The plaintiff seeks damages for violations of sections 1 and 2 of the Sherman Act arising out of the road-repair business in southern Illinois. 15 U.S.C. sec.sec. 1, 2. There are two groups of defendants: the road contractors themselves, called "applicators," and producers of the emulsified asphalt that the applicators apply to the surface of the roads. After the plaintiff, itself an applicator, settled with all three of the producers and three of the six applicator defendants, the district court granted summary judgment for the remaining applicator defendants, who are the appellees in this court.

Before going further, we note a problem of appellate jurisdiction, namely that two claims against one of the applicator defendants were dismissed without prejudice, and indeed with express leave to reinstate should this appeal fail. The circuits and indeed the cases in this court are divided over whether this form of dismissal affects the finality of the district court's judgment, but most hold, we think correctly (see also Rebecca A. Cochran, "Gaining Appellate Review by 'Manufacturing' a Final Judgment Through Voluntary Dismissal of Peripheral Claims," 48 Mercer L. Rev. 979 (1997)), that such a form of dismissal does not terminate the litigation in the district court in any realistic sense and so is not a final decision within the meaning of 28 U.S.C. sec. 1291, which authorizes the appeal of such decisions. Compare Union Oil Co. v. John Brown E & C, 121 F.3d 305 (7th Cir. 1997); Horwitz v. Alloy Automotive Co., 957 F.2d 1431, 1435-36 (7th Cir. 1992); Construction Aggregates, Ltd. v. Forest Commodities Corp., 147 F.3d 1334 (11th Cir. 1998) (per curiam); Chappelle v. Beacon Communications Corp., 84 F.3d 652 (2d Cir. 1996); Dannenberg v. Software Toolworks Inc., 16 F.3d 1073 (9th Cir. 1994); Cook v. Rocky Mountain Bank Note Co., 974 F.2d 147 (10th Cir. 1992); Ryan v. Occidental Petroleum Corp., 577 F.2d 298 (5th Cir. 1978), with United States v. Kaufmann, 985 F.2d 884, 890-91 (7th Cir. 1993); Division 241 Amalgamated Transit Union v. Suscy, 538 F.2d 1264, 1266 and n. 1 (7th Cir. 1976) (per curiam); State ex rel. Nixon v. Coeur d'Alene Tribe, 164 F.3d 1102, 1106 (8th Cir. 1999); Hicks v. NLO, Inc., 825 F.2d 118, 120 (6th Cir. 1987) (per curiam). That makes this an interlocutory appeal, and one that does not fit into any of the pigeonholes that permit such appeals in the federal courts. But when we raised this point at argument, the plaintiff's lawyer quickly agreed that we could treat the dismissal of the two claims as having been with prejudice, thus winding up the litigation and eliminating the bar to our jurisdiction. Health Cost Controls of Illinois, Inc. v. Washington, 187 F.3d 703, 708 (7th Cir. Aug. 10, 1999). And so we can proceed to the merits of the appeal.

The plaintiff presented evidence both that the applicator defendants had agreed not to compete with one another in bidding on local government contracts and that the producers had agreed not to compete among each other either, both agreements being (if proved) per se violations of section 1 of the Sherman Act. Palmer v. BRG of Georgia, Inc., 498 U.S. 46 (1990) (per curiam); Israel Travel Advisory Service, Inc. v. Israel Identity Tours, Inc., 61 F.3d 1250, 1256 (7th Cir. 1995); Metro Industries, Inc. v. Sammi Corp., 82 F.3d 839, 843-44 (9th Cir. 1996). There is a long history of bid-rigging and related practices of collusion in the road construction and road maintenance business. See, e.g., Gulf Oil Corp. v. Copp Paving Co., 419 U.S. 186 (1974); United States v. Azzarelli Construction Co., 612 F.2d 292 (7th Cir. 1979); In re Western Liquid Asphalt Cases, 487 F.2d 191 (9th Cir. 1973). (In recent years most criminal antitrust prosecutions have been of road contractors.) These are local markets, with a limited number of competitors, selling a rather standardized service to local governments constrained to give their business to the lowest bidder, a constraint that makes it easy for colluding bidders to determine whether one of their number is cheating on the agreement to divide markets. The conditions are thus ripe for effective collusion, making it unsurprising that there is evidence that the applicator defendants in fact colluded with one another to allocate the applicator business in their region. (In discussing the evidence, we emphasize that we are construing it in the light most favorable to the plaintiff, as we are required to do in reviewing the grant of summary judgment to the defendants.)

As for the producers of the asphalt used by these applicators, the record contains evidence that the product is both heavy relative to value and prone to deteriorate when transported long distances, and that as a result the practical radius within which a plant can supply applicators is only about 70 miles. This has limited to three the number of producers that can supply applicators in the region served by the plaintiff and by the applicator defendants. The plants are specialized to the production of emulsified asphalt, meaning that they can't readily be switched to producing other products. This gives the producers an incentive to produce emulsified asphalt up to the capacity of their plants (because there is no profitable use of the plants other than producing this product), and, since it is a fungible product, about the only way of increasing output is by cutting price. But since the demand for emulsified asphalt is inelastic--that is, lower prices do not yield commensurate increases in volume--the effect of price competition would be to diminish profits. So the producers, like the applicators, have much to gain by eliminating competition among themselves. And since the product is standard and the number of competing producers few, an agreement not to compete should not be too difficult to enforce; that is, at the producer level as at the applicator level, cheating should be readily observable and hence quickly checked by a retaliatory price cut. Therefore a cartel agreement would not be quickly eroded by cheating, and so again the conditions for collusion are ripe and again the record contains evidence of such collusion.

But the only defendants remaining in the case are applicators, which is to say the plaintiff's competitors. The producers have settled out. Suppose that all that the plaintiff were complaining about, therefore, was a conspiracy of the applicators to raise prices or (what amounts to the same thing) to allocate customers or markets. The complaint would fail at the threshold. A conspiracy of a firm's competitors to do these things could only help the firm, by providing an umbrella under which it could sell a large quantity at a supracompetitive price generating supracompetitive profits just by setting price in between the conspirators' price and the lower, competitive price that would prevail in the absence of the conspiracy. Israel Travel Advisory Service, Inc. v. Israel Identity Tours, Inc., supra, 61 F.3d at 1256-57. You want your competitors to charge high prices. The fact that it was in JTC's interest that the other applicators agree not to compete with each other would be irrelevant if the Department of Justice were suing the applicators, but it is not; this is a private suit by a competitor that, since it cannot show harm to itself from its competitors' eliminating competition among themselves, cannot obtain relief under the antitrust laws. Id.; Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574, 586 (1986). To put it bluntly, you cannot obtain damages without proving injury. (This is a general principle of both common law tort law and statutory tort law, rather than anything special to antitrust.) As a purchaser of emulsified asphalt, JTC could of course be hurt by a producers' cartel, but, to repeat, no producer remains as a defendant.

So what JTC has tried to show is that the applicators enlisted the producers in their conspiracy, assigning them the role of policing the applicators' cartel by refusing to sell to applicators who defied the cartel--such as JTC, which has bid for jobs that the cartel had assigned to other applicators. JTC, a maverick, was a threat to the cartel--but only if it could find a source of supply of emulsified asphalt. The claim is that the applicators got the producers to deny JTC this essential input into its business, and as a result injured it. The producer was the cat's paw; the applicators were the cat.

Matsushita Electric Industrial Co. v. Zenith Radio Corp., supra, 475 U.S. at 587, however, teaches that an antitrust claim which makes no economic sense can on that ground be dismissed on summary judgment. See also In re Brand Name Prescription Drugs Antitrust Litigation, 123 F.3d 599, 614 (7th Cir. 1997). And it might seem to make no sense from the producers' standpoint to shore up a cartel of their customers. Cartels, as we have pointed out, raise price above the competitive level and by doing so reduce the demand for their product. The less asphalt the members of the applicators' cartel sell (perhaps because the higher, cartel price induces municipalities to defer road maintenance), the less they will buy, and so the producers will be hurt. But if the producers have nowhere else to turn to sell their product, as may be the case here because of the specialized character of their plants and the limited radius within which they can ship their product from the plant, the applicator defendants may be able to coerce them into helping to police their cartel by threatening to buy less product from them or pay less for it, as in the well-known case of Eastern States Retail Lumber Dealers' Ass'n v. United States, 234 U.S. 600 (1914); see also Business Electronics Corp. v. Sharp Electronics Corp., 485 U.S. 717, 730 n. 4 (1988); United States v. General Motors Corp., 384 U.S. 127 (1966); Morrison v. Murray Biscuit Co., 797 F.2d 1430, 1438 (7th Cir. 1986); Rossi v. Standard Roofing Inc., 156 F.3d 452, 462 (3d Cir. 1998).

