1985 AHS sued HPL for breach of the distribution agreement.
Under the terms of the distribution agreement, if AHS desired to terminate the agreement it was required to furnish written notice ninety days prior to September 1, 1985 (June 3, 1985). On June 3, 1985 HPL requested that AHS confirm whether it intended to renew the distribution agreement. AHS responded the same day by observing that since notice had not been given the agreement was consequently renewed. AHS gave these assurances even though it had already filed suit against HPL claiming that the same agreement had already been breached. On June 4 HPL informed AHS that in its opinion AHS's conduct and the April 30, 1985 meeting constituted a repudiation of the distribution agreement. On June 5, 1985 and then again on June 7, 1985 AHS repeated its belief that the agreement remained in full force. Nevertheless, on June 7, 1985 SCI sent a Mailgram to AVM's customers stating that AVM was no longer authorized to distribute SCI's stapling products. In the same Mailgram HPL announced the commencement of a direct sales campaign of its products.
On June 18, 1985 AHS moved for a temporary restraining order prohibiting HPL from selling its products directly in the United States. On June 19, 1985 the United States District Court for the Northern District of Illinois granted the restraining order. HPL thereupon terminated its employees and a receiver was appointed on June 25, 1985. On July 2 and 3, 1985 a preliminary injunction hearing was held. The district court granted AHS's motion for a preliminary injunction on July 8, 1985.
II
Parties seeking a preliminary injunction must meet four requirements. They must show that: (1) they have no adequate remedy at law and will suffer irreparable harm if the relief is not granted; (2) the irreparable harm they would suffer outweighs the irreparable harm defendants would suffer from an injunction; (3) they have some likelihood of success on the merits; and (4) the injunction would not disserve the public interest. Palmer v. City of Chicago, 755 F.2d 560, 576 (7th Cir. 1985); Godinez v. Lane, 733 F.2d 1250, 1257 (7th Cir. 1984).
Adequate Legal Remedy and Irreparable Harm
In granting the preliminary injunction the district court found that AHS had no adequate remedy at law and would suffer irreparable harm without a preliminary injunction. The district court found that the June 7 HPL Mailgram harmed AHS and AVM "significantly," and further found that the evidence was "clear" that AVM's customers refused to buy the millions of dollars of SCI products which AVM had on hand as a result of being informed that AVM was no longer an authorized distributor.
The district court ignored the fact that AHS did not, and indeed could not, prove a significant loss of sales. AHS's 1984 Annual Report listed sales of just under $3.5 billion. Even if all $12.8 million of HPL products AHS had agreed to purchase in 1985 were rendered unmarketable by the direct sales campaign lost sales would represent less than 0.4% of AHS's total sales. Of course, the mere fact that AHS is a giant corporation with enormous annual sales does not preclude a finding that AHS will be irreparably harmed by HPL's actions. At the same time, however, we cannot overlook the relative insignificance of HPL products to AHS revenues. Moreover, AHS submitted no evidence that the Mailgram had led to any actual loss of sales. In addition, at oral argument AHS admitted that if it later won on the merits it would be able to return all HPL inventory.
The majority concludes that a threat of irreparable harm to AHS was established "although the dollar amount of that harm is not known with any precision and we hesitate to call it great." The majority, however, does not rely on the district court's findings. Instead, the court hypothesizes a series of imagined harms that might have, but did not, furnish the basis for the grant of the preliminary injunction. The court suggests that the Mailgram "might have impaired American Hospital Supply's goodwill." There is no evidence of this in the record. The district court merely made a conclusory finding that AHS's alleged loss of goodwill would, if proven, be difficult to calculate. The majority suggests that the "suddenness of the termination and the urgent mode of announcement might have made the dealers think that American Hospital Supply must have engaged in unethical or unreasonable conduct." Again, there is no support for this statement in the record and even the majority concedes its theory is "speculative." The majority further suggests that the Mailgram jeopardized AHS's investment "because dealers might be reluctant to buy Hospital Products' goods from American Hospital Supply, not wanting to become enmeshed in a legal dispute between it and its supplier or perhaps even fearing that there might be some defect in the particular items being sold by American Hospital Supply." There is, again, simply no indication in the record of such a concern on the part of anyone. The task of appellate courts is simply to review the findings of the district courts and not to construct new findings of fact as they see fit.
