These consolidated appeals bring up to us several rulings, made in 1978 and 1981, disposing of various aspects of a patent suit. See 453 F. Supp. 743 (N.D.Ill.1978); 514 F. Supp. 213 (N.D.Ill.1981). The district court has certified these rulings for immediate appeal under Fed.R.Civ.P. 54(b). The rulings involve the res judicata effect of patent consent decrees, fraud as a defense to res judicata, and the lawfulness of differential patent royalties under antitrust and patent-misuse principles.

SPS, a manufacturer of industrial fasteners, owned a patent, issued in 1963, on a patch-type self-locking industrial fastener. In 1969 it sued USM, a competing manufacturer of fasteners, for infringement. After a trial on the issue whether USM had a valid license under the patent by virtue of a grant-back clause in a licensing agreement between the parties, the district court held that USM did not have a valid license. Standard Pressed Steel Co. v. Coral Corp., 168 U.S.P.Q. 741 (N.D.Ill.1971). The parties then settled the case by entry of a consent judgment in which USM acknowledged that the patent was valid and had been infringed. As part of the settlement SPS granted USM a license which allowed USM to continue using the patent but required it to pay royalties to SPS.

In 1974, three years after SPS's suit had been settled, USM brought the present suit, seeking to invalidate SPS's patent and get back the royalties it had paid since the settlement. For the first time USM alleged that SPS had procured the patent by a fraud on the Patent Office, and alleged that SPS had concealed this fraud from the district court in the previous suit and by so doing had committed a fraud on the court as well. The district court held after a bench trial that SPS had committed a fraud on the Patent Office, but not on the court, and that USM was entitled to have the patent declared void as of the date the second suit had been filed and to get back the royalties it had paid since then, plus attorneys' fees. The court declined to reconsider a ruling in its 1978 opinion that res judicata prevented USM from getting back any of the royalties it had paid before the second suit was filed--that is, between 1971, the date of the license agreement, and 1974.

In these cross-appeals, USM challenges the district court's conclusion that there was no fraud on the court in the first suit and argues that in any event a consent judgment in a patent case should have no res judicata effect, while SPS challenges the district court's conclusion that it committed a fraud on the Patent Office and argues that in any event res judicata bars USM from any relief based on that fraud. If SPS's res judicata argument is correct, we need not consider whether the district court correctly found fraud on the Patent Office; so it is with the res judicata issue that we begin. We cannot end there, though. USM's suit not only challenges the validity of the patent but also alleges that certain terms that first appeared in the license agreement entered into at the termination of the first suit constitute patent misuse. As these terms could not have been challenged in the first suit, USM's challenge to them cannot be barred by res judicata. The district court granted summary judgment for SPS on this aspect of the case in 1978, and USM's appeal brings this ruling up to us along with the rulings on fraud and res judicata.

American Equipment Corp. v. Wikomi Mfg. Co., 630 F.2d 544 (7th Cir.1980), is the authoritative decision in this circuit on the issue of the res judicata effect of consent decrees in patent suits. American had sued Wikomi for patent infringement and the case had been settled--just as SPS's suit against USM was settled--by the entry of a consent decree that recited validity and infringement. Also as in this case, the infringer, Wikomi, had received a license to continue using the patent. Wikomi's successor corporation refused to pay the royalties due under the license, and American sued. We held that American was entitled to summary judgment because a patent consent decree which recites both validity and infringement is res judicata in any subsequent proceeding between the parties (or those in privity of contract with them) over the validity or infringement of the patent. This holding is supported by other recent decisions and by scholarly commentary. See, e.g., Interdynamics, Inc. v. Firma Wolf, 653 F.2d 93, 96-98 (3d Cir.1981); Vulcan, Inc. v. Fordees Corp., 658 F.2d 1106, 1111 (6th Cir.1981); Note, "To Bind or Not to Bind": Bar and Merger Treatment of Consent Decrees in Patent Infringement Litigation, 74 Colum.L.Rev. 1322 (1974). We decline USM's invitation to reexamine it.