Alternatively, and more plausibly (at least on this record), the cartelists may have been paying the producers to perform the policing function, rather than coercing them, by threats, to do so. In re Brand Name Prescription Drugs Antitrust Litigation, supra, 123 F.3d at 614. If by refusing to sell to mavericks the producers increase the profits of the applicators' cartel, they create a fund out of which the cartel can compensate them, in the form of a higher price for the purchase of the product, for their services to the cartel. Cf. Elizabeth Granitz & Benjamin Klein, "Monopolization by 'Raising Rivals' Costs': The Standard Oil Case," 39 J. Law & Econ. 1 (1996). The record contains evidence that one of the producers obtained from applicators in the cartel area prices that were 4 to 18 (or maybe even 28) percent higher than the prices it obtained from presumably noncolluding applicators in the adjacent region, though there is no suggestion that the producer's costs were any higher in that region. The evidence is contested by the defendants, but the resolution of the contest is for trial. There is also evidence (again contested, and again this is irrelevant to whether summary judgment was properly granted) that the reasons the producers gave for refusing to sell to JTC were pretextual, as in Eastman Kodak Co. v. Image Technical Services, Inc., 504 U.S. 451, 484 (1992); McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973)--for example, that JTC was not a good credit risk, even though when JTC offered to pay cash the producers still refused to sell to it. This suggests that the real reason for the refusal was one that the producers didn't want to acknowledge--namely that they were being compensated by the cartel for refusing to sell to a customer whom otherwise they would have been happy to sell to. The combination of the price difference with the evidence of pretext support an inference that the producers were being compensated by the applicators for shoring up the cartel by boycotting an applicator that was competing with the cartel. If so--if the producers were working for the cartel--they were part of the applicators' conspiracy, and for the injury that they inflicted on JTC as agents of the applicators' cartel by denying JTC a source of supply the members of the cartel, three of which are the remaining defendants, would be culpable under elementary principles of both conspiracy law and agency law. E.g., Valley Liquors, Inc. v. Renfield Importers, Ltd., 678 F.2d 742, 743 (7th Cir. 1982); Rossi v. Standard Roofing Co., supra, 156 F.3d at 472.

There may be an innocent explanation for why producers would charge lower prices elsewhere or why they refused to sell to JTC. But the only issue for us, in reviewing the grant of summary judgment for these defendants, is whether a rational jury, having before it the evidence developed to date, could conclude (construing the evidence as favorably to the plaintiff as the record permits) that the reason for the producers' refusal to deal with JTC was that they were in cahoots with the cartel to discourage competition in the applicator market. Given the evidence of cartelization at both the applicator and producer level, the suspicious producer price behavior (suggestive of the producers' having been "paid off" by the cartel to boycott JTC and other upstarts), and the pretextual character of the reasons the producers gave for the refusal to deal, a rational jury could conclude that JTC was indeed the victim of a producers' boycott organized by the applicator defendants. All the evidence that we have discussed is circumstantial, but of course an inference of conspiracy--of, in this case, an informal agreement among the applicators and the producers to deny supply to firms that tried to break into the applicators' cosily divided market--can be drawn from circumstantial evidence as well as from admissions or other direct testimony of the conspirators' communications with each other. Monsanto Co. v. Spray-Rite Service Corp., 465 U.S. 752, 764 (1984); Market Force Inc. v. Wauwatosa Realty Co., 906 F.2d 1167, 1171-73 (7th Cir. 1990); Petruzzi's IGA Supermarkets, Inc. v. Darling-Delaware Co., 998 F.2d 1224, 1233 (3d Cir. 1993). JTC has some direct evidence as well. It strikes us as equivocal, and we have not thought it necessary to discuss it; but we do not mean to suggest that it should not be admitted at the trial to which, we conclude, JTC was entitled.

We have finally to consider whether JTC is entitled to proceed to trial on its section 2 claims as well. One claim is that the defendants engaged in a conspiracy to monopolize the local applicator market. See, e.g., Great Escape, Inc. v. Union City Body Co., 791 F.2d 532, 540-41 (7th Cir. 1986). This is a puzzling offense, as it seems entirely duplicative of a claim under section 1 of a conspiracy to restrain trade. IIIA Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law: An Analysis of Antitrust Principles and Their Application para. 809, pp. 370-71 (1996). But there are cases with facts like those of the present case in which section 2 claims have been allowed, e.g., Full Draw Productions v. Easton Sports, Inc., 182 F.3d 745, 753, 756-57 (10th Cir.1999); Movie 1 & 2 v. United Artists Communications, Inc., 909 F.2d 1245, 1247-48, 1255-56 (9th Cir. 1990), and the appellees here do not argue that the jury would be confused to be instructed under section 2 as well as under section 1. Cf. Gehring v. Case Corp., 43 F.3d 340, 343 (7th Cir. 1994); McDonald v. Sandvik Process Systems, Inc., 870 F.2d 389, 395 (7th Cir. 1989).

The second section 2 claim pressed by JTC is that the defendants attempted to monopolize the local applicator market. Attempt to monopolize differs from conspiracy to monopolize primarily in dispensing with the requirement of an agreement, and the usual case therefore is one in which the defendant has or seeks to obtain monopoly power all by itself. E.g., Elliott v. United Center, 126 F.3d 1003 (7th Cir. 1997). There is no suggestion of that here. JTC's argument rather is that the defendants were acting jointly to exclude JTC from the market; there is no evidence that had any single one of the defendants boycotted a producer who sold emulsion to JTC, the producer would have kowtowed. So this theory of violation merges with JTC's last and most audacious theory, that section 2 can be violated by purely tacit collusion. Each of the defendants knew that if all of them boycotted a producer who supplied JTC, that producer would knuckle under; it was thus in their common interest to act jointly. But if they acted jointly in this manner without any agreement to do so, it would be novel to characterize their behavior as conspiratorial within the meaning of either section 1 or section 2 of the Sherman Act.

The question whether purely tacit collusion, or as it is sometimes called oligopolistic interdependence, might violate the Sherman Act has been much mooted in academic circles, as discussed in IIIA Areeda & Hovenkamp, supra, para.para. 810 et seq. A number of cases treat it as open. Morgenstern v. Wilson, 29 F.3d 1291, 1295 n. 2 (8th Cir. 1994); Harkins Amusement Enterprises, Inc. v. General Cinema Corp., 850 F.2d 477, 490 (9th Cir. 1988); see also Springfield Terminal Ry. v. Canadian Pacific Ltd., 133 F.3d 103, 108 (1st Cir. 1997). We cannot find a case that squarely supports the theory; the Second Circuit appears to have rejected it, H.L. Hayden Co. v. Siemens Medical Systems, Inc., 879 F.2d 1005, 1018 (2d Cir. 1989); Crimpers Promotions Inc. v. Home Box Office, Inc., 724 F.2d 290, 291 n. 1 (2d Cir. 1983) (Friendly, J.); we threw some cold water on it in Indiana Grocery Inc. v. Super Value Stores, Inc., 864 F.2d 1409, 1416 (7th Cir. 1989); and the Ninth Circuit rejected it outright in Rebel Oil Co. v. Atlantic Richfield Co., 51 F.3d 1421, 1443 (9th Cir. 1995). We cannot agree that it derives any support at all from the fact that section 2, unlike section 1 (which is limited to combinations, contracts, and conspiracies), does not require an agreement (unless the specific charge is conspiracy to monopolize). If oligopolistic interdependence can be described as "joint monopolization" or "attempted joint monopolization" (JTC's terms), equally it can be described as a combination or a (tacit) conspiracy.

The most compelling objection to JTC's theory has nothing to do with the language of the Sherman Act but rather is the difficulty of formulating effective relief without transforming the district court into a regulatory agency, here for example charged with compelling producers of emulsion to sell to JTC. At all events the theory is a novel one and when a litigant wants a court to buy a novel theory it has to do more than assert it, wholly ignoring the objections that have been made to it and the cases that have questioned or rejected it. Huntzinger v. Hastings Mutual Ins. Co., 143 F.3d 302, 308 n. 7 (7th Cir. 1998); Colburn v. Trustees of Indiana University, 973 F.2d 581, 593 (7th Cir. 1992); Turner v. Chicago Housing Authority, 969 F.2d 461, 463 (7th Cir. 1992); Burdett v. Miller, 957 F.2d 1375, 1382 (7th Cir. 1992). The Second Circuit has put it well: "We have no obligation to review issues that are raised, but not independently and sufficiently developed, in an appellant's main brief." Karibian v. Columbia University, 14 F.3d 773, 777 n. 1 (2d Cir. 1994). Indeed not; and so JTC will have to prove an agreement in order to prevail on remand.

Reversed.