In addition, the sending of the Mailgram was motivated, in part, by AHS's actions in bringing the breach of contract suit on April 30, 1985. HPL terminated AHS only after AHS had filed suit. AHS can thus hardly have been surprised when HPL proceeded to act as if the agreement had already been breached.
Both the district court and the majority assert that what made the threatened loss to AHS irreparable was HPL's insolvency. The evidence, however, does not show that HPL was on the "brink of insolvency" before the alleged failure of AHS to fulfill its contractual obligations. At worst, the evidence indicates that HPL was a profitable, albeit highly leveraged, concern. Just what brought about the financial collapse of HPL is the essence of the underlying dispute between the parties. The district court's determination that HPL's insolvency was irrelevant is thus a perfect example of question-begging. The insolvency dispute will be more effectively resolved within the framework of the ongoing bankruptcy proceedings. A preliminary injunction hearing in the district court is a clumsy vehicle for resolving factual disputes. If it later develops that HPL's bankruptcy was due, in significant part, to AHS's actions, AHS will have succeeded in benefitting from its own misconduct. Moreover, it is not clear to what extent the granting of the preliminary injunction itself insured that HPL would become insolvent. Finally, even if I were to concede that HPL's imminent insolvency rendered the recovery of damages questionable, Roland requires that a party moving for a preliminary injunction establish both a threat of irreparable harm and the lack of an adequate remedy at law. Even if the district court established the lack of an adequate remedy at law it cannot be permitted to bootstrap that finding into a finding of a threat of irreparable harm as well. Irreparable harm and lack of an adequate remedy at law are separate but equally necessary components of the Roland test.
Balancing of Harms
The district court found that the balance of harms weighed in AHS's favor. The district court acknowledged that the injunction might drive HPL out of business, but concluded that "defendants may not excuse their unlawful actions by their actual or threatened insolvency." The district court adopted by reference its previous analysis of the issue as set forth in the June 19, 1985 memorandum order granting the temporary restraining order. The June 19 order conceded that the harm to HPL caused by the injunction would be "significant." Nevertheless, it concluded the balance of harms weighed in AHS's favor for four reasons: (1) insolvency is no shield for unlawful conduct; (2) HPL owes its existence to the financial commitments of AHS; (3) the bond it required AHS to post would adequately protect HPL's interests; and (4) AHS has more than enough resources to compensate HPL in damages should it later be determined that the injunction was erroneously granted.
The district court's findings are an inadequate basis for injunctive relief. The court rejected HPL's contention that its probable bankruptcy in the event the injunction should issue was a factor to be considered in weighing the balance of harms. The district court erred in rejecting the importance of HPL's insolvency. The majority opinion states clearly "that the effect of a preliminary injunction in precipitating insolvency is a factor arguing against the grant of the injunction . . . reflection on this overlooked point might conceivably persuade the district judge to grant a motion to dissolve the preliminary injunction, should such a motion be made, as it can be at any time." The district court, of course, did not merely "overlook the point"; it drew a clearly erroneous conclusion of law. That conclusion was a crucial aspect of its decision to grant the preliminary injunction. The majority's decision, in light of its recognition of the district court's error, is inexplicable.
The second factor mentioned in the district court's June 19 order was that HPL owed its continuing existence to the financial commitments of AHS. It is not clear what the district court intended to imply by this remark except perhaps that AHS had acquired some vague equity rights by its decision to loan HPL money. This is clearly an improper basis upon which to grant a preliminary injunction. The moral obligations incurred by HPL because of the loans from AHS are not at issue here. The majority does not suggest why this particular finding of the district court is legally significant, but elsewhere states its belief, with which I agree, that AHS provided financial assistance to HPL "not out of altruism . . . but because it liked Hospital Products' product." AHS should derive no benefits because of a previous business decision to loan HPL money. Like any creditor AHS evaluated the risks involved in making the loans and judged them to be manageable. The fact that those loans may have turned out to be unwise is irrelevant. Thus, the second purported basis for the district court's order granting the preliminary injunction is also without justification.
In granting the temporary restraining order and the preliminary injunction the district court also relied on the capacity of AHS's posted bond and AHS's vast financial resources to compensate HPL for any harm resulting from the issuance of the injunction. These claims of the district court are discussed at length in the majority opinion and I will not repeat that discussion here. I agree with the majority, however, that the district court's reasoning in this matter is "not entirely satisfactory." As the majority points out the injunction bond and HPL's counterclaim for damages will not be an adequate source of compensation for the loss of a going business concern. HPL was one of only two manufacturers of surgical stapling products in the non-communist world. The company had acquired an international reputation in the surgical supply industry. The grant of the preliminary injunction may have helped destroy a company whose value cannot be measured by an injunction bond. To paraphrase Judge Friendly: HPL wants to make surgical supplies, not to live on the income from a damages award. Semmes Motors v. Ford Motor Co., 429 F.2d 1197, 1205 (2d Cir. 1970).
The majority cannot decide whether the district court was correct in concluding that the balance of harms weighed in AHS's favor or whether no clear balance of harms was shown. The majority merely concludes that in either event the injunction was properly issued. With all due respect the majority's treatment of the district court's findings is too cavalier. If AHS can establish only a parity of harms its right to injunctive relief and its likelihood of ultimate success on the merits is seriously jeopardized. See discussion infra at 607. We cannot affirm a district court's decision to grant an injunction when substantial questions remain concerning the district court's assessment of one of the four parts of the traditional preliminary injunction test.
In my opinion the balance of harms weighed in HPL's favor if for no other reason than the grant of the preliminary injunction was certain to precipitate HPL's slide into bankruptcy.
Likelihood of Success on the Merits
The district court also found that AHS and AVM have "some likelihood" of succeeding on the merits. This finding was based on the district court's belief that HPL's conduct in sending the June 7 Mailgram constituted a breach of the distribution agreement. HPL contends that the mailgram was justified because by June 7 AHS had already breached and repudiated the distribution agreement. A party repudiates a contract when there is "a positive and unequivocal manifestation of intention that the party will not render the promised performance when the time fixed in the contract arrives." B C Electric v. Pullman Bank Trust, 96 Ill.App.3d 321, 328, 51 Ill.Dec. 698, 421 N.E.2d 206 (1st Dist. 1981). The district court found no evidence of an intention to repudiate the agreement. It is true that AHS, on several occasions, said the distribution agreement was in full force and that it would comply with its contractual obligations. The district court relied on these representations in making its preliminary assessment that HPL, and not AHS, breached the agreement. Speaking with great respect, I believe the district court was clearly wrong to have placed so much reliance on what AHS might have said about the agreement. A party that intends to breach a contract is not likely to admit to the contemplated breach beforehand.
The district court should have more closely examined AHS's conduct. "Anticipatory breach is not founded only upon the spoken words of the breaching party. Were this so, the right to action would founder ineffectively upon the false or evasive verbiage of the wrongdoer." Tobriner v. Mayfair Extension, Inc., 250 F. Supp. 614, 617 (D.D.C. 1966), aff'd sub nom. Luxemberg v. Mayfair Extension, Inc., 382 F.2d 475 (D.C. Cir. 1967). AHS failed to pay $3.4 million in receivables for products shipped prior to April 30, 1985. AHS refused to accept delivery on $7.3 million in non-cancellable orders. HPL presented into evidence several documents indicating AHS's desire to alter or terminate its relationship with HPL. AHS brought a breach of contract suit against HPL one month before the fateful Mailgram. These actions go to the very heart of the question of whether there was an anticipatory breach of the distribution agreement. The district court was simply wrong to characterize HPL's allegations as "differences of opinion over the interpretation of an agreement."
The majority's review of this aspect of the district court's opinion misses the point entirely. The majority, after briefly and inconclusively discussing the merits of HPL's anticipatory breach claim, abruptly abandons its analysis, characterizing the dispute as "legal badminton." In order to affirm the decision below the majority merely concludes that the "able and experienced" district judge was on "solid ground" in resolving the legal uncertainty in AHS's favor. But whether the distribution agreement in question was breached is clearly a legal question courts of appeals need not refrain from answering. We owe deference to the factual findings, not the legal conclusions, of district courts. In this case the majority affirms the district court's ultimate finding that AHS established some likelihood of success on the merits while simultaneously expressing serious reservations regarding the district court's discussion of the merits of the crucial contract dispute.
