This is an appeal from the granting of a preliminary injunction that prevents the LTV Corporation from proceeding with a tender offer to acquire, in stages, all of the shares of Grumman Corporation. The District Court for the Eastern District of New York (Jacob Mishler, Judge) issued the injunction after finding that Grumman had proved a likelihood of success on its claims that the proposed acquisition would violate § 7 of the Clayton Act, 15 U.S.C. § 18 (1976), and that LTV's Offer to Purchase violates the Williams Act, §§ 14(d) and 14(e) of the Securities Exchange Act of 1934, as amended, 15 U.S.C. §§ 78n(d) and 78n(e) (1976). Grumman Corp. v. LTV Corp., --- F.Supp. ---- (E.D.N.Y.1981). We agree that the preliminary injunction was properly issued on antitrust grounds and therefore affirm, without reaching the Williams Act claims.1

On September 24, 1981, LTV commenced a tender offer for the acquisition by CKH Corporation, its wholly owned subsidiary, of common stock, convertible preferred stock, and convertible subordinated debentures of Grumman, representing or convertible into approximately 70% of Grumman's outstanding voting shares. The price was $45 per share of common stock, approximately double the market price of Grumman stock during the week preceding the offer. LTV also announced plans subsequently to acquire the remaining 30% of Grumman stock. Grumman initiated this litigation on September 28. Following limited discovery, including 19 depositions, a limited evidentiary hearing on the motion for a preliminary injunction was held on October 6 and 7. Judge Mishler rendered a detailed opinion on October 14. Pursuant to an expedited appeal, we heard oral argument on October 28.

Mindful that target companies are quick to seek refuge in § 7 against an unfriendly takeover, see Missouri Portland Cement Co. v. Cargill, Inc., 498 F.2d 851, 854 (2d Cir.), cert. denied, 419 U.S. 883, 95 S. Ct. 150, 42 L. Ed. 2d 123 (1974), we also recognize that when a takeover threatens horizontal integration, the courts have a clear duty to weigh carefully the claim for a preliminary injunction, F. & M. Schaefer Corp. v. C. Schmidt & Sons, Inc., 597 F.2d 814 (2d Cir. 1979) (per curiam ); Gulf & Western Industries, Inc. v. Great Atlantic & Pacific Tea Co., 476 F.2d 687 (2d Cir. 1973); Allis-Chalmers Mfg. Co. v. White Consolidated Industries, Inc., 414 F.2d 506 (3d Cir. 1969), cert. denied, 396 U.S. 1009, 90 S. Ct. 567, 24 L. Ed. 2d 501 (1970), even though the true concerns of the "private attorney general" may be more "private" than "attorney general." If the effect of a proposed takeover may be substantially to lessen competition, the target company is entitled to fend off its suitor. Our focus is therefore not upon Grumman's motivation for bringing this suit, but upon the adequacy of its preliminary showing that the proposed takeover will violate § 7.

Judge Mishler ruled that Grumman had shown the requisite probability of success on its § 7 claims with respect to three distinct product markets: the carrier-based aircraft market, the major airframe subassembly market, and the nacelle (aircraft engine structure) market. For purposes of this appeal, LTV does not contest the appropriateness of these three product markets. Its primary contention is that the District Court misapprehended a likelihood of diminished competition within these markets, and it also contends that the scope of the markets was incorrectly defined. LTV does not dispute that during the past two decades Vought Corporation (LTV's subsidiary) and Grumman have been substantial competitors in the production of airplanes and aircraft components. The issue is whether there currently exists a prospect of significant future competition in the three relevant product markets and whether the nature of those markets is such that the proposed acquisition will probably have an unlawful anti-competitive effect.