Unless the district court's finding that SPS did not commit fraud on the court that approved the consent decree is overturned, a matter to which we shall turn shortly, Wikomi requires us to reject the district court's Solomonic approach of giving the consent decree some but not complete res judicata effect--that is, up to the date on which the second suit was filed. Wikomi held that the second suit was barred in its entirety by the consent judgment in the first. True, the consent judgment in Wikomi included an injunction against infringement and the one here did not, but that makes no difference; it is the fact that a cause of action has been adjudicated rather than the specific relief granted the prevailing party that brings res judicata into play. Lambert v. Conrad, 536 F.2d 1183, 1185 (7th Cir.1976); 1 Restatement (Second) of Judgments Sec. 25(2) (1982).

We are asked to create an exception to Wikomi for cases where the patent is alleged not merely to be invalid but to have been procured by fraud on the Patent Office. (The Wikomi opinion does not indicate the basis on which Wikomi's successor attacked the validity of the patent in the second action.) But such an exception, rather than deterring patent fraud, would reduce the incentive of an infringer to prove fraud when first sued; he would have nothing to lose by biding his time until the terms of the settlement became onerous to him. True, res judicata, like any other threshold defense, may immunize some patent frauds--the one alleged in this case, for example--from challenge by some licensees. But the immunity will be incomplete, since res judicata can be invoked only against a party to a previous litigation and those in privity with that party--against USM and those in privity with it, but not against SPS's other licensees (there are some, as we shall see), or the Department of Justice, which can sue to cancel a patent procured by fraud. United States v. Saf-T-Boom Corp., 431 F.2d 737 (8th Cir.1970) (per curiam).

As we are not persuaded that patent fraud will on balance be less common if there is an exception to res judicata for cases in which such fraud is alleged, the general policy in favor of finality of litigation leads us to reject the exception. And since it is immaterial that in this case fraud was not alleged in the earlier suit--fraud is just another ground for contesting the validity of a patent, and res judicata prevents the litigation of any ground that could have been advanced in the earlier suit in support of the claim (patent invalidity) made there, e.g., Diaz v. Indian Head, Inc., 686 F.2d 558, 562 (7th Cir.1982); 1 Restatement, supra, Sec. 25(1)--all that remains to be considered on the issue of patent validity is whether the consent decree, as distinct from the patent, was procured fraudulently. SPS concedes that if it was, it has no res judicata effect.

Illustrative of SPS's allegedly fraudulent conduct is its counsel's refusal, in the course of pretrial discovery in the first suit, to turn over to USM a patent application that the inventor of the patented device in issue, Mr. Villo, had submitted to the Patent Office some years before that patent was applied for but that he had later withdrawn. SPS's failure to draw this application to the Patent Office's attention was one of the acts that the court below deemed a fraud on the Patent Office. If this was fraud, and if SPS fraudulently withheld the application from opposing counsel in the first lawsuit, a strong argument can be made that SPS obtained the consent decree confirming the validity of its patent by fraud also.

At his deposition, Villo mentioned having filed an earlier patent application. USM's counsel asked SPS's counsel "if we could see a copy of that application." SPS's counsel refused, saying that Villo had been describing "an earlier form of the invention .... It isn't the form of invention that is presently disclosed and claimed. I don't think that application is relevant, if there was one at that time." USM dropped the matter. The district court found that USM knew enough about the application from this and other stray bits of information that if it had "elected to make further discovery inquiries" it would have obtained the contents of the application itself and any other information it needed to prove a fraud on the Patent Office.

At first glance this finding may seem to import into a fraud inquiry, where it normally would not belong, a concept of contributory negligence: if USM had been more diligent, it would not have been fooled. Contributory negligence is a defense only to unintentional torts, and fraud is an intentional tort. See, e.g., Cenco Inc. v. Seidman & Seidman, 686 F.2d 449, 454 (7th Cir.1982). But the district court was not alone in thinking that a party's diligence--before as well as after the entry of judgment--is relevant to deciding whether the party may get the judgment set aside on grounds of fraud. One of the "critical considerations" bearing on that decision is "whether in the original action the victim had pursued reasonable precautions against deception." 2 Restatement, supra, Sec. 70, at p. 182. Fraud in responding to a request for discovery should mean a more active obstruction of the inquiry than simply putting opposing counsel to the burden of getting a discovery order--which is all that happened here. SPS's counsel did not deny the existence of the patent application; and though he denied its relevance, that would not have prevented USM's counsel from obtaining the document if he had tried harder. Counsel merely refused to produce the document voluntarily, no doubt in the hope, which proved to be well founded, that USM's counsel would not pursue the matter further.

Fraud is just a name for the misrepresentations and omissions that legislators or judges want to punish; the concrete question we have to answer is whether the kind of conduct that SPS's counsel engaged in should be punished--and by denying finality to consent decrees, a heavy sanction. We are unwilling to judge the conduct of SPS's counsel by the standards that would be appropriate if he had been responding to a request from his client or from someone else to whom he owed fiduciary obligations. His relationship with USM's counsel was adversary rather than fiduciary. The American system of justice has been built on the premise that truth, at least the sort of truth that is relevant to legal rights and remedies, is likeliest to emerge from a vigorously competitive contest between opposing counsel. In any competitive contest--even war--there are constraints on the adversaries. In litigation one of these is that the adversaries may not resort to fraud; but that as we have said is a conclusion rather than a standard. It would be psychologically unrealistic, given the adversary setting, to call a failure to go out of one's way to produce damaging documents a "fraud" on opposing counsel and so, perhaps, on the court. To respond to specific discovery requests is one thing; to construe them broadly is another, and goes against the trial lawyer's grain. We agree that a lawyer's "failure to disclose an instrument which he could have supposed reasonably--although, as it now appears, erroneously--to have been known to his adversary" is not fraud on the court. Kupferman v. Consolidated Research & Mfg. Corp., 459 F.2d 1072, 1081 (2d Cir.1972).

Hazel-Atlas Glass Co. v. Hartford-Empire Co., 322 U.S. 238, 64 S. Ct. 997, 88 L. Ed. 1252 (1944), where the Supreme Court pushed the concept of fraud on the court in patent cases to its furthest, and some think an unjustified, extreme, see 7 Moore & Lucas, Moore's Federal Practice 512-13 (2d ed. 1982); 11 Wright & Miller, Federal Practice and Procedure Sec. 2870, at pp. 255-56 (1973), does not require a different result. A patent applicant's lawyer wrote an article extolling as a revolutionary advance the method of molding glass that was the subject of the application, and persuaded the president of the glass blowers' union to sign the article as its author. The lawyer's role was not disclosed. The patent was issued and a federal court of appeals relied on the article in holding the patent valid.

That lawyer's misrepresentation was more egregious than the conduct here, which conformed to the well-nigh universal practice of construing an adversary's document requests narrowly and yielding ground only slowly and grudgingly. We would be less tolerant if there had been no adversary contest, and the parties had jointly concealed information from the judge who had to approve the consent decree. The level of candor required in the parties' dealings with the court is higher than that required in their dealings with each other in the heat of forensic combat.

Other acts of fraud on the court are alleged besides the failure to hand over Villo's earlier patent application, but they are similar or at least no worse, and we agree with the district court that neither singly nor in combination do they warrant setting aside the consent decree. Hence the 1971 decree was valid; it prevents the parties from relitigating the validity of SPS's patent and USM's alleged infringement of it; and so we need not consider whether the district court was correct in concluding that SPS had obtained its patent by a fraud on the Patent Office.

The remaining issue is whether SPS committed patent misuse by including a differential royalty schedule in the license agreement entered into as part of the settlement of the earlier suit. The agreement requires USM to remit to SPS 25 percent of any royalties it obtains by sublicensing SPS's patent, except that if USM should happen to sublicense any of four companies that SPS had previously licensed directly USM must remit 75 percent of the royalties obtained from the sublicensee(s).

The doctrine of patent misuse has been described as an equitable concept designed to prevent a patent owner from using the patent in a manner contrary to public policy. Morton Salt Co. v. G.S. Suppiger Co., 314 U.S. 488, 62 S. Ct. 402, 86 L. Ed. 363 (1942). This is too vague a formulation to be useful; taken seriously it would put all patent rights at hazard; and in application the doctrine has largely been confined to a handful of specific practices by which the patentee seemed to be trying to "extend" his patent grant beyond its statutory limits. An early example was fixing the price at which the purchaser of the patented item could resell it. See Bauer & Cie. v. O'Donnell, 229 U.S. 1, 33 S. Ct. 616, 57 L. Ed. 1041 (1913). The courts reasoned (in rather a circular fashion, one must admit) that once the patent owner had given up title to the patented item his patent rights were at an end, and any further restriction on the purchaser would extend the patent beyond its statutory bounds. Similar thinking lies behind the most common application of the doctrine, which is to prevent the patent owner from requiring his licensees to buy an unpatented staple item used with the patented device--for example, ink with a mimeograph machine. See generally Dawson Chem. Co. v. Rohm & Haas Co., 448 U.S. 176, 188-93, 100 S. Ct. 2601, 2609-11, 65 L. Ed. 2d 696 (1980).

Both examples--resale price maintenance and tying--suggest an overlap between misuse and antitrust principles. But although resale price maintenance by patentees was condemned as misuse shortly after Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373, 31 S. Ct. 376, 55 L. Ed. 502 (1911), held that the Sherman Act forbade resale price maintenance in nonpatent cases, see Bauer & Cie. v. O'Donnell, supra, and patent tie-ins were condemned as misuse shortly after the enactment of the tying provision (section 3) of the Clayton Act, 15 U.S.C. Sec. 14, in 1914, see Motion Picture Patents Co. v. Universal Film Mfg. Co., 243 U.S. 502, 517-18, 37 S. Ct. 416, 420-21, 61 L. Ed. 871 (1917), in both instances the condemnation of the patentee's conduct was based on the doctrine of patent misuse rather than on antitrust law. More recently the doctrine has been used to forbid the patentee to require his licensees to pay royalties beyond the expiration of the patent, Brulotte v. Thys Co., 379 U.S. 29, 85 S. Ct. 176, 13 L. Ed. 2d 99 (1964), or to measure royalties by the sales of unpatented end products containing the patented item, Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 133-40, 89 S. Ct. 1562, 1581-85, 23 L. Ed. 2d 129 (1969), or to require licensees not to make any items competing with the patented item, Stewart v. Mo-Trim, Inc., 192 U.S.P.Q. 410 (S.D.Ohio 1975).

As an original matter one might question whether any of these practices really "extends" the patent. The patentee who insists on limiting the freedom of his purchaser or licensee--whether to price, to use complementary inputs of the purchaser's choice, or to make competing items--will have to compensate the purchaser for the restriction by charging a lower price for the use of the patent. If, for example, the patent owner requires the licensee to agree to continue paying royalties after the patent expires, he will not be able to get him to agree to pay as big a royalty before the patent expires.

In all of these cases the patentee's total income may be higher--why else would he impose the restriction? But there is nothing wrong with trying to make as much money as you can from a patent. True, a tie-in can be a method of price discrimination. It enables the patent owner to vary the amount he charges for the use of the patent by the intensity of each user's demand for the patent (e.g., the mimeograph), as measured by the user's consumption of the tied product (e.g., the ink). Heaton-Peninsular Button-Fastener Co. v. Eureka Specialty Co., 77 F. 288, 296 (6th Cir.1896); Stigler, The Theory of Price 210-11 (3d ed. 1966); Bowman, Patent and Antitrust Law 55, 116-19 (1973). But since, as we shall see, there is no principle that patent owners may not engage in price discrimination, it is unclear why one form of discrimination, the tie-in, alone is forbidden.

But whether decided rightly or wrongly these are all cases where the license purports to enlarge the licensee's obligations beyond the limits of the patent grant. There is nothing of that sort here. But we must also consider whether the patent-misuse doctrine goes beyond these specific practices and constitutes a general code of patent licensing distinct from antitrust law.

The doctrine arose before there was any significant body of federal antitrust law, and reached maturity long before that law (a product very largely of free interpretation of unclear statutory language) attained its present broad scope. Since the antitrust laws as currently interpreted reach every practice that could impair competition substantially, it is not easy to define a separate role for a doctrine also designed to prevent an anticompetitive practice--the abuse of a patent monopoly. One possibility is that the doctrine of patent misuse, unlike antitrust law, condemns any patent licensing practice that is even trivially anticompetitive, at least if it has no socially beneficial effects. This might seem to explain cases such as Duplan Corp. v. Deering Milliken, Inc., 444 F. Supp. 648, 697 (D.S.C.1977), aff'd in relevant part, 594 F.2d 979 (4th Cir.1979), which held that a patent tie-in agreement is misuse per se unless the patentee shows that he had some nonmonopolistic reason for the tie-in, such as protection of goodwill. To prove a tie-in prima facie unlawful under the antitrust laws all you have to show is that the defendant has some economic power in the market for the tying product, United States Steel Corp. v. Fortner Enterprises, Inc., 429 U.S. 610, 97 S. Ct. 861, 51 L. Ed. 2d 80 (1977), and Duplan eliminates this requirement in misuse cases. But if a patentee has no market power (and, of course, not every patent confers market power, SCM Corp. v. Xerox Corp., 645 F.2d 1195, 1203 (2d Cir.1981)) he cannot use a tie-in to practice price discrimination, which presupposes market power. Stigler, supra, at 211. Much less can he lever his way into a dominant position in the market for the tied product. The logical presumption in such a case is that the tie-in promotes efficiency--and there is no lack of hypotheses as to how it might do that. See Bork, The Antitrust Paradox 375-81 (1978). It is hard to understand why in these circumstances, where if any presumption is warranted it is that the tie-in promotes efficiency rather than reduces competition, the burden of proof on the issue of misuse should be shifted to the patentee.

But probably cases like Duplan--which was, like Motion Picture Patents Co., supra, a tie-in case--are best understood simply as applications of the patent-misuse doctrine within its conventional, rather stereotyped boundaries. Outside those boundaries there is increasing convergence of patent-misuse analysis with standard antitrust analysis. See, e.g., Carter-Wallace, Inc. v. United States, 449 F.2d 1374, 1378-82 (Ct.Cl.1971); Congoleum Indus., Inc. v. Armstrong Cork Co., 366 F. Supp. 220, 227-32 (E.D.Pa.1973), aff'd, 510 F.2d 334 (3d Cir.1975); SCM Corp. v. Xerox Corp., 463 F. Supp. 983, 997-98 (D.Conn.1978) (the lengthy subsequent history of this case is irrelevant). One still finds plenty of statements in judicial opinions that less evidence of anticompetitive effect is required in a misuse case than in an antitrust case. See, e.g., Transitron Electronic Corp. v. Hughes Aircraft Co., 487 F. Supp. 885, 892-93 (D.Mass.1980), aff'd, 649 F.2d 871 (1st Cir.1981). But apart from the conventional applications of the doctrine we have found no cases where standards different from those of antitrust law were actually applied to yield different results. For example, the issue in Transitron was whether patent misuse is a tort; the court held it was not.

If misuse claims are not tested by conventional antitrust principles, by what principles shall they be tested? Our law is not rich in alternative concepts of monopolistic abuse; and it is rather late in the day to try to develop one without in the process subjecting the rights of patent holders to debilitating uncertainty. Cf. Hensley Equipment Co. v. Esco Corp., 383 F.2d 252, 261-62 n. 19, amended, 386 F.2d 442 (5th Cir.1967).