The district court was correct in stating that Roland requires a district judge ruling on a motion for a preliminary injunction to find that the movant has shown "some likelihood of success" on the merits. I have indicated my belief that the district court was wrong in finding that the balance of harms weighed in AHS's favor. At most, the record supports a finding of a parity of harms. The majority does not disagree with this assessment. Under Roland, when the moving party has shown only a parity of harms, the district court must apply a stricter standard than "some likelihood of success." Roland mandates that when "there is no clear balance of hardships in favor of the injunction" the moving party "must show that it is more likely than not to win." Id. at 392. Only then would the error of denying AHS a preliminary injunction should it later win on the merits be more costly than the error of granting the injunction should AHS later lose on the merits. Id. Because the district court clearly erred in finding that the balance of harms weighed in AHS's favor, the district court necessarily compounded its error by failing to apply the stricter standard of "more likely than not to win" required when only a parity of harms has been shown. AHS has not shown that it is more likely than not to ultimately prevail on the merits of this dispute. The contract issues of this case are obviously too close to call. The district court's finding that AHS had shown some likelihood of success on the merits applied an incorrect legal standard and constitutes reversible error.
I expressed my reservations concerning this court's recent reevaluation of the preliminary injunction standard in my dissenting opinion in Roland, 749 F.2d at 396. Events since have only reinforced my belief that the majority's discussion in Roland of the preliminary injunction standard represents a "fundamental misunderstanding of the role of preliminary injunctive relief in our legal system and the role of the district courts in dispensing that relief." Id. at 397.
To recapitulate: the district court found that AHS had no adequate remedy at law and would suffer irreparable harm without an injunction. The majority agrees, but only after making new findings of fact not supported in the record. The district court found that the balance of harms weighed in AHS's favor. The majority's position on this point is unclear, but ultimately it decides that any error was harmless. The district court found that AHS had some likelihood of success on the merits. The majority agrees, but with apparent reluctance. It would appear that in this circuit, despite vigorous protestations to the contrary, the standard of review of the grant or denial of a preliminary injunction is effectively de novo. When reviewing a grant or denial of an injunction, this court will independently evaluate the evidence. If our evaluation agrees with that of the district court, we will affirm. If our evaluation disagrees, we will reverse.
The majority twice attempts to excuse the inadequacy of the district court's findings in this case by reminding the reader of the haste with which a judge requested to issue a preliminary injunction must act. As a former district judge I am well aware of the extraordinary responsibilities of the district courts in granting or denying preliminary injunctions. But these responsibilities cannot be used to justify the erroneous issuance of an injunction. Fed.R.Civ.P. 65(d) requires that an order granting an injunction "shall set forth the reasons for its issuance." That requirement is for the benefit of both the reviewing court and the party enjoined. Small v. Kiley, 567 F.2d 163, 164 (2d Cir. 1977). From what appears in the record there is simply no basis upon which the district court could have made the requisite findings required by this court's governing law of injunctions. Recently, we have been far less deferential to the difficult choices made by district courts in deciding whether to issue an injunction. See, e.g., McCall-Bey v. Franzen, 777 F.2d 1178 (7th Cir. 1985). In my view the district court was affirmed because the result it reached met with the approval of the majority.
This summary of the majority's decision may seem harsh, and so in order to demonstrate the accuracy of my criticism I have attached, as an appendix to this dissent, the opinions of the district court in this case.
III
I would have preferred to avoid commenting on the majority's attempt to reduce the well-developed and complex law of preliminary injunctions to a "simple" mathematical formula. But because of the potentially far-reaching and baneful consequences of today's decision, I must regretfully voice my concerns.
Henceforth, the district courts of this circuit should grant a preliminary injunction if, "but only if," P X H[p] (1 — P) X H[d]: where P is the probability that denial of the injunction would be an error; where H[p] is the harm to the plaintiff if the injunction is denied; and where H[d] is the harm to defendant if the injunction is granted.