1. Carrier-Based Aircraft. As both sides point out, the carrier-based aircraft market has unusual characteristics. In this country there is only one buyer, the Defense Department. It decides which companies will be invited to submit bids for new planes. Bidding competitions occur infrequently and entail enormous costs. Grumman's evidence indicated that development of a proposal for the VTX, a carrier-based training aircraft, would cost more than $30 million. As for sales of currently available models, the District Court accepted Grumman's evidence that in the past three years the total for carrier aircraft was approximately $11 billion, divided among only four manufacturers: Grumman, 41.5%, Vought (alone and with Lockheed), 7.7%, and McDonnell Douglas, 50.8%. Plainly, this market is a "tight oligopoly," Stanley Works v. Federal Trade Commission, 469 F.2d 498, 504 (2d Cir. 1972), cert. denied, 412 U.S. 928, 93 S. Ct. 2750, 37 L. Ed. 2d 155 (1973), in which it is important to prevent "even slight increases in concentration." United States v. Continental Can Co., 378 U.S. 441, 462, 84 S. Ct. 1738, 1749, 12 L. Ed. 2d 953 (1964).

LTV challenges the Court's assignment to Vought of a 7.7% market share. That share represents $734 million in sales of the A-7 and $110 million in sales of the S-3 during the past three years. Apparently not disputing $533 million in revenue from A-7's in 1979 and 1980, LTV challenges the $201 million figure for 1981 on the ground that $112.6 million of the 1981 total represents A-7K's, which are land-based trainers for the Air National Guard, and $79.5 million represents modifications of existing aircraft. Grumman responds that the A-7K's are properly included since the planes are designed for carrier use and their sale keeps in operation a production line for carrier-based planes. Grumman also contends that modifications of existing aircraft are properly included within the carrier-based aircraft market. LTV also challenges inclusion of $110 million for the S-3's, since Lockheed is the prime contractor for these planes. Grumman responds that Vought and Lockheed team up to produce the S-3's and that Vought's experience with carrier-based planes was critical to the awarding of the contract. While the challenges to the 7.7% share may merit further exploration during a plenary trial, we are satisfied that the Vought share of the carrier-based aircraft market was properly found to be of sufficient significance reasonably to apprehend a lessening of competition in the event of a Vought-Grumman combination.

LTV's more fundamental challenge is that whatever the record shows as to past sales, there is an insufficient basis to believe that Vought will be a competitive factor in the future. LTV points out that the Navy's last order for A-7's was submitted in 1978 and that the Defense Department has no current plans to purchase A-7's from Vought. Judge Mishler did not ignore these facts, but concluded nonetheless that Vought can reasonably be expected to provide competition in the carrier-based aircraft market. Obviously, the District Court could not say with certainty that the Defense Department will resume purchase of A-7's at a precise date or in a precise amount. The Court noted several factors that increase the likelihood of such purchases: problems associated with the A-18 aircraft produced by McDonnell Douglas, the Navy's plans to increase the number of carriers in operation, Vought's own marketing strategy, and the heightened tensions in the Middle East.

LTV contends the Court was impermissibly considering mere possibilities. To the extent that this criticism is valid, it simply reflects an inevitable aspect of an unusual market. Carrier-based planes do not roll off assembly lines like television sets or automobiles. In a market with a single domestic purchaser, which buys intermittently, a court assessing the anti-competitive effect of a horizontal combination must consider future possibilities in assessing whether there exists a significant probability of decreased competition. Whether or not Vought will sell more A-7's to the Defense Department, the fact remains that it was properly found to be competing to do so. It hopes to keep its A-7 production line in operation by selling more A-7K's to the Air National Guard. It is attempting to sell A-7's abroad. As recently as June of this year it had pending with the Navy a proposal for a modified version of the A-7, the A-7X. The Navy's rejection of the proposal does not lessen the significance of Vought's capacity and desire to make it. "Unsuccessful bidders are no less competitors than the successful one." United States v. El Paso Natural Gas Co., 376 U.S. 651, 661, 84 S. Ct. 1044, 1049, 12 L. Ed. 2d 12 (1964). In a planning document Vought submitted to LTV's board of directors just prior to the tender offer, Vought characterized itself as "STRONG" in the naval tactical aircraft market.

Assessing all of the evidence before him, Judge Mishler concluded that in light of the history, structure, and probable future of the market for carrier-based aircraft, a merger between Grumman and Vought might tend to substantially lessen competition. We are satisfied that Grumman established at least the requisite probability of success in proving this proposition at trial to warrant a preliminary injunction.

2. Major Airframe Subassemblies. The District Court determined that there existed a relevant market for the sale of major airframe assemblies for large civilian aircraft. The key items sold in this market are the fuselage, wings, and empennage (tail assembly) of an airplane. The Court found the market limited to seven manufacturers, of which Grumman and Vought were considered to be two of the strongest competitors. The Court determined that Vought currently has a 15% share of a $1 billion market and is projected to have a 14.3% share of a $1.5 billion market by 1984; comparable figures for Grumman are 9.8% and 32%. The Court found that Grumman and Vought have engaged in head-to-head competition for sections of the fuselage for the Boeing 757 and the empennage for the Lockheed CX and that a merger would substantially weaken competition in the market.

LTV contends that the District Court erred in calculating both the aggregate market and the respective shares of Grumman and Vought, and that the actual competition between the two companies affects only an insubstantial percentage of the airframe subassembly market. First, LTV objects to the exclusion of export sales to foreign manufacturers. We see no reason why the District Court may not assess the effect of the proposed merger upon the market consisting of sales to United States purchasers.

Somewhat more substantial is the claim that the Court should have included in the market the subassembly work that the major prime contractors, Boeing, Lockheed, and McDonnell Douglas, perform for themselves. Judge Mishler excluded this work because the major contractors never award subcontracts to each other. LTV responds that the subcontractors, including Grumman, regard the prime contractors as competitors for the subassembly work in the important sense that the subcontractors must meet the price at which the prime contractors could do the work.

Courts have divided on whether in-house activity should be included in market measurement. Compare Spectrofuge Corp. v. Beckman Instruments, Inc., 575 F.2d 256, 278, 285-86 (5th Cir. 1978) (including in-house servicing of medical instruments with services rendered by independent organizations), cert. denied, 440 U.S. 939, 99 S. Ct. 1289, 59 L. Ed. 2d 499 (1979), with United States v. Blue Bell, Inc., 395 F. Supp. 538, 543 (M.D.Tenn.1975) (excluding, from market for sale of industrial rental garments to laundries, garments sold by manufacturers to their own affiliated laundries). Since the ultimate issue is whether competition may be substantially lessened in the "area of effective competition," United States v. E.I. du Pont de Nemours & Co. (General Motors), 353 U.S. 586, 593, 77 S. Ct. 872, 877, 1 L. Ed. 2d 1057 (1957), the appropriateness of including in-house activity must turn on an assessment of the economic realities of the competitive situation. For example, the contractors may be unlikely to decrease subassembly work to take advantage of subcontractors' price reductions if the in-house work is done primarily to maintain workforce and production line activity. And the contractors may lack the capacity to increase their in-house work in the event that subcontractors' prices rise. In such circumstances, the in-house work would not be in the area of "effective" competition. On the other hand, under other circumstances the decision of each prime contractor to set the level of in-house work may be highly responsive to price changes by subcontractors, in which event the fact that the primes will not do work for each other would have only a slight bearing on effective competition for subassembly work, including work done in-house.

While the limited inquiry conducted at the hearing on the preliminary injunction did not permit full exploration of the economic realities of the in-house activity, the District Court did find from the expert testimony that the prime contractors "are not capable of constructing all portions of their major airframe assemblies (the fuselage, the wings and the empennage) in house." (Emphasis in original). This finding suffices, at this stage of the litigation, to exclude in-house subassembly production, although the resulting market shares of the subcontractors must nevertheless be evaluated in light of whatever competitive effect is provided by the in-house activity.

LTV also challenges the exclusion of subassembly work performed on United States aircraft by foreign subcontractors. Unrebutted evidence showed that such subcontracts are awarded as a precondition for selling planes to airlines controlled by foreign governments. Judge Mishler was entitled to conclude that such work was not realistically available for competitive bidding and therefore properly excluded from the market definition.

Finally, LTV contends that the District Court distorted the Grumman subassembly volume and market share by aggregating volume for both major subassemblies and nacelles, even though the latter were considered to be in a separate submarket. The point seems to be well taken, but appropriate adjustments, made from the same data credited by the District Court,2 would not alter the significance of the Court's ultimate conclusions. In a 1981 subassembly market of $700 million, Vought, which sold no nacelles, would have a 21% share, although the Grumman share would be only 5%. However, in a 1984 market of $1 billion, Vought's projected sales of $214 million of subassemblies would represent a 21% share,3 and Grumman's projected sales of $182 million would represent an 18% share. Judge Mishler's concern for the effect upon competition of a merger of subcontractors with these market shares remains entirely valid. LTV's point that Vought and Grumman have submitted bids on the same subassembly contracts in only eight instances in the last five years, totaling a not insignificant $251 million, ignores the competitive effect they exert simply by being available to compete. Vought was not misleading LTV's board when it characterized its own position in the market as "VERY STRONG" and Grumman's as "GOOD."

3. Nacelles. The District Court determined that the market for subcontracting of nacelles currently totals $350-$500 million. At present Rohr Corporation dominates the market with a share estimated at between 60-80%. The Court acknowledged that neither Grumman nor Vought has a significant share of the market. Grumman's 1981 nacelle revenue is estimated at $63 million, and Vought anticipates no revenue from nacelles in 1981.

The Court's assessment of the anti-competitive effect of a Grumman-Vought merger focused quite properly on the "probable future" of the market, United States v. General Dynamics Corp., 415 U.S. 486, 498, 94 S. Ct. 1186, 1194, 39 L. Ed. 2d 530 (1974). The market is expected to experience rapid growth because of the demand for fuel-efficient engines. Vought's internal documents estimate a market of.$2.7 billion of nacelle sales for aircraft produced in the United States by 1996. Though Vought currently derives no revenue in this market, its status as an active competitor is indicated by its present contract to manufacture nacelles for the Canadair CL-600 and by its aggressive plans for the future. LTV exhibits excessive modesty in this litigation in emphasizing Vought's current minor role in the market, while Vought's internal documents state that in the expanding market of the future Vought's only two "viable nacelle competitors" will be Rohr and Grumman. Rohr's present dominance is expected to decline because its principal customer, Boeing, has set up its own nacelle facility. Vought is aiming to win the re-engining program for the Boeing 727-200, which is expected to be subcontracted, and in that contest Vought sees Grumman as the "major competition." In fact, from Vought's perspective, Grumman has "competed against us for every opportunity." Looking to the future in the expanding market, Vought intends to become "the No. 2 nacelle supplier by 1986."

In a concentrated market expected to see total volume grow and the share of the dominant firm decrease, the District Court was properly concerned with maintaining small aggressive competitors in the market. See United States v. Aluminum Co. of America, 377 U.S. 271, 280-81, 84 S. Ct. 1283, 1289-1290, 12 L. Ed. 2d 314 (1964); Stanley Works v. Federal Trade Commission, supra, 469 F.2d at 507-08. Judge Mishler was entitled to conclude that removing one competitor from this market would tend to substantially lessen competition.

4. Appropriateness of a Preliminary Injunction. Relying on Buffalo Forge Co. v. Ampco-Pittsburgh Corp., 638 F.2d 568 (2d Cir. 1981), and Missouri Portland Cement Co. v. Cargill, Inc., supra, LTV contends that the District Court misapplied the standards for issuance of a preliminary injunction. LTV urges that Grumman has not shown any likelihood of irreparable injury and that the equities favor allowing the tender offer to proceed, especially the concern of the Grumman shareholders who stand to reap a 100% gain.

We note at the outset that Buffalo Forge was an affirmance of the denial of a preliminary injunction sought to bar a tender offer. The District Court in that case had found the § 7 claim to be "speculative, theoretical, and weak." Id. at 569. While Missouri Portland reversed the grant of a preliminary injunction, Judge Friendly's opinion pointedly emphasized that the case "differs totally from the horizontal merger, illustrated by United States v. Philadelphia National Bank, 374 U.S. 321, 83 S. Ct. 1715, 10 L. Ed. 2d 915 (1963)." 498 F.2d at 858; see also id. at 869: "Here again it must be emphasized that, in contrast to a case where the acquirer is a competitor of the target company, Cargill (the acquirer) would have no interest other than the promotion of MP's (the target's) welfare."

In assessing the appropriateness of a preliminary injunction pursuant to § 16 of the Clayton Act, 15 U.S.C. § 26 (1976), the District Court conscientiously considered the harm to Grumman, as well as the equities of other private interests and the overriding public interest. In doing so, the Court faithfully weighed the factors identified in Gulf & Western Industries, Inc. v. Great Atlantic & Pacific Tea Co., supra, a case, like this one, over which "the atmosphere of horizontal integration hung heavily." Missouri Portland Cement Co. v. Cargill, Inc., supra, 498 F.2d at 868. With respect to the interests of Grumman, Judge Mishler noted that the company's president had been with the company for 35 years and was concerned that "efficiencies which stem from that unique association ... would be destroyed." While such "jitters in executive suites," Missouri Portland, supra, 498 F.2d at 869 n.36, are commonplace and not necessarily weighty, we cannot disregard the District Court's ultimate conclusion that consummation of LTV's tender offer "would seriously disrupt Grumman's business." The Court expressed particular concern about the risk that LTV would liquidate Grumman's nonaerospace components.

Moreover, apart from the inevitable consequences of a takeover, such as gaining access to the target's confidential information, e. g., Allis-Chalmers Manufacturing Co. v. White Consolidated Industries, Inc., supra, 414 F.2d at 516, irreparable damage exists here because of the high probability that Grumman will cease to be an effective competitor in three aerospace product markets of vital interest to the nation's economy and defense. Of course the damage from diminished competition is borne most immediately by the buyers, the Defense Department, with respect to carrier-based planes, and airplane manufacturers, with respect to airframe subassemblies and nacelles. If free to combine, two competitors might be expected to prefer the advantages of diminished competition. But in reality it is only the resulting entity that would enjoy such advantages. The target company is entitled under § 16 to preserve its separate existence as a competitor.4 And it is entitled to obtain preliminary relief if it can show, as Grumman did here, that the threat to the public interest from the loss of competition is serious and not likely to be undone by a divestiture in the event the acquisition is found to be unlawful after it has occurred. Judge Mishler carefully assessed the public interest in maintaining Grumman and Vought as competing entities and found that it was "greater than in the ordinary case."

We need not assess the parties conflicting claims as to the reality of renewing the tender offer in the event that a permanent injunction is denied upon a plenary trial. Compare Missouri Portland, supra, 498 F.2d at 870 n.38 (noting that the preliminary injunction ended the tender offer in Gulf & Western Industries ), with Ronson Corp. v. Liquifin Aktiengesellschaft, 370 F. Supp. 597 (D.N.J.) (offer renewed when preliminary injunction vacated after trial), aff'd, 497 F.2d 394 (3d Cir.), cert. denied, 419 U.S. 870, 95 S. Ct. 129, 42 L. Ed. 2d 108 (1974). The Grumman shareholders are not entitled to a gain obtained from a sale that presents a substantial likelihood of violating § 7.

The issuance of a preliminary injunction is affirmed.