We come at last to the particulars of USM's charge of patent misuse, which the district court dismissed on summary judgment and which for the reasons just explained we think must be evaluated under antitrust principles. The basic charge is simply that SPS has set a discriminatory royalty schedule. But no general principle of antitrust law forbids charging different prices to different customers, what is often but loosely called "price discrimination." (The technical economic definition of price discrimination is disparity of price-cost ratios rather than of prices alone. Stigler, supra, at 209.) It is not illegal per se, even under section 2(a) of the Clayton Act as amended by the Robinson-Patman Act, 15 U.S.C. Sec. 13(a). O. Hommel Co. v. Ferro Corp., 659 F.2d 340, 346 (3d Cir.1981); American Oil Co. v. FTC, 325 F.2d 101, 106 (7th Cir.1963). It might in a particular case be condemned as an attempt to monopolize or as an act of monopolization under section 2 of the Sherman Act, 15 U.S.C. Sec. 2, or as a violation of the Rule of Reason under section 1 of that Act, but USM has made no effort to prove the elements of any of these offenses.

Specifically, there is no antitrust prohibition against a patent owner's using price discrimination to maximize his income from the patent. Bela Seating Co. v. Poloron Prods., Inc., 438 F.2d 733, 738 (7th Cir.1971). The furthest the courts have gone in condemning patent price discrimination under antitrust principles is in a series of cases involving a patented process for machine peeling of shrimp. See La Peyre v. FTC, 366 F.2d 117 (5th Cir.1966); Laitram Corp. v. King Crab, Inc., 244 F. Supp. 9, modified, 245 F. Supp. 1019 (D.Alaska 1965); Peelers Co. v. Wendt, 260 F. Supp. 193 (W.D.Wash.1966). The patentee leased the machines at twice the price to Pacific Northwest shrimp processors as to Gulf Coast processors, because hand peeling the smaller Pacific Northwest shrimp required twice as much labor. This was price discrimination in its economic sense. The lease rate varied according to the different intensities of the buyers' demands for the patented process (greater in the Pacific Northwest because of higher labor costs replaced by the patented process), rather than according to any difference in the patentee's costs of dealing with the two regions. But the focus of concern of the Federal Trade Commission and the courts was not on the evils of price discrimination in any abstract sense but on the effect of the discrimination in limiting the competition that the Pacific Northwest industry would have been able to offer the Gulf Coast industry if equal lease rates had been charged--an effect that seemed all the more sinister because the patentee had an interest in a Gulf Coast processor.

The decisions have been criticized. See Bowman, supra, at 105-10; Baxter, Legal Restrictions on Exploitation of the Patent Monopoly: An Economic Analysis, 76 Yale L.J. 267, 280-99 (1966). They require a patentee to establish a pricing schedule that will increase competition in the industries that use the invention--as if the function of antitrust law were to compel firms to maximize competition (between customers, no less), rather than to prevent them from restricting it. There is a difference between positive and negative duties, and the antitrust laws, like other legal doctrines sounding in tort, have generally been understood to impose only the latter. The pricing schedule did not harm the Pacific Northwest industry, but merely left it in the same position relative to the Gulf Coast industry that it had occupied before the patented process was invented.

But whether they were decided correctly or incorrectly, the shrimp peeler cases are distinguishable from the present case; no competitive effects in the market of the patentee's customers have been shown here. See Bela Seating Co. v. Poloron Prods., Inc., supra, 438 F.2d 733 at 738-39. USM has made no offer to prove that competition in the manufacture or sale of the products made by SPS's licensees and sublicensees (corresponding to the shrimp peelers) would be greater but for the royalty differential; and it is unlikely that it would be. The main differential is in the amount of royalties retained by USM rather than in the amount paid by the sublicensees. True, there potentially is some differential in that amount. While the usual royalty rate is 4 percent, of which USM retains 3 percent and SPS gets 1 percent, if USM sublicenses the four companies already licensed by SPS and those companies do not drop their SPS licenses, the royalty rate rises to 5 percent, with USM retaining 1 percent and SPS getting the other 4 percent. But a one percent cost difference is too small to give rise to an inference of significant competitive effect.

Conceivably the much larger difference between the amount of royalties retained by USM and the amount retained by SPS could affect competition not among the sublicensees but between these two firms. USM presented evidence that its technology embodying SPS's patent was superior to SPS's own technology and that SPS had imposed the royalty retention differential because it knew that without it USM would outcompete SPS to license the four companies. Even if this were true, it would not get USM very far in making out an antitrust case; as the district court pointed out, the essence of the patent grant is to allow the patentee to exclude competition in the use of the patented invention or, within broad limits not apparently exceeded here, to license competitors only on such terms as he sees fit. In any event, USM made no effort to present evidence of actual or probable anticompetitive effect in a relevant market, as is required in every Rule of Reason antitrust case in the Seventh Circuit. Dos Santos v. Columbus-Cuneo-Cabrini Medical Center, 684 F.2d 1346, 1352 (7th Cir.1982). There is no argument that the royalty differential is unlawful per se. Patent licensing agreements between competitors are sometimes struck down under antitrust law, of course, but only upon proof of an anticompetitive effect beyond that implicit in the grant of the patent. See Priest, Cartels and Patent License Arrangements, 20 J. Law & Econ. 309 (1977).

Moreover, the licensing agreement entitles the four companies licensed directly by SPS to obtain a sublicense from USM on the same terms as USM's other sublicensees. If USM's technology really were better than SPS's--enough better at any rate to make it worth their while to pay an additional one percent royalty to be able to use it--the four companies would have taken up their right to get sublicenses from USM. That USM would have been worse off if they had done so than it would have been had it negotiated a different licensing agreement with SPS in settlement of the earlier litigation is not in itself a basis for finding a violation of the antitrust laws. Those laws are solicitous not of the individual firm but of the competitive process. When, four years into the case (it was filed in 1974, and summary judgment on the misuse issue was granted in 1978), USM had presented no evidence of actual or probable anticompetitive effect, the dismissal of its misuse claim on a motion for summary judgment was proper. See Weit v. Continental Illinois Nat'l Bank & Trust Co., 641 F.2d 457, 464 (7th Cir.1981); Products Liability Ins. Agency, Inc. v. Crum & Forster Ins. Cos., 682 F.2d 660, 663 (7th Cir.1982).

The fact that the four direct licensees of SPS had the right at little or no additional cost to demand sublicenses from USM casts doubt, moreover, on USM's explanation of why SPS imposed the royalty retention differential. An alternative explanation is supplied in a deposition submitted by SPS in support of its motion for summary judgment, though not referred to by the district court. The deposition suggests that the royalty retention differential was an effort to overcome a "free rider" problem. SPS rather than USM had licensed the four companies in question and wanted a fair return on its efforts in doing so. There are costs to lining up licensees, as USM itself has emphasized in contending that 1 percent is too little to compensate it for sublicensing SPS's four direct licensees. Otherwise SPS would not allow USM to keep 75 percent of the royalties on sublicenses obtained by USM. That is compensation for USM's efforts in arranging for the use of the patent. It is overcompensation if the efforts are SPS's, as apparently was the case with the four companies in question; if, in other words, USM wants to reap where SPS has sown. In these circumstances the royalty differential would not even be "discriminatory" in any interesting sense. And antitrust law increasingly is tolerant of contractual arrangements that reduce free-rider problems and thereby increase competition (here, competition to line up patent users). See, e.g., United States Trotting Ass'n v. Chicago Downs Ass'n, Inc., 665 F.2d 781, 789 (7th Cir.1981) (en banc); Muenster Butane, Inc. v. Stewart Co., 651 F.2d 292, 297 (5th Cir.1981).

Admittedly there is irony in our recitation of the reasons that the challenged features of the licensing agreement may actually be procompetitive and in any event are not anticompetitive, when the district court found, in findings that we have not reviewed, that the patent was procured by fraud and was therefore invalid. If the patent really is invalid, and well it may be, the licensing agreement may be altogether more sinister than our discussion implies; USM and SPS are, after all, competitors. But unless we are to overrule Wikomi, which we have no mind to do, we must approach the misuse issue on the assumption that the patent is valid, for the defense of res judicata prevents USM from showing the contrary. Of course nothing we say in this opinion is intended to prejudge any other challenge that may be brought against SPS's patent; and, in any event, the patent has now expired, and can no longer restrain trade.

To sum up, we vacate the order of the district court holding SPS's patent invalid and granting USM other relief, and otherwise we affirm the orders appealed from, with costs in this court to SPS.

SO ORDERED.