The majority describes its formula as a procedural counterpart to Judge Hand's negligence formula first appearing in United States v. Carroll Towing, 159 F.2d 169, 173 (2d Cir. 1947). Carroll Towing was an admiralty case in which a shipowner's duty to provide against injuries resulting from the breaking of a vessel's moorings was expressed in algebraic terms. In Hand's formula the liability of the shipowner depends on whether B PL, where P is the probability that the ship will break away; where L is the gravity of the resulting injury if she does; and where B is the burden of adequate precautions. Various attempts have been made to apply the Hand formula, or some derivation of it, to areas other than negligence. Professor Kronman, for instance, now at the Yale Law School, has attempted to apply the balancing standard of Carroll Towing to governmental claims that information is exempt from disclosure under Exemption 6 of the Freedom of Information Act. 5 U.S.C. § 552(b)(6). Kronman, The Privacy Exemption to the Freedom of Information Act, IX(4)-J. Legal Stud. 727 (1980); see also Washington Post v. United States Department of Health, 690 F.2d 252, 276 (D.C. Cir. 1982) (Tamm, J., dissenting). Most courts, however, have continued to view the Carroll Towing opinion as a negligence formula. See, e.g., Complaint of Paducah Towing, 692 F.2d 412 (6th Cir. 1982); Maine Yankee Atomic v. NLRB, 624 F.2d 347 (1st Cir. 1980). In recent years the Carroll Towing opinion has undergone a renewed popularity in this circuit. Llaguno v. Mingey, 763 F.2d 1560, 1564 (7th Cir. 1985); United States Fidelity Guaranty v. Jadranska Slobodna Plovidba, 683 F.2d 1022, 1026 (7th Cir. 1982); Evra Corp. v. Swiss Bank Corp., 673 F.2d 951, 958 (7th Cir. 1982). My quarrel, however, is not with Carroll Towing but rather with the majority's attempt today to create its equitable analogue. A quantitative approach may be an appropriate and useful heuristic device in determining negligence in tort cases, but it has limited value in determining whether a preliminary injunction should issue. Proceedings in equity and cases sounding in tort demand entirely different responses of a district judge. The judgment of the district judge in a tort case must be definite; the judgment of the district judge in an injunction proceeding cannot, by its very nature, be as definite. The judgment of a district judge in an injunction proceeding must be flexible and discretionary — within the bounds of the now settled four-prong test.
I question the necessity and the wisdom of the court's adoption of a mathematical formula as the governing law of preliminary injunctions. The majority claims that its formula is merely a distillation of the traditional four-prong test. But if nothing is added to the substantive law, why bother? The standard four-prong test for determining whether a preliminary injunction should issue has survived for so many years because it has proven to be a workable summation of the myriad factors a district court must consider in deciding whether to grant an injunction. The test articulated in Technical Publishing v. Lebhar-Friedman, 729 F.2d 1136, 1138 (7th Cir. 1984), and in countless other cases, may not exhibit the "precision" the majority seems to demand, but such "precision" is antithetical to the underlying principles of injunctive relief. Equity, as the majority concedes, involves the assessment of factors that cannot be quantified. A district court faced with the task of deciding whether to issue a preliminary injunction must to some extent, the majority concedes, rely on the "feel" of the case. See generally my discussion in Roland, 749 F.2d at 396. The majority's formula will not assist the district courts in their assessment of this aspect of the decision to grant a preliminary injunction. The traditional element of discretion residing in the decision of a trial court to grant a preliminary injunction has been all but eliminated by today's decision.
Ironically, the majority never attempts to assign a numerical value to the variables of its own formula. We are never told how to measure P or H[p] or H[d]. I believe, and the majority appears to concede, that a numerical value could never be assigned to these variables. Who can say, for instance, what exactly the probability is that the granting of the injunction was an error? How then will the majority's formula ease in a meaningful way the responsibilities of the district courts? Judges asked to issue a preliminary injunction must, in large part, rely on their own judgment, not on mathematical quanta.
We must, of course, be mindful not to vest too much imprecision in the preliminary injunction standard, for law implies a system of known and generally applicable rules. See Fiss Rendleman, Injunctions 104 (2d ed. 1984). The existing four-prong test, however, represents the historical balance struck by the courts between the rigidity of law and the flexibility of equity.
The majority disavows any effort to force the district courts into a "quantitative straitjacket," but I suspect that today's decision may lead to just that. District judges operate under enormous pressure to be decisive and precise. Much rides on their smallest decisions. Like a Homeric Siren the majority's formula offers a seductive but deceptive security. Moreover, the majority's formula invites members of the Bar to dust off their calculators and dress their arguments in quantitative clothing. The resulting spectacle will perhaps be entertaining, but I do not envy the district courts of this circuit and I am not proud of the task we have given them.
I would reverse the district court's issuance of the preliminary injunction.
HPL retains the option of returning to the district judge and moving for a modification of the injunction on the basis of bankruptcy. Having sent the corrective Mailgram HPL has now allowed AHS to sell its inventory of HPL products. In return HPL should be permitted to sell its product in the United States until the contract dispute is resolved.