BYNUM v. FMC CORP

RANDALL, Circuit Judge:

Plaintiff-appellant Daniel Edward Bynum brought this product liability action in district court seeking damages for injuries sustained when a M-548 cargo carrier in which he was riding fell off a bridge and crashed into the creek below. At the time of the accident, Bynum was a member of the Mississippi National Guard taking part in a seventeen-day training mission at Fort Stewart, Georgia. The district court, 599 F. Supp. 155, holding the government contractor defense applicable under Mississippi law, granted summary judgment in favor of defendant-appellant FMC Corporation, and Bynum appeals. On appeal, we do not address the merits of the district court's Erie guess. Rather, for the reasons given below, we hold that, at least under the circumstances presented here, federal common law provides the basis for FMC Corporation's assertion of the government contractor defense. Because all the elements of the defense have been established through either uncontroverted affidavit or the stipulations of the parties, we affirm the summary judgment.

I. FACTUAL AND PROCEDURAL BACKGROUND.

On July 13, 1978, Bynum, then a member of the Mississippi National Guard, was participating in a tactical road march at Fort Stewart, Georgia, as part of a training mission for the Mississippi National Guard. Bynum's role in the march was to ride in the M-548 cargo carrier ("M-548"), a tracked United States Army vehicle, as an air gunner. While crossing a bridge, the M-548 swerved to the left and plunged into the creek bed below. As a result of the accident, Bynum sustained severe injuries, including two broken legs caused when the M-548's turret gun rotated freely during the fall.

Thereafter, on May 4, 1982, Bynum filed the instant complaint in the District Court for the Northern District of Mississippi, alleging negligence, breach of express and implied warranties, and strict liability theories of recovery. Defendant FMC Corporation ("FMC") answered and initiated discovery, which continued for over a year until terminated by the district court on August 4, 1983. At that point, FMC moved for summary judgment on the ground that the government contractor defense barred the action. In support of its motion, FMC relied primarily on the deposition testimony of Bynum's expert witness, Dr. Malcolm Newman, and the affidavit of William H. Stites, who was the Engineering Manager of FMC's Ordnance Division. Bynum filed no evidentiary response to the summary judgment motion.

General Motors Corporation ("GM") was also an original defendant in this action. GM was granted summary judgment on June 27, 1983, and that order has not been appealed.

FMC also sought summary judgment on a variety of other grounds, none of which are relevant here. The district court granted summary judgment to FMC with respect to Bynum's warranty claims. Bynum does not challenge that order on appeal.

At his deposition, Newman testified that he knew of no evidence of negligence or negligent manufacture by FMC. Rather, Newman stated, the accident was most probably caused by design defects in the tracking system and the gun-locking mechanism and a failure to warn as to the proper manner in which to operate the vehicle under the conditions existing at the time of the accident. Deposition of Malcolm Newman, at 66-74. In his affidavit, Stites described the function and capabilities of the M-548 as follows:

Newman's deposition has been made a part of the record on appeal by stipulation of the parties.

The M548 is a tracked vehicle that is used by the United States Army in a variety of ways. Its primary function is to carry ammunition both for tanks and for self-propelled artillery. The vehicle must be capable of carrying 6 tons of ammunition while traveling across all types of terrain in all types of weather. It follows behind the tanks or large cannon, such as the 155 Howitzer, mounted on tracks. The vehicle has to be air droppable, that is, capable of being dropped by parachute from airplanes; so it must have outside dimensions that allow it to fit inside the airplanes from which it would be dropped. The vehicle also must be amphibious. Not only must it be able to float when it enters the water while carrying its load of ammunition; but also it must be able to propel itself through the water.

Record at 316. Stites also explained that the specifications for the M-548 were the work product of the United States Army Tank-Automatic Command ("TACOM"), which was responsible for the designing and testing of wheeled and tracked vehicles procured by the government for military use. According to Stites, TACOM supplied FMC with a technical data package containing over 2500 sheets of detailed drawings that were to be used in the manufacture of the M-548. Under the contract, Stites stated, FMC was obligated to comply strictly with these design specifications in producing the M-548. If FMC desired to alter any particular portion of the design, the proposed modification had to be submitted to TACOM for review and testing. Any unapproved deviation from the government's specifications would result in rejection of the vehicles by the government's inspectors who, in accordance with the contract, worked full time at the FMC plant.

Finally, Stites stated in his affidavit that the particular vehicle that was involved in the accident had been manufactured in strict compliance with the government's plans and specifications. The type of event that Newman described as being the most likely cause of the accident, Stites asserted, had never before occurred within the knowledge of FMC.

On February 1, 1984, Chief Judge L.T. Senter, Jr., denied FMC's motion for summary judgment on the ground that the government contractor defense had not been recognized or adopted by any controlling precedent. The case was then transferred to Judge Biggers on April 11, 1984. At the beginning of trial, on October 1, 1984, Bynum moved in limine to prevent FMC from introducing any evidence concerning the government contractor defense. In response, FMC orally renewed its motion for summary judgment based on our intervening decision in Hansen v. Johns-Manville Products Corp., 734 F.2d 1036, 1045 (5th Cir. 1984), cert. denied, ___ U.S. ___, 105 S.Ct. 1749, 84 L.Ed.2d 814 (1985). After reviewing the record in light of that opinion, Judge Biggers stated in chambers that he was inclined to recognize and apply the defense in the instant case. That announcement evidently prompted the parties to enter into the following two stipulations:

See infra Part III(A)(3).

(1) [The district court] has the authority to reconsider a previous ruling on a motion for summary judgment; and

(2) The vehicle in issue in the present case was manufactured in accordance with precise design specifications furnished by the United States Government, the vehicle conformed to the specifications, and the manufacturer FMC knew of no patent dangers in the vehicle that were not known to the United States Government.

Bynum v. General Motors Corp., 599 F. Supp. 155, 156 (N.D.Miss. 1984).

Based on these stipulations, as well as on the pleadings, affidavits, memoranda, and other submissions of the parties, Judge Biggers granted summary judgment. In the published opinion accompanying the order, Judge Biggers explained that the basis of his decision was an Erie guess that Mississippi courts would apply the government contractor defense, as set forth in Hansen, if and when such a situation was presented to them. Judge Biggers found that by their stipulations the parties established that the following three prerequisites for the application of the defense were met: (1) establishment of product specifications by the government, (2) manufacture in accordance with those specifications, and (3) government knowledge equal to that of the manufacturer of the hazards. Id. at 158.

From the order granting summary judgment, Bynum appeals.

II. LEGAL BACKGROUND.

Before we examine the legal basis of the modern government contractor defense and its applicability to the case at hand, a brief overview is in order of the defense's historic analogues and the reasons provided by federal and state courts for the adoption of the modern defense. Toward this end, we discuss first the Feres-Stencel doctrine, which prompted the formulation of the defense. Then we examine the defenses traditionally available to manufacturers complying with design specifications supplied by the government and the deficiencies of those defenses in the current context. Finally, we review the justifications upon which recently adopted versions of the modern government contractor defense have been based.

A. The Feres-Stencel Doctrine.

In Feres v. United States, 340 U.S. 135, 71 S.Ct. 153, 95 L.Ed. 152 (1950), the Supreme Court held that the Federal Tort Claims Act ("FTCA"), 28 U.S.C. §§ 1346(b), 2671 et seq., did not waive the government's sovereign immunity with respect to "injuries to servicemen where the injuries arise out of or are in the course of activity incident to service." Id. at 146, 71 S.Ct. at 159. Although this issue was clearly one of statutory construction, because nothing in the history of the Act was directly on point, the Supreme Court focused on the policies of the Act and the consequences of allowing such a suit. Specifically, the Court gave three reasons for barring a serviceman's suit against the government to recover for service-related injuries. First, the Supreme Court found that "[t]he relationship between the government and members of its Armed Forces is `distinctively federal in character.'" Id. at 143, 71 S.Ct. at 158. By this, the Court meant that "the scope, nature, legal incidents and consequences of the relation between persons in service and the government are fundamentally derived from federal sources and governed by federal authority." Id. at 143-44, 71 S.Ct. at 158 (citations omitted). Consequently, the Court reasoned, it would make no sense in providing for those disabled in service "to leave them dependent upon geographic considerations over which they have no control and to laws which fluctuate in existence and value." Id. at 143, 71 S.Ct. at 158. Second, the Court observed that the purpose of the Act was not to create new causes of action but to allow suit against the government in those circumstances when, in an analogous context, a claim was traditionally recognized against a private defendant. Therefore, the Supreme Court concluded, because the military context has no similar counterpart in the private sector, the Act by itself could not be construed as authorizing the action. Id. at 141-42, 71 S.Ct. at 157. Finally, the Court noted that the government had already adequately provided for injured servicemen by enacting the Veterans' Benefits Act, 38 U.S.C. §§ 321 et seq., which establishes a no-fault military compensation scheme. "The compensation system, which normally requires no litigation, is not negligible or niggardly. . . . The recoveries compare extremely favorably with those provided by most workman's compensation statutes." Id. at 145, 71 S.Ct. at 159.

See 28 U.S.C. § 1346(b) (providing that the law of the place where the actionable act or omission occurred applies in actions brought under the FTCA).

An additional reason for the holding in Feres was suggested in United States v. Brown, 348 U.S. 110, 75 S.Ct. 141, 99 L.Ed. 139 (1954). In Brown, the Supreme Court expressed concern that allowing Feres-type actions alleging negligence on the part of military authorities would undermine military discipline and impair the effectiveness of the armed services. Said the Court:

The peculiar and special relationship of the soldier to his superiors, the effects of the maintenance of such suits on discipline, and the extreme results that might obtain if suits under the [Federal] Tort Claims Act were allowed for negligent orders given or negligent acts committed in the course of military duty, led the [ Feres] Court to read that Act as excluding claims of that character.

Id. at 112, 75 S.Ct. at 143.

The Supreme Court carried its Feres decision one step further in Stencel Aero Engineering Corp. v. United States, 431 U.S. 666, 97 S.Ct. 2054, 52 L.Ed.2d 665 (1977). In that case, a National Guardsman sued the manufacturer of a faulty ejection system for injuries sustained when the egress life-support system of his fighter aircraft malfunctioned during a midair emergency. The manufacturer in response filed a third-party claim against the United States. The sole issue before the Supreme Court was whether the FTCA waived the government's immunity from suit by a manufacturer seeking indemnity for any damages it may be required to pay to a National Guardsman for service-related injuries.

The Supreme Court began its consideration of this question by reviewing its holding in Feres. The Court found that its Feres decision was based on essentially three factors: (1) the distinctive federal character of the relationship between the government and members of the armed services, (2) the availability of "generous pensions to injured servicemen" through the Veterans' Benefits Act, and (3) the effect that a suit by a member of the armed services against the government would have on discipline. Id. at 671, 97 S.Ct. at 2057. The Court then framed the issue presented in the case as whether these factors were similarly implicated in a third-party indemnity action brought by a manufacturer against the government arising from service-related injuries. With respect to the first factor, the Court stated that "the relationship between the government and its suppliers of ordnance is certainly no less distinctively federal in character than the relationship between the government and its soldiers." Id. Thus, according to the Court, it would make little sense to permit the fortuity of the situs of the alleged negligence to affect the liability of the government to the military contractor. Id. Second, the Court found that the Veterans' Benefits Act creates an upper limit of liability for the federal government as to service-connected injuries. As a result, to permit an indemnity action for this type of injuries "would be to judicially admit at the back door that which has been legislatively turned away at the front door." Id. at 673, 97 S.Ct. at 2059 (citations omitted). Finally, the Stencel Court pointed out that the litigation spawned by the indemnity claim would take virtually the identical form as that of litigation on a primary claim directed against the government. "The trial would, in either case, involve second-guessing military orders, and would often require members of the Armed Services to testify in court as to each others' decisions and actions." Id. Thus, because the considerations underlying the Feres decision were largely applicable to indemnity actions as well, the Supreme Court held that the right of a third party to recover in an indemnity action against the United States was limited "by the rationale of Feres." Id. at 674, 97 S.Ct. at 2059.

Bynum argues that in Stencel the Supreme Court also suggested that a member of the armed services could sue the manufacturer for defects in government designs of military equipment. In making this argument, Bynum refers specifically to footnote 8 of that opinion, which reads: "Since the first Circuit case to hold [indemnity] actions barred by Feres was decided in 1964 . . . petitioner no doubt had sufficient notice so as to take this risk into account in negotiating its contract for the emergency eject system at issue here." 431 U.S. at 674 n. 8, 97 S.Ct. at 2059 n. 8. We decline to adopt Bynum's construction of this footnote. As we have already noted, the question of the manufacturer's liability was not before the Stencel Court. The purpose of the footnote was merely to respond to the manufacturer's argument that disallowing the indemnity claim would be unfair. To this, the Court replied in the footnote that any unfairness that might result was a risk that presumably was taken into account during the contract negotiations. The Court intimated no view as to whether the primary suit against the manufacturer might actually be barred.

Since Stencel, it has become clear that the third factor described above is the principal justification for the Feres-Stencel doctrine. Most recently, for example, in United States v. Shearer, ___ U.S. ___, 105 S.Ct. 3039, 87 L.Ed.2d 38 (1985), the Supreme Court stated explicitly that the other factors mentioned in the Feres and Stencel decisions were "no longer controlling." Id. at ___ n. 4, 105 S.Ct. at 3043 n. 4. Rather, to the Shearer Court, the central questions have become "whether the suit requires the civilian court to second-guess military decisions and whether the suit might impair essential military discipline." Id. (citations omitted); see also Scales v. United States, 685 F.2d 970, 973 (5th Cir. 1982), cert. denied, 460 U.S. 1082, 103 S.Ct. 1772, 76 L.Ed.2d 344 (1983). That the Shearer Court should place so much importance on judicial restraint in military matters is of course not surprising. It has long been recognized that interference by civilian courts with military authority inevitably raises both questions about judicial competency in this area and separation of powers concerns. The Supreme Court has recently reaffirmed the importance of these considerations in holding that enlisted military personnel could not maintain a suit to recover damages from their superior officers for alleged violations of constitutional rights:

See also Lockheed Aircraft Corp. v. United States, 460 U.S. 190, 103 S.Ct. 1033, 74 L.Ed.2d 911 (1983). In Lockheed, the Supreme Court construed the Federal Employees' Compensation Act, 5 U.S.C. §§ 8101 et seq., to uphold an indemnity claim asserted by a manufacturer against the government arising from the death of a civilian employee of the Navy caused by the crash of an Air Force aircraft. There are many similarities between Lockheed and Stencel. In the factual contexts of both Stencel and Lockheed, for example, to allow indemnity claims against the government would subject the government to widely differing rules of liability. Further, indemnity in cases such as Lockheed and Stencel would permit the primary plaintiffs to avoid the limitations on government liability established by the respective no-fault compensation schemes. Nonetheless, the Lockheed Court found Stencel distinguishable on the basis of the "military nature of the action" in Stencel. Id. at 197 n. 8, 103 S.Ct. at 1037 n. 8. While this distinction is not altogether clear, it strongly implies that the essential reason for the Stencel decision is the possible adverse consequences that litigating disputes essentially between servicemen and the government in civilian courts may have on military effectiveness. See Hillier v. Southern Towing Co., 714 F.2d 714, 723 (7th Cir. 1983), cert. denied, ___ U.S. ___, 105 S.Ct. 961, 83 L.Ed.2d 966 (1985).

"It would be difficult to think of a clearer example of the type of governmental action that was intended by the Constitution to be left to the political branches directly responsible — as the Judicial Branch is not — to the electoral process. Moreover, it is difficult to conceive of an area of governmental activity in which the courts have less competence. The complex, subtle, and professional decisions as to the composition, training, equipping, and control of a military force are essentially military judgments, subject always to civilian control of the Legislative and Executive Branches. The ultimate responsibility for these decisions is appropriately vested in branches of the government which are periodically subject to electoral accountability."

Chappell v. Wallace, 462 U.S. 296, 302, 103 S.Ct. 2362, 2366, 76 L.Ed.2d 586 (1983) (emphasis in original) (quoting Gilligan v. Morgan, 413 U.S. 1, 10, 93 S.Ct. 2440, 2445, 37 L.Ed.2d 407 (1973)); see also Rostker v. Goldberg, 453 U.S. 57, 64-65, 101 S.Ct. 2646, 2651-2652, 69 L.Ed.2d 478 (1981); Orloff v. Willoughby, 345 U.S. 83, 93-94, 73 S.Ct. 534, 540, 97 L.Ed. 842 (1953); see generally Warren, The Bill of Rights and the Military, 37 N.Y.U.L.Rev. 181 (1962).

In sum, under the Feres-Stencel doctrine, the government is immune from both a primary suit by a member of the armed services for service-connected injuries and a third-party indemnity action arising from the same injuries. In either case, to allow such a suit would require a civilian court to second-guess military judgments or otherwise to intervene in military matters at the possible expense of military discipline and effectiveness. See Shearer, ___ U.S. at ___, 105 S.Ct. at 3043.

B. Traditional Defenses.

Historically, the contractor who strictly complied with design specifications provided by the government had two rather effective defenses to liability for any damage arising from such specifications. Neither defense, however, has retained its vitality in the context of a modern product liability action against a manufacturer of military equipment. The defenses, as well as their deficiencies, are well documented in the literature on the subject and therefore are only briefly reviewed below.

See, e.g., Note, Government Contract Defense: Sharing the Protective Cloak of Sovereign Immunity After McKay v. Rockwell Int'l Corp., 37 Baylor L.Rev. 181, 183, 192-94 (1985) [herein after cited as Note, Government Contract Defense]; Note, Liability of a Manufacturer for Products Defectively Designed by the Government, 23 B.C.L.Rev. 1025, 1031-55 (1982) [hereinafter cited as Note, Liability of a Manufacturer]; Note, The Government Contract Defense: Should Manufacturer Discretion Preclude its Availability?, 37 Me.L.Rev. 187, 188, 200 (1985) [hereinafter cited as Note, Manufacturer Discretion].

The first traditional "defense" was not actually a defense at all but a standard of conduct applicable to contractors working pursuant to either government or private contracts. Based on negligence principles, this standard provided that a contractor would not be liable for damages resulting from specifications provided by another unless those specifications were so obviously defective and dangerous that a contractor of reasonable prudence would be put on notice that the work was dangerous and likely to cause injury. Restatement (Second) of Torts § 404 comment a (1965); see also Davis v. Henderlong Lumber Co., 221 F. Supp. 129 (N.D.Ind. 1963); Ryan v. Feeney Sheehan Building Co., 239 N.Y. 43, 145 N.E. 321 (1924). The basis for this rule seems to have been the belief that the average contractor usually could not be expected to have sufficient expertise or knowledge to reevaluate design specifications provided by the owner. Johnston v. United States, 568 F. Supp. 351, 354 (D.Kan. 1983); Liability of a Manufacturer, supra note 8, at 1034-35. Special knowledge possessed by the contractor, however, would in certain instances subject the contractor to a higher standard of care. See, e.g., Johnston, 568 F. Supp. at 354; Person v. Cauldwell-Wingate Co., 187 F.2d 832 (2d Cir.) (Hand, J.), cert. denied, 341 U.S. 936, 71 S.Ct. 855, 95 L.Ed. 1364 (1951).

Courts are divided over whether this common law standard of liability provides protection to the contractor against a product liability action brought under strict liability principles. While some courts have extended the defense to cover strict liability claims on various public policy grounds, others have held that the negligence standard of liability simply has no relevance in a strict liability action, especially one in which a design defect is alleged. In Challoner v. Day Zimmermann, Inc., 512 F.2d 77 (5th Cir.), vacated on other grounds, 423 U.S. 3, 96 S.Ct. 167, 46 L.Ed.2d 3 (1975), for example, we found that under Texas law the plaintiff in a strict liability action had the burden of showing only that the product at some point was within the defendant's control, not that the defendant's act or omission was the cause of the defect. Id. at 83. Because the claim in Challoner was based upon strict liability theory, we concluded that the common law standard of care for contractors had no application. Id. The at best uncertain availability of this bar to liability under the several states' product liability laws makes it a precarious defense upon which the government contractor can rely.

See, e.g., McCabe Powers Body Co. v. Sharp, 594 S.W.2d 592, 594-95 (Ky. 1980) (illogical to hold nondesigner liable for design defect); Hunt v. Blasius, 55 Ill.App.3d 14, 12 Ill.Dec. 813, 370 N.E.2d 617, 620-22 (1977) (applying defense in strict liability case to encourage bidding and to keep government expenses down), aff'd on other grounds, 74 Ill.2d 203, 23 Ill.Dec. 574, 384 N.E.2d 368 (1978).

The second traditional defense fares no better. In Yearsley v. W.A. Ross Construction Co., 309 U.S. 18, 60 S.Ct. 413, 84 L.Ed. 554 (1940), the Supreme Court held that a contractor was not liable for the erosion of the plaintiff's waterfront property because the damage was caused by the contractor's construction of certain dikes in accordance with a contract with the government. While the basis of the contractor's defense is not altogether clear from the text of the opinion, the Supreme Court's reference to the contractor as an "agent or officer of the Government," id. at 21, 60 S.Ct. at 414, indicates that as such the contractor was entitled to share in the government's sovereign immunity. In any event, it has since become well established that, at least with respect to public works projects, when contractors as agents or officers of the federal government or of individual states work according to government specifications, they are entitled to assert the government's sovereign immunity in suits arising from such activities. See, e.g., Myers v. United States, 323 F.2d 580 (9th Cir. 1963); Green v. ICI American, Inc., 362 F. Supp. 1263 (E.D.Tenn. 1973); Dolphin Gardens, Inc. v. United States, 243 F. Supp. 824 (D.Conn. 1965). Of course, contractors that fail to follow government specifications or otherwise mismanufacture a product are not entitled to raise the defense. See Foster v. Day Zimmermann, Inc., 502 F.2d 867, 874 n. 5 (8th Cir. 1974) ("The government's specifications did not call for the defendants to assemble a defectively made grenade.").

The problem with applying the Yearsley defense in the context of the military contractor is the apparent requirement that the contractor possess an actual agency relationship with the government. While federal courts certainly have not always required such a relationship, see, e.g., Pratt v. Hercules, Inc., 570 F. Supp. 773, 803 (D.Utah 1982), it is the law at least in this Circuit that the contractor must be an agent of the government before invoking its immunity. Thus, in Whitaker v. Harvell-Kilgore Corp., 418 F.2d 1010 (5th Cir. 1969), we rejected a claim of immunity asserted by manufacturers of grenades and fuses made according to government specifications on the ground that the defendants — although having close contractual ties with the government — ultimately were independent contractors. The difficulty of establishing a traditional agency relationship with the government makes the derivative sovereign immunity defense illsuited to many manufacturers of military equipment.

The literature also is divided over whether an agency relationship is a prerequisite to the application of the Yearsley defense. Compare Note, Government Contract Defense, supra note 8, at 192 and Note, The Government Contract Defense in Strict Liability Suits for Defective Design, 48 U.Chi.L.Rev. 1030, 1046 (1981) [hereinafter cited as Note, Strict Liability Suits] with Note, Liability of a Manufacturer, supra note 8, at 1049-55.

In Whitaker, the manufacturer of grenades, Day Zimmermann, Inc., operated a government-owned plant and used material supplied by the government. Assembly of the grenades rigidly followed government specifications. Further, the contract itself called for indemnification by the government for losses arising from performance under the contract. Nonetheless, the panel, citing Powell v. United States Cartridge Co., 339 U.S. 497, 70 S.Ct. 755, 94 L.Ed. 1017 (1950), held that the manufacturer was "in no wise an agent of the Government." Whitaker, 418 F.2d at 1014. The panel similarly disposed of the immunity claim of Harvell-Kilgore Corp., the manufacturer of the fuses, stating that "Kilgore was clearly an independent contractor and is not entitled to sovereign immunity." Id. at 1015.

C. Modern Government Contractor Defense.

From the foregoing, it is evident that in the last few decades, while the government's immunity from suits having some connection with service-related injuries of members of the armed services has been growing more expansive and defined, the traditional defenses of military contractors have been becoming less effective in shielding them from liability for the design defects of their military products. Consequently, especially in recent years, military contractors have been threatened more frequently with liability that could not be passed on to the party ultimately responsible for the plaintiffs' injuries — the government. Military contractors, in response, have sought both judicial and legislative relief from their plight. See generally Souk, Government Contracts and Tort Liability: Time for Reform, 30 Fed.B. News J. 70 (1983). One limited form of relief that has been made available, referred to herein as the "government contractor defense," is essentially a judicial amalgamation of the two traditional defenses, see Johnston, 568 F. Supp. at 353, and has been adopted to some extent by almost every court, both state and federal, that has considered the matter. See, e.g., In re Air Crash Disaster at Mannheim Germany, 769 F.2d 115, 121 (3d Cir. 1985); Tillett v. J.I. Case Co., 756 F.2d 591, 597 (7th Cir. 1985); McKay v. Rockwell International Corp., 704 F.2d 444, 451 (9th Cir. 1983), cert. denied, ___ U.S. ___, 104 S.Ct. 711, 79 L.Ed.2d 175 (1984); Shaw v. Grumman Aerospace Corp., 593 F. Supp. 1066, 1074 (S.D.Fla. 1984); In re Related Asbestos Cases, 543 F. Supp. 1142, 1151-52 (N.D.Cal. 1982); In re "Agent Orange" Product Liability Litigation, 534 F. Supp. 1046, 1055 (E.D.N.Y. 1982), modified, 597 F. Supp. 740, 849 (E.D.N.Y. 1984); McLaughlin v. Sikorsky Aircraft, 148 Cal.App.3d 203, 195 Cal.Rptr. 764, 768 (1983); Sanner v. Ford Motor Co., 144 N.J.Super. 1, 364 A.2d 43, 47 (1976), aff'd, 154 N.J.Super. 407, 381 A.2d 805, 806 (1977); Casabianca v. Casabianca, 104 Misc.2d 348, 350, 428 N.Y.S.2d 400, 402 (Sup.Ct. 1980). But cf. Nobriga v. Raybestos-Manhattan, Inc., 683 P.2d 389, 392 (Hawaii 1984) (rejecting defense in asbestos case on ground that "product manufactured was inherently dangerous because of a quality in the materials specified for use").

This dilemma of the military contractor is made all the more severe by (1) the contractor's frequent inability to deviate from the specifications supplied by the government in the manufacture of a product and (2) the excessive risks that are frequently inherent in the use of essential military equipment. See generally Polinsky, Products Liability and the United States Government Contractor, 14 Pub.Cont.L.J. 313 (1984); Tobak, A Case of Mistaken Liability: The Government Contractor's Liability for Injuries Incurred by Members of the Armed Forces, 13 Pub.Cont.L.J. 74 (1982).

Courts have relied upon numerous reasons of policy in recognizing the defense. In our view, the most important of these is that a trial between a serviceman and a military contractor in which government specifications are at issue would inevitably implicate the same concerns that underlie the Supreme Court's Feres and Stencel decisions. Litigation involving defective designs in military products would take the identical form regardless of whether the named defendant happens to be the government or the military contractor. In either case, members of the armed services would be allowed to question military decisions and obtain relief from actions of military officials. Moreover, civilian courts would be compelled to second-guess professional military judgment concerning, at least, the proper equipping of the armed services. As noted in Part I(A), there are few areas of government activity in which courts have less competence or that raise such significant separation of powers concerns. In re Air Crash Disaster, 769 F.2d at 121; Tillett, 756 F.2d at 597; McKay, 704 F.2d at 449; Note, Government Contract Defense, supra note 8, at 187-88.

Other policy reasons suggested by the courts also give strong support to the fashioning of a government contractor defense. First, as many courts have recognized, holding military contractors liable for defective designs supplied by the government would circumvent the government's immunity under Feres and Stencel. "[D]espite the government's immunity, [military suppliers] would pass the cost of accidents off to the United States through cost overrun provisions in equipment contracts, through reflecting the price of liability insurance in the contracts, or through higher prices in later equipment sales." McKay, 704 F.2d at 449; see also In re Air Crash Disaster, 769 F.2d at 121; In re "Agent Orange" Product Liability Litigation, 506 F. Supp. 762, 793-94 (E.D.N.Y.), rev'd on other grounds, 635 F.2d 987 (2d Cir. 1980); Note, McKay v. Rockwell International Corp.: No Compulsion Required for Government Contractor Defense, 28 St. Louis U.L.J. 1061, 1073 (1984) [hereinafter cited as Note, No Compulsion Required].

Second, military contractors are often unable through negotiation to alter the design specifications of their military products because of military efforts in certain contexts to push technology to the limits even if to do so would incur risks beyond those that would ordinarily be acceptable for consumer goods. Without the government contractor defense, military contractors would be discouraged from bidding on essential military projects. Note, Liability of a Manufacturer, supra note 8, at 1067-68. And, if a contractor were compelled by federal law to manufacture the product, the contractor would "face the untenable position of choosing between severe penalties for failing to supply products necessary to conduct a war, and producing what the government requires but at a contract price that makes no provision for the need to insure against potential liability for design flaws in the government's plans." In re "Agent Orange" Product Liability Litigation, 506 F. Supp. at 794. See also McKay, 704 F.2d at 450; Tobak, supra note 12, at 75-78; Note, Government Contract Defense, supra note 8, at 192.

See Defense Production Act, 50 U.S.C. app. 2071(a) (giving executive authority to require private contractor to accept military contracts).

Third, by conditioning the application of the government contractor defense on full disclosure to the government of all known risks, military contractors would have an incentive to work closely with military authorities in the development and testing of equipment. Tillett, 756 F.2d at 597; McKay, 704 F.2d at 450. As a result, military procurement decisions would be made based on all readily available information. Koutsoubos v. Boeing Vertol, 755 F.2d 352, 355 (3d Cir. 1985).

Finally, principles of fairness dictate that an innocent contractor should not be ultimately liable for a dangerous design when the responsibility properly lies elsewhere. Brown v. Caterpillar Tractor Co., 696 F.2d 246, 252-53 (3d Cir. 1982); Note, Government Contract Defense, supra note 8, at 191-92; Note, Liability of a Manufacturer, supra note 8, at 1072-73. In this respect, an important consideration is that injured servicemen are at least partially compensated through payments under the Veterans' Benefits Act. See Feres, 340 U.S. at 145, 71 S.Ct. at 159.

These policies underlying the government contractor defense, although broad in nature, have nonetheless limited its application. One version of the defense that is particularly well tailored to the policies that it serves was formulated by the Ninth Circuit in McKay v. Rockwell International Corporation, a case arising under the Death on the High Seas Act, 46 U.S.C. §§ 761-767. To invoke the government contractor defense under the McKay test, the military contractor must demonstrate:

(1) [that] the United States is immune from liability under Feres and Stencel,

(2) . . . that the United States established, or approved, reasonably precise specifications for the allegedly defective military equipment,

(3) [that] the equipment conformed to those specifications, and

(4) [that] the supplier warned the United States about the patent errors in the government's specifications or about dangers involved in the use of the equipment that were known to the supplier but not to the United States.

McKay, 704 F.2d at 451.

Although the McKay test has been widely adopted, see, e.g., In re Air Crash Disaster, 769 F.2d at 122; Tillett, 756 F.2d at 597; Shaw, 593 F. Supp. at 1074; McLaughlin, 148 Cal.App.3d 203, 195 Cal.Rptr. at 768, certain aspects of the McKay test have been questioned, and other significant issues, not addressed by the Ninth Circuit in McKay, remain. See, e.g., In re All Maine Asbestos Litigation, 575 F. Supp. 1375, 1378-80 (D.Me. 1983) (collecting issues). Some of these issues are discussed infra. We of course express no opinion on any issue concerning the government contractor defense that is not directly raised by the circumstances of this case.

Another often-cited version of the modern government contractor defense is the one set forth by the District Court for the Eastern District of New York in In re "Agent Orange" Product Liability Litigation, 534 F. Supp. 1046 (E.D.N. Y. 1982). Under this formulation, a government contractor is not liable if (1) the government established the specifications for the military product, (2) the product met the government's specifications in all material respects, and (3) the government knew as much as or more than the military contractor about the hazards to people that accompanied use of the military product. Id. at 1055. A comparison between the McKay and Agent Orange tests reveals that there is not much difference between the two. Although the McKay version explicitly requires the application of the Feres and Stencel doctrines, the Agent Orange opinion never states that the application of the doctrine is not necessary, and it is clear that in the Agent Orange context the doctrine is applicable. See Koutsoubos v. Boeing Vertol, 755 F.2d at 355 (stating that the Ninth Circuit in McKay rephrased the Agent Orange test); see also infra note 28.
In 1984, the Agent Orange court modified its version of the government contractor defense. See In re "Agent Orange" Product Liability Litigation, 597 F. Supp. 740, 849 (E.D.N.Y. 1984). The reasons for the revised test do not appear to be applicable in the case at bar, see infra note 29, and the revised test has not been as widely accepted as the earlier version.

III. ANALYSIS.

Bynum on appeal challenges in two general respects the district court's opinion and order granting summary judgment to FMC. First, Bynum argues that there is nothing in Mississippi caselaw to suggest that Mississippi courts would adopt the government contractor defense. Thus, Bynum asserts, the district court erred in applying it as a matter of Mississippi law. Second, Bynum contends that, even if the government contractor defense were recognized, the particular version adopted by the district court would provide more protection to the military contractor than is required by the legitimate policies of the defense. Because we find that the interests that justify the creation of the government contractor defense also compel its application as a matter of federal common law, we do not reach the state law question. Further, for the reasons outlined above and discussed more fully below, we adopt the McKay test, at least insofar as it applies to the circumstances of the instant case, and affirm the summary judgment.

The application of the government contractor defense as a matter of federal common law was urged by FMC and was fully briefed by both parties.

A. Federal Common Law.

There is no question that jurisdiction in this product liability suit is based solely on diversity of citizenship and therefore state law generally governs the resolution of the controversy. See Jackson v. Johns-Manville Sales Corp., 750 F.2d 1314, 1323-27 (5th Cir. 1985) (en banc); In re "Agent Orange" Product Liability Litigation, 635 F.2d 987, 993-95 (2d Cir. 1980). That a plaintiff's claim arises under state law, however, does not preclude the formulation of a federal defense in cases raising issues of uniquely federal concern. See Howard v. Lyons, 360 U.S. 593, 597, 79 S.Ct. 1331, 1334, 3 L.Ed.2d 1454 (1959); In re Diamond Shamrock Chemicals Co., 725 F.2d 858, 861 n. 2 (2d Cir.), cert. denied, ___ U.S. ___, 104 S.Ct. 1417, 79 L.Ed.2d 743 (1984). With regard to the government contractor defense, most of the courts that have considered the matter have found that, at least when military design specifications provided by the government are at issue, product liability actions are likely to involve matters that are subject to exclusive federal control and necessitate the limited imposition of federal common law. See, e.g., In re "Agent Orange" Product Liability Litigation, 597 F. Supp. at 847; In re Related Asbestos Cases, 543 F. Supp. at 1151; McLaughlin v. Sikorsky Aircraft, 148 Cal.App.3d 203, 195 Cal.Rptr. at 768. We think these cases were correctly decided and hold that, in suits by members of the armed services against military contractors for defects in designs supplied by the government, federal common law provides a basis for the application of the government contractor defense.

But see Brown v. Caterpillar Tractor Co., 696 F.2d at 247-49. Brown is discussed infra note 17.

Under the applicable caselaw, we must apply a two-tier analysis in determining whether federal common law provides a government contractor defense under the circumstances of this case. First, as a threshold matter, we must consider whether the question at bar substantially bears upon "uniquely federal" interests. Texas Industries, Inc. v. Radcliff Materials, Inc., 451 U.S. 630, 640, 101 S.Ct. 2061, 2067, 68 L.Ed.2d 500 (1981); Jackson v. Johns-Manville Sales Corp., 750 F.2d at 1323. If federal interests are so implicated, we must then determine whether the court should choose federal common law or state law as the applicable federal rule. United States v. Kimbell Foods, Inc., 440 U.S. 715, 728, 99 S.Ct. 1448, 1458, 59 L.Ed.2d 711 (1979); Georgia Power Co. v. Sanders, 617 F.2d 1112, 1115 (5th Cir. 1980) (en banc), cert. denied, 450 U.S. 936, 101 S.Ct. 1403, 67 L.Ed.2d 372 (1981). This inquiry essentially involves focusing on the effect that applying state law would have on the federal interests and the extent to which imposition of federal common law would disrupt important and legitimate state policies. Kimbell, 440 U.S. at 728-29, 99 S.Ct. at 1458-59; Sanders, 617 F.2d at 1118; see generally Miree v. Dekalb County, 433 U.S. 25, 30-33, 97 S.Ct. 2490, 2494-95, 53 L.Ed.2d 557 (1977); Wallis v. Pan American Petroleum Corp., 384 U.S. 63, 67-68, 86 S.Ct. 1301, 1303-1304, 16 L.Ed.2d 369 (1966). While basic considerations of federalism require a presumption against the displacement of state law, if the adoption of state law as the federal rule would frustrate federal policies or otherwise interfere with the authority and duties of the United States as sovereign in our constitutional system, federal common law must be applied irrespective of state interests. Sanders, 617 F.2d at 1118.

1. Uniquely Federal Interests.

In Texas Industries, Inc. v. Radcliff Materials, supra, the Supreme Court described the kinds of cases that raise issues of uniquely federal concern:

[A]bsent some congressional authorization to formulate substantive rules of decision, federal common law exists only in such narrow areas as those concerned with the rights and obligations of the United States, interstate and international disputes implicating the conflicting rights of States or our relations with foreign nations, and admiralty cases. In these instances, our federal system does not permit the controversy to be resolved under state law, either because the authority and duties of the United States as sovereign are intimately involved or because the interstate or international nature of the controversy makes it inappropriate for state law to control.

Id. 451 U.S. at 641, 101 S.Ct. at 2067. See also Jackson v. Johns-Manville Sales Corp., 750 F.2d at 1323-27 (national interests alone, no matter their significance, cannot justify the application of federal common law).

The most obvious example in which the authority and duties of the United States as sovereign would be intimately involved is a controversy whose outcome will have an immediate effect on the federal treasury. Clearfield Trust Co. v. United States, 318 U.S. 363, 366-67, 63 S.Ct. 573, 574-75, 87 L.Ed. 838 (1943). It is well recognized, however, that uniquely federal interests may exist even when the government is not a party to the litigation and when the federal fisc is not directly implicated. Banco Nacional v. Sabbatino, 376 U.S. 398, 423-24, 84 S.Ct. 923, 937-38, 11 L.Ed.2d 804 (1964); see also Miree v. Dekalb County, 433 U.S. at 31-32, 97 S.Ct. at 2494-95; Wallis v. Pan American Petroleum Corp., 384 U.S. at 71-72, 86 S.Ct. at 1305-06; Georgia Power Co. v. Sanders, 617 F.2d at 1115-18.

We have already seen that the primary justification for the government contractor defense is the same as that for the Feres-Stencel doctrine: members of the armed services should not be able to challenge the decisions and actions of military authorities by thrusting the judiciary into the role of adjudicating what are essentially military matters. The plenary control of military judgments by the legislative and executive branches of the federal government, however, is more than a separation of powers concern. The composition, training, equipping, and management of our military forces is a matter exclusively within the rights and duties of the federal government and, as a result, any interference with the federal authority over national defense and military affairs implicates uniquely federal interests of the most basic sort. This was one of the lessons of In re Tarble, 80 U.S. (13 Wall) 397, 20 L.Ed. 597 (1871), in which the Supreme Court held that a state court had no power to issue a writ of habeas corpus for the discharge of a petitioner within the custody of the United States Army. The Supreme Court stated:

Now, among the powers assigned to the National government, is the power "to raise and support armies," and the power "to provide for the government and regulation of the land and naval forces." The execution of these powers falls within the line of its duties; and its control over the subject is plenary and exclusive. It can determine, without question from any State authority, how the armies shall be raised, whether by voluntary enlistment or forced draft, the age at which the solider shall be received, and the period for which he shall be taken, the compensation he shall be allowed, and the service to which he shall be assigned. And it can provide the rules for the government and regulation of the forces after they are raised, define what shall constitute military offenses, and prescribe their punishment. No interference with the execution of this power of the National government in the formation, organization, and government of its armies by any State officials could be permitted without greatly impairing the efficiency, if it did not utterly destroy, this branch of the public service.

Id. at 408, 20 L.Ed. 597. See also United States v. Standard Oil Co., 332 U.S. 301, 305-07, 67 S.Ct. 1604, 1607-08, 91 L.Ed. 2067 (1947) (federal common law governs in cases involving interference with relation between member of the armed services and the government).

The application of these principles to a soldier's product liability litigation involving the merits of military design specifications supplied by the government is clear. There can be no question that the design of military equipment is, at bottom, a military decision. Often dangerous designs must be used in the military context to meet the exigencies of our national defense, and even military equipment that is relatively safe for every day use may have to be operated on occasion under dangerous conditions or in a manner creating a high risk of harm. In this respect, we think that the District Court for the Eastern District of New York was correct in observing:

The purpose of a government contract defense . . . is to permit the government to wage war in whatever manner the government deems advisable, and to do so with the support of suppliers of military weapons. Considerations of cost, time of production, risks to participants, risks to third parties, and any other factors that might weigh on the decisions of whether, when, and how to use a particular weapon are uniquely questions for the military and should be exempt from review by civilian courts.

In re "Agent Orange" Product Liability Litigation, 534 F. Supp. at 1054 n. 1 (E.D.N.Y. 1982). Furthermore, as the Supreme Court concluded in Stencel, suit by a serviceman alleging that the government defectively designed its military equipment poses a threat to military discipline. Regardless of whether the named defendant is the government or the military contractor, the litigable issue would be the same, and the respective parties would have to introduce the same evidence and make the same arguments. In either case, the controversy would raise significant federal interests in the management of the military and specifically in the relationship between members of the armed services and military command. Cf. Chappell v. Wallace, 462 U.S. at 304, 103 S.Ct. at 2367.

We note, however, that the Third Circuit has apparently adopted an opposite view. In Brown v. Caterpillar Tractor Co., supra, the court held that federal interests do not justify the creation of a government contractor defense as a matter of federal law. In so holding, the court, applying the factors outlined by the Supreme Court in Feres and Stencel, stated that "such suits generally do not necessitate the second-guessing of military decisions." 696 F.2d at 249. The basis for this rather conclusory statement was not explained.

In In re "Agent Orange" Product Liability Litigation, 506 F. Supp. at 746-47, the district court also found uniquely federal interests "in the liability of war contractors to soldiers, since the extent of a contractor's liability may, undoubtedly will, affect future dealings between the contractor and the government." Id. at 747. In a footnote, the district court stated further:

Speculative federal interests in the obligations of war contractors are numerous. War contractors might be expected to increase the price of war materials to correspond to any extension in their potential liability. Such adjustments might have a significant effect on the federal treasury. If potential liability increased dramatically, future war contractors might attach conditions to the use of their products, or balk at supplying the military with any products whatsoever. Thus, the government's military capacities might be affected by this litigation.

Id. at 747 n. 5. See also Note, Tort Remedies for Servicemen Injured by Military Equipment: A Case for Federal Common Law, 55 N.Y.U.L.Rev. 601, 617-18 (1980).

We find the present situation to be not unlike that of Banco Nacional v. Sabbatino, supra. At issue in Banco Nacional was the legal basis and scope of the "act of state" doctrine, which precludes judicial inquiry into the public acts of a recognized foreign sovereign power committed within its own territory. The Supreme Court held that, because of the essential political nature of foreign expropriation and its significance in international affairs, the basic relations between the judicial and political branches of government required that the judiciary not pass on the validity of the expropriations. Id. 376 U.S. at 435-37, 84 S.Ct. at 944-45. Further, the Court made clear that the same considerations underlying the adoption of the act of state doctrine compelled the displacement of state law. The Court stated:

Whatever considerations are thought to predominate, it is plain that the problems involved are uniquely federal in nature. If federal authority, in this instance this Court, orders the field of judicial competence in this area for the federal courts, and the state courts are left free to formulate their own rules, the purposes behind the doctrine could be as effectively undermined as if there had been no federal pronouncement on the subject.

Id. at 424, 84 S.Ct. at 938.

Similarly, the government contractor defense is rooted in separation of powers concerns that would be frustrated by the application of state law and policies. Like the act of state doctrine, the government contractor defense embodies "a basic choice regarding the competence and function of the Judiciary and the National Executive" and "should not be left to divergent and parochial state interpretations." Id. at 425, 84 S.Ct. at 938.

Thus, we find that a suit by a serviceman against a military contractor in which at issue are government design specifications for military equipment does give rise to uniquely federal interests sufficient to warrant the imposition of federal law. Such a suit involves the second-guessing of military decisions by civilian courts and threatens to impair essential military discipline. The states, as well as the judiciary, are "ill-equipped to determine the impact upon military effectiveness that any particular intrusion upon military authority may have," see Chappell v. Wallace, 462 U.S. at 305, 103 S.Ct. at 2368 (quoting Warren, supra, at 187), and, as a consequence, the management of the military is of paramount federal concern.

2. Conflict with State Policy.

The overriding federal interest at stake here concerns the preservation of exclusive control over military matters in the legislative and executive branches of the federal government. Thus, if state laws permitting or requiring judicial intervention into such matters were applied, the federal interest would be entirely frustrated. This conflict, by itself, suffices to preclude adoption of state law as the federal rule. Sanders, 617 F.2d at 1120.

There is a significant federal interest in uniformity for its own sake here as well. If a contractor's liability for the accident costs of its military equipment depends upon which state law is applied, the contractors that are less likely to be subject to such liability would have an unfair competitive advantage in the bidding process. Consequently, it may be more difficult to identify and award military contracts to the most efficient contractor.

We nonetheless also examine the effect of a federal government contractor defense on important and legitimate state interests. In this regard, we find most telling the fact that a clear majority of courts that have considered the availability of the government contractor defense under applicable state law have decided to adopt the defense. See, e.g., Tillett v. J.I. Case Co., 756 F.2d at 599-00 (Wisconsin); Brown v. Caterpillar Tractor Co., 696 F.2d at 249-54 (Pennsylvania); Hunt v. Blasius, 55 Ill. App.3d 14, 12 Ill.Dec. at 818, 370 N.E.2d at 622; Sanner v. Ford Motor Co., 144 N.J. Super. 1, 364 A.2d at 46-47; Casabianca v. Casabianca, 104 Misc.2d at 350, 428 N.Y.S.2d at 402. Moreover, the modern government contractor defense itself is based on two historic counterparts that were both recognized at common law. See supra Part II; Annot., 9 A.L.R.3d 382 (1966).

Nor would the traditional policies that support strict liability be substantially advanced by application of the doctrine in a case where the military contractor merely followed the design specifications of the government. The parties draw our attention to four principal reasons for imposing strict liability: enterprise liability, market deterrence, compensation, and implied representation of safety. See also McKay, 704 F.2d at 451; Note, Protecting the Buyer of Used Products: Is Strict Liability for Commercial Sellers Desirable?, 33 Stan.L.Rev. 535, 536 n. 7 (1981). In McKay, the Ninth Circuit discussed each of these justifications and convincingly demonstrated that none is fully implicated in a situation where the government contractor defense would be applicable.

According to enterprise liability principles, strict liability should be imposed on producers in order to force them to incorporate accident costs into the price of their products. The theory is that the increased prices will then discourage consumers from purchasing risky products and thereby lower total accident costs to society. Further, imposing the costs of accidents generally on manufacturers and consumers allows the costs of product-caused injuries to be spread over a class of users. In the context of the government contractor defense, however, neither of these goals would be met. First, as the McKay court noted, the demand for military equipment, unlike consumer goods, is relatively inelastic. Decisions to purchase military equipment are usually made far in advance and are often based on considerations of military and political strategy as well as the government's assessment of the risks and the benefits involved in the use of the equipment. "Meeting adequately the needs of national defense, not accident costs, is the ultimate standard by which purchases of military equipment must be measured." McKay, 704 F.2d at 452. Increasing prices of military equipment, therefore, would not often result in the use of safer substitutes but might pose the risk of a diminished military capability. Second, the loss-spreading goal of strict liability is built on the presumption that the manufacturer sells a large volume of products to a large number of customers. This is simply not the case with respect to military equipment purchased by the government. The effect of strict liability here would be to shift the entire cost of the accident to the military contractor or, more likely, the government, thus subverting the Feres-Stencel doctrine. Note, Liability of a Manufacturer, supra note 8, at 1080.

Strict liability also often functions to deter manufacturers from marketing unreasonably dangerous products. When military equipment is at issue, though, there is a strong government interest in encouraging the military contractor to agree to manufacture potentially dangerous products for the government and to comply strictly with government design specifications. Nor do these policies deprive the user of all protection from unintended design defects. As is illustrated by the record in this case, the government "has both the ability to recognize safety problems in military equipment and to negotiate with the suppliers to remedy those problems." McKay, 704 F.2d at 452. There is thus strong reason to believe, especially in view of the relatively inelastic demand for military equipment, that requiring the contractor to test war materials would merely increase defense costs without resulting in significantly safer products. Id.

As recounted in Part I, the M-548 cargo carrier at issue in this case was designed by TACOM, a division of the Department of the Army. According to the uncontroverted affidavit of William H. Stites, submitted in conjunction with FMC's motion for summary judgment, TACOM has a staff of over 5500 personnel, both civilian and military, who are engaged in the design and testing of wheeled and tracked vehicles procured for military use. TACOM personnel include power plant engineers, power train engineers, suspension engineers, material engineers (ballistics experts, etc.), welding engineers, quality engineers, packing engineers, and other specialists. It is TACOM's reported function to make the military decisions and the engineering and design decisions necessary to insure that the M-548 as well as other military vehicles are able to perform the military missions for which they are conceived. Record at 318.

The third justification, compensation, also does not compel the imposition of strict liability for defective designs supplied by the government. As the Supreme Court noted in Stencel, "the Veterans' Benefits Act establishes, as a substitute for tort liability, a statutory no fault compensation scheme which provides generous pension to injured servicemen." Stencel, 431 U.S. at 671, 97 S.Ct. at 2057. While strict liability would obviously increase that compensation, to the extent the military contractor has to bear the cost of the increase, the burden would fall on an innocent defendant. And to the extent that the burden can be shifted to the government through higher prices, the added compensation would be in conflict with one of the primary purposes of the Veterans' Benefits Act: to establish "an upper limit of liability for the Government as to service-connected injuries." Stencel, 431 U.S. at 673, 97 S.Ct. at 2059.

Finally, strict liability usually serves to vindicate the reasonable expectations of the consumer that a product purchased by a seller will not be unreasonably dangerous. As the McKay court observed, however, "[members of the armed services] recognize when they join the armed forces that they may be exposed to grave risks of danger." McKay, 704 F.2d at 453. Moreover, even if servicemen have the same expectations of safety as those of ordinary consumers, these expectations would be directed toward the government, not the military contractor, when the government is responsible for the formulation of the design specifications in question and accepts the product once it is manufactured. See Note, Liability of a Manufacturer, supra note 8, at 1080.

3. Fifth Circuit Precedent.

Our holding today, adopting the government contractor defense as a matter of federal common law, is not inconsistent with either Challoner v. Day Zimmermann, Inc., supra, or Hansen v. Johns-Manville Products Corp., supra. In Challoner, a serviceman sustained injuries when a howitzer round prematurely exploded during combat with North Vietnamese. He subsequently brought a strict liability tort action against the manufacturer in federal district court. The manufacturer responded that it was immune from liability because the design of the shell was within the exclusive control of the government. 512 F.2d at 78-79. On appeal from judgment for the serviceman, we refused to apply federal common law on the ground that we could discern no specific federal policy implicated in the action. Id. at 80 n. 2. Applying Texas state law, we then held that the defense raised by the manufacturer was not a bar under Texas law against a strict liability claim. Id. at 83.

Challoner is distinguishable on several points. First, we considered in that case only the application of federal common law to the entire action, not the creation of merely a federal defense. Cf. Jackson v. Johns-Manville Sales Corp., 750 F.2d at 1323-27. Thus, the Challoner court never considered the issue addressed by us today. Second, and more importantly, Challoner was issued before the Supreme Court's Stencel decision. Stencel reaffirmed and clarified the proposition that strong federal interests exist in avoiding judicial second-guessing of military decisions as well as any other type of judicial intervention into military matters that would impair military discipline and effectiveness. These interests, which support the imposition of federal common law, were not brought to the attention of the Challoner court. Third, as already noted, the Texas defense asserted in Challoner was based on negligence principles and thus had little application in the strict liability context. The Challoner court itself duly noted this point in distinguishing the cases cited by the manufacturer: "They involved attempts to demonstrate negligence and stand only for the proposition that there is no negligence in following the design of another unless the design is such that the defectiveness was sufficiently obvious to alert a reasonably competent technician to the danger." Id. at 83; see Note, Government Contract Defense, supra, note 8, at 215.

Challoner can also be construed as a mismanufacture case since the most likely cause of the explosion was the existence of a cavitation in the explosive material poured into the artillery round. 512 F.2d at 79. The government contractor defense, of course, does not apply when the actionable injuries result from manufacturing defects.

Hansen v. Johns-Manville Products Corp. involved a diversity suit brought by an employee of a shipyard against the manufacturer of asbestos for injuries received allegedly as a result of his work with asbestos products. The manufacturer in defense claimed that it produced the asbestos in conformity with government specifications and thus under Texas law should be immune from liability. We refused to recognize the government contractor defense in that case in part because Texas courts had not indicated that they would adopt the defense under similar circumstances. 734 F.2d at 1044-45.

Hansen, like Challoner, is distinguishable on the ground that we also were not presented with the federal common law question. Both parties concerned themselves solely with the application of Texas law and policies. Furthermore, the plaintiff in Hansen was not a member of the armed services. Thus, the federal interest in military discipline was not implicated. That this factor is a vital component in the analysis cannot be questioned in view of the Supreme Court's recent refusal to apply the Feres-Stencel doctrine to cases involving military-related injuries of civilians. See Lockheed Aircraft Corp. v. United States, 460 U.S. 190, 103 S.Ct. 1033, 74 L.Ed.2d 911 (1983). Finally, the manufacturer in Hansen performed work under government contract for only a short period of time relative to the duration of the plaintiff's exposure to asbestos. As we stated in Hansen, "[w]e do not think that the defense would apply in these circumstances." 734 F.2d at 1045.

We pretermit the question whether asbestos can be considered "military equipment."

4. Conclusion.

In sum, we hold that uniquely federal interests are implicated in actions brought by members of the armed services against military contractors where at issue are design specifications supplied by the government for the manufacture of military equipment. The federal interests, rooted in notions of the limits of the judicial function and in separation of powers concerns, require the creation of a federal common law government contractor defense. We now turn to address some of the specifics of that defense and its applicability in the case at hand.

B. Government Contractor Defense Applied.

At least with respect to the circumstances of this case, we think that the McKay formulation of the government contractor defense, set forth in Part II, substantially reflects the pertinent federal interests. To establish the government contractor defense, a military contractor must first demonstrate that the government is immune from liability under the Feres-Stencel doctrine. This requirement ensures that the federal interests in avoiding judicial and state intrusion upon military authority are present and warrant the imposition of federal common law. Second, the military contractor must prove that the government established reasonably precise specifications for the allegedly defective military equipment and that the equipment conformed to those specifications. This element makes clear that federal law provides no defense to the military contractor that mismanufactures military equipment or that is itself ultimately responsible for the design defect. Finally, it must be shown that the military contractor warned the government about errors in the government specifications or dangers involved in the use of the equipment that were known to the contractor but not to the government. The primary purpose of this requirement is to enable the government to make determinations as to the design and use of military equipment based on all readily available information. See McKay, 704 F.2d at 451; cf. In re "Agent Orange" Product Liability Litigation, 534 F. Supp. at 1055.

Courts and commentators disagree over the extent to which the government must participate in generating the design specifications of military equipment before the government contractor defense would be applicable. See, e.g., In re Air Crash Disaster, 769 F.2d at 121; McKay, 704 F.2d at 450-51; Price v. Tempo, Inc., 603 F. Supp. 1359, 1363 (E.D.Penn. 1985); Note, Government Contract Defense, supra note 8, at 221-22; Note, Manufacturer Discretion, supra note 8, at 200. We need not address this issue. Before the district court, both parties stipulated that the product at issue here "was manufactured in accordance with precise design specifications furnished by the United States government." 599 F. Supp. at 156. This certainly is enough to satisfy the requirement.

Because the policies underlying the government contractor defense apply equally well regardless of whether the action is predicated on an allegedly dangerous defect in the design of the product or a failure to warn, the government contractor defense functions as an affirmative defense against both claims of tort liability. See McKay, supra (defense applies in cases brought under Restatement (Second) of Torts § 402A); Note, Liability of a Manufacturer, supra note 8, at 1063-64.

Bynum asserts three objections to our adoption of the McKay test. First, Bynum contends that the government contractor defense should be limited to those circumstances in which a military contractor was required by law to manufacture the product and had no ability to refuse to enter into the contract. Cf. Note, Strict Liability Suits, supra note 10, at 1048-51 (advocating adoption of similar type of compulsion prerequisite). Such a requirement, however, would bear no connection with the uniquely federal interests that the government contractor defense is designed to protect. Moreover, a requirement of compulsion of this sort would discourage military contractors from bidding on government projects or, alternatively, give the contractors incentive to pressure the military to purchase safer equipment. Either result would be contrary to sound public policy, which, as we have seen, requires that contractors be encouraged to enter into contracts with the government for the production of military hardware and to adhere fully to specifications furnished by the government. While it is reasonable to expect a military contractor to apprise the government of any risks relating to the equipment of which it has knowledge, the contractor, like the judiciary, should not be thrust into the position of second-guessing military decisions. As noted above, it is the needs of national defense, not accident costs, that must be the ultimate standard by which the purchase of military equipment is measured. McKay, 704 F.2d at 452; see also Tillett, 756 F.2d at 597-98. Accordingly, we refuse to adopt compulsion as an element of the government contractor defense.

Although we hold that compulsion to contract with the government is not a prerequisite to raising the government contractor defense, we express no opinion as to whether a precondition of the defense should be that the contractor was compelled under the contract to follow the government's design specifications. We note that this question is another one that has been the subject of considerable debate among courts and commentators. See, e.g., Brown v. Caterpillar Tractor Co., 741 F.2d 656, 662 (3d Cir. 1984); McKay, 704 F.2d at 450; Merritt, Chapman Scott Corp. v. Guy F. Atkinson Co., 295 F.2d 14, 16 (9th Cir. 1961); Note, Government Contract Defense, supra note 8, at 222 n. 318; Note, Manufacturer Discretion, supra note 8, at 208-09; Note, No Compulsion Required, supra, at 1084. Here, it is clear that FMC had no discretion to deviate from the design specifications provided by the government, although FMC could of course suggest modifications. See, e.g., Record at 316-21 (affidavit of William H. Stites).

Second, Bynum argues that the government contractor defense should only be available when the product at issue incorporates the newest technology or when the formulation of the designs requires "special military expertise." Again, we decline to limit the scope of the defense in the manner that Bynum suggests. That a decision requires consideration of no specialized knowledge does not preclude it from being, in the final analysis, a military judgment. Military authorities in deciding whether, when, and how to use military equipment must balance considerations of cost, time of production, risks of harm, the needs of our national defense, and other factors that touch upon such decisions. It is the balancing of these factors, not merely the nature of the information upon which that balance is predicated, that the government contractor defense seeks to exempt from challenge by servicemen in civilian courts. See In re "Agent Orange" Product Liability Litigation, 534 F. Supp. at 1054 n. 1. As long as the design specifications at issue are those for an instrumentality of war and were supplied by the government, the federal interests requiring the displacement of state law and judicial restraint in military matters are fully implicated.

What Bynum means by the phrase "special military expertise" is not altogether clear. We assume that the term refers to expert knowledge of military affairs, strategy, or performance requirements.

Again, we express no opinion on issues not directly raised by the instant case, such as whether military expertise might be needed to trigger the defense when the government merely approves a design furnished by the contractor. Cf. Shaw v. Grumman Aerospace Corp., 593 F.2d at 1074.

Lastly, Bynum argues that, in order to assert the government contractor defense, the military contractor should have to demonstrate that it warned the government about all defects or dangers of which the contractor knew or should have known. It is suggested that such a duty would increase the information flow between the military contractor and the government and allow the government to make better informed military determinations. While we do not wish to understate the importance of this goal, we think that the "known or should have known" standard is not warranted here. Such a duty would compel the military contractor to reevaluate the design specifications furnished by the government and to engage in testing not required under the government contract. Such reevaluations and additional testing would mean delay and an increase in defense costs not contemplated by the military authorities. Furthermore, as we noted in our discussion of market deterrence, the military has the ability to recognize safety problems or at least negotiate with the contractors for further testing, especially with respect to those design specifications that it generates itself. See McKay, 704 F.2d at 452. When to require additional testing of military equipment, and at what cost, are decisions that are better left to the military and the political branches of government. Id. at 451; In re "Agent Orange" Product Liability Litigation, 534 F. Supp. at 1055; Note, Government Contract Defense, supra note 8, at 223 n. 322.

At one point, Bynum also contends that the McKay court itself intended to adopt a "known or should have known" standard, at least with respect to open and obvious design defects. In support, Bynum refers to the Ninth Circuit's statement establishing the fourth element of the defense: "the supplier [must have] warned the United States about patent errors in the government's specifications or about dangers involved in the use of the equipment that were known to the supplier but not to the United States." McKay, 704 F.2d at 451. Upon reading the decision as a whole, however, we think that the McKay court intended only to adopt a duty to warn of known dangers and that the term "patent" in the decision must be construed as meaning "known." First, the statement which immediately follows the one relied upon by Bynum refers to the test as merely imposing a duty to warn of those dangers that are known: "The imposition of this duty to warn of known defects is necessary to enable the United States to balance the risks and benefits inherent in the use of the equipment." Id. (emphasis added). Second, all courts that have applied the McKay test, including the McKay court, have inquired only into whether the military contractor warned of defects that were actually known to it but not the government. See, e.g., Tillett, 756 F.2d at 599; McKay, 704 F.2d at 453; cf. Koutsoubos, 755 F.2d at 355 (stating that the McKay court merely rephrased the Agent Orange test — which contains only a duty to warn of known risks). We note further that a number of commentators have also construed McKay as requiring only the warning of errors in the government's specifications or any dangers involved in the use of the military equipment that were known to the military contractor but not to the government. See, e.g., Note, The Government Contractor Defense and Manufacturer of Military Equipment: McKay v. Rockwell International Corp., 21 Hous. L.Rev. 855, 872 (1984); Note, No Compulsion Required, supra, at 1080; Comment, 57 Temp. L.Q. 697, 702-03 (1984).

We recognize that in In re "Agent Orange" Product Liability Litigation, 597 F. Supp. 740, 849 (E.D.N.Y. 1984), the district court modified its earlier version of the government contractor defense, see supra note 14, to incorporate a limited "known or should have known" duty to warn. As the Agent Orange court explained, however, the heightened standard of conduct there was justified because (1) the manufacturer had years of experience with the components of the product before it was manufactured for the military and (2) the product was highly technical in nature. 597 F. Supp. at 849. The district court felt that a duty to warn only of known dangers under these circumstances would encourage the contractor to know as little as possible about the risks posed by the product. Id. In the instant case, however, the record reveals that the product involved, the M-548 cargo carrier, was designed by TACOM, a division of the Department of the Army, well versed in the design of and the problems inherent in tracked military vehicles. See supra note 20. The factors that prompted the modification of the Agent Orange test therefore are not present here.

It therefore remains for us to determine whether the district court erred in granting FMC's motion for summary judgment. Under rule 56(c) of the Federal Rules of Civil Procedure, summary judgment is proper when, viewed in the light most favorable to the non-moving party, "the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact." Fed.R.Civ.P. 56(c); see also Galindo v. Precision American Corp., 754 F.2d 1212, 1216 (5th Cir. 1985); Russell v. Harrison, 736 F.2d 283, 287 (5th Cir. 1984). The moving party carries the burden of showing the absence of a genuine issue of material fact. United States v. An Article of Drug, 725 F.2d 976, 984 (5th Cir. 1984). Once this burden is discharged, the burden shifts to the opposing party to demonstrate through opposing affidavits or other competent evidence that a genuine issue of material fact exists for trial. Id. Mere allegations or affidavits setting forth ultimate or conclusory facts and conclusions of law are not enough. Fed.R.Civ.P. 56(e); Galindo, 754 F.2d at 1216.

The first element of the McKay test, that the Feres-Stencel doctrine applies, is not here in dispute. Both parties agree that Bynum was a member of the armed services at the time of the accident and that the accident itself was service-related. Record at 350-51. Similarly, the second and third elements of the test have been established through the express stipulations of the parties. According to these stipulations, the M-548 was manufactured in accordance with precise design specifications furnished by the government and conformed to those specifications.

The National Guard, of course, is a component of the armed services of the United States, see 10 U.S.C. §§ 331 et seq.; 32 U.S.C. §§ 101 et seq.; Gilligan v. Morgan, 413 U.S. 1, 7, 93 S.Ct. 2440, 2444, 37 L.Ed.2d 407 (1973), and suits by members of the National Guard are subject to the Feres-Stencel doctrine. See Stencel, supra (applying doctrine to indemnity claim in action brought by National Guardsman against the manufacturer).

With respect to the fourth element of the test, however, the stipulations of the parties fall short. Although the parties have established through their stipulations that FMC knew of no patent dangers in the vehicle that were not known to the government, FMC can successfully assert the defense only if it knew of no dangers in the vehicle, either patent or latent, about which the government did not know. Nonetheless, we think that the affidavit of William Stites, the Engineering Manager of FMC's Ordnance Division, taken together with the stipulations, meets FMC's burden of demonstrating the absence of any genuine issue of material fact. In his affidavit, Stites stated that he had been employed by FMC since it agreed to manufacture the M-548 cargo carrier for the Army and that the M-548 was part of his responsibility. Stites further stated that he had read the deposition of Dr. Malcolm Newman, Bynum's expert witness, which contained Newman's description of how the accident probably occurred. In Stites' opinion, that type of event "has never occurred within the knowledge of FMC." Moreover, according to Stites, "FMC has never received from TACOM or any other source reports that the type of event has occurred with any M-548 or M113 family member manufactured by FMC." Record at 321.

We are perhaps being overly cautious on this point. Bynum conceded during oral argument that the parties intended to stipulate that the facts of the instant case satisfied all the elements of the test for the application of the government contractor defense set out in In re "Agent Orange" Product Liability Litigation, 534 F. Supp. at 1055. One of the elements of this test is "[t]hat the government knew as much as or more than the defendant about the hazards to people that accompanied use of [the military product]." Id. This element is equivalent to the fourth element of the McKay test. See supra note 28. Nonetheless, in the light of the principle that a court must draw all reasonable inferences in favor of a party opposing summary judgment, we find that the stipulations of the parties are insufficient by themselves to establish the fourth element of the McKay test.

Bynum did not introduce affidavits or any other evidence to counter FMC's prima facie proof that it had no knowledge of dangers involved respecting the M-548 that were not known to the government. Indeed, Bynum even on appeal does not allege that FMC possessed such knowledge. We, therefore, conclude that Bynum did not discharge his summary judgment burden and that FMC is entitled to judgment as a matter of law.

At oral argument, Bynum stated that he was unprepared for the renewed summary judgment contest because of his reliance upon the initial ruling of the court denying FMC's summary judgment motion on the ground that the government contractor defense was not available as a matter of law. Bynum, however, stipulated on the record that the district court had the power to reconsider the original summary judgment ruling. Further, Bynum failed to request additional time to obtain counter affidavits. Finally, the court reconsidered the summary judgment motion on the day of trial, some time after discovery had been ordered terminated and when all the evidence in the case should have been relatively accessible for use in demonstrating a genuine issue of material fact. Under these circumstances, Bynum is not entitled to relief from his summary judgment burden.

IV. CONCLUSION.

For the aforementioned reasons, we hold that, under the circumstances of this case, federal common law provides a basis for the fashioning of a government contractor defense. Because FMC has established all four elements of this defense, with no genuine issue of material fact remaining, the summary judgment in favor of FMC is affirmed.

AFFIRMED.