PATRICK E. HIGGINBOTHAM, Circuit Judge:
We face the question of whether the doctrine of comparative fault applies in a products liability suit maintained under the maritime jurisdiction of the federal courts. We are persuaded that it does.
I
Alfred Lewis was injured when working as a member of a crew furnished by his employer, Timco, Inc., to Atwood Oceanies, Inc. for work aboard Oceanies’ drilling barge, the Vicksburg. At the time of the accident, the Vicksburg was in Louisiana's territorial waters.
Lewis operated hydraulic tongs used to “make up” tubing joints to be placed in a well. These tongs were owned and supplied by Rebel Rentals, Inc. and were manufactured by Joy Manufacturing, Inc. On the day before the accident, equipment was accidentally dropped in the drilling hole. Edwards Rental and Fishing Tools, Inc. furnished an employee to retrieve the equipment from the hole with a special fishing tool. Lewis was using the hydraulic tongs to assist in the “make up” of the fishing tool. Because of a design defect, these tongs failed to shut off when Lewis released their throttle and a snubbing cable attached to the tongs wrapped around Lewis, seriously injuring him.
[1427] A trial to the court resulted in an award for Lewis’s serious and permanently disabling injuries. The trial court found multiple causes for the injury. It found that Lewis was negligent in attempting to make up the fishing tool joint without adjusting the length of the snubbing line. It found that the tongs manufactured by Joy Manufacturing had a design defect that allowed them to continue operating when the throttle was released. It also found that Rebel’s representatives were negligent in failing to instruct Lewis as to the proper method of synchronizing the tong controls. Finally, it found that the Edwards employee had been negligent in not advising Lewis to shorten the snub line. The trial court apportioned 20 percent of the fault each to Joy Manufacturing and Rebel Rentals, 10 percent to Edwards Rental, and 50 percent to Lewis.
On appeal a panel of this court affirmed all but the district court’s reduction of Lewis’s award against Joy Manufacturing by the amount of his fault. 697 F.2d 1252 (5th Cir.1983). We granted a petition to rehear en banc the manufacturer’s entitlement to the reduction. Lewis argues that Joy Manufacturing’s liability for the product defect should not be reduced by that part of his injury caused by his own negligence. He argues alternatively that if comparative fault be applied the trial court’s assessment of 50 percent was clearly erroneous. The panel having concluded that comparative fault was not to be applied did not reach the question of whether there was sufficient evidence to sustain that level of fault. We find that issue appropriate for decision by the panel and return the case to it for that review. We decide only that the trial court was correct in its decision that the maritime principle of comparative fault is applicable in maritime eases that urge strict liability for defects in products.
We will review comparative fault as applied under the maritime law, then turn to its application in products cases where liability rests on the principle of strict liability.1 Finally, we will explain the basic policy choice we make. We turn first to comparative fault in maritime jurisprudence, pausing to explain our jurisdiction and the relevance of state law.
II
The citizenship of the parties was not diverse and Lewis’s suit by the time of trial was footed solely upon maritime jurisdiction. There is such jurisdiction because the injury was sustained on board a drilling barge, a “vessel,” in the navigable territorial waters of Louisiana. In maritime tort cases courts traditionally apply principles of maritime law, as informed by common law tort developments, Sea-Land Services, Inc. v. Gaudet, 414 U.S. 573, 94 S.Ct. 806, 39 L.Ed.2d 9 (1974), unless a policy determination has been made by the Congress. Mobil Oil Corp. v. Higginbotham, 436 U.S. 618, 98 S.Ct. 2010, 56 L.Ed.2d 581 (1978). Admiralty courts make their own decisions but, true to legal analogical processes, do so with an awareness of other courts’ solutions to similar problems, sensitive to whether a “significant policy” of the state within whose territorial waters the injury occurred “would be frustrated by such an application.” See Watz v. Zapata Off-Shore Co., 431 F.2d 100, 113 (5th Cir.1970).
III
Admiralty courts have long engaged in the exercise of comparing plaintiffs’ negligence to both fault and non-fault based liability of defendants. For example, comparative fault is applied in the strict liability action for unseaworthiness, Pope & Talbot, Inc. v. Hawn, 346 U.S. 406, 408-09, 74 S.Ct. 202, 204, 98 L.Ed. 143 (1953), in personal injury actions under the Jones Act, 46 U.S.C. § 688, in actions brought under the [1428] Death on the High Seas Act, 46 U.S.C. § 766, and in longshoremen’s suits against vessels under the Longshoremen’s and Harbor Workers’ Compensation Act, 33 U.S.C. § 901 et seq. Gay v. Ocean Transport & Trading, Ltd., 546 F.2d 1233, 1238 (5th Cir. 1977). “The admiralty rule in personal injury cases is, in effect, one of comparative negligence.” G. Gilmore & C. Black, The Law of Admiralty 500 .n. 70 (2d ed. 1975).
Lewis's desire to except maritime products cases from this consistent application of comparative fault also overlooks the fact that maritime law traditionally resists doctrinal change that might balkanize its uniformity and generality. Most notably, courts applying maritime law have repeatedly rejected choice of law notions that would reference state tort doctrines. State workers’ compensation schemes were held to be inapplicable to personal injury claims arising from maritime related work on vessels in navigable waters. Southern Pacific Co. v. Jensen, 244 U.S. 205, 37 S.Ct. 524, 61 L.Ed. 1086 (1917). State negligence law was held not applicable in a maritime personal injury suit, Carlisle Packing Co. v. Sandanger, 259 U.S. 255, 42 S.Ct. 475, 66 L.Ed. 927 (1922), including state law defenses of contributory negligence, Pope & Talbot, Inc. v. Hawn, 346 U.S. at 409-11, 74 S.Ct. at 204-06. In 1948 Congress extended admiralty jurisdiction shoreward with the Extension of Admiralty Act, 46 U.S.C. § 740, which provides that maritime jurisdiction shall include loss caused by a vessel on navigable water even if the injury is finally suffered on land. Even resort to states’ wrongful death statutes ended with recognition of a general maritime law right of recovery for wrongful death. Moragne v. States Marine Lines, 398 U.S. 375, 90 S.Ct. 1772, 26 L.Ed.2d 339 (1970); see Matter of S/S Helena, 529 F.2d 744 (5th Cir. 1976).
In sum, comparative fault has long been the accepted risk-allocating principle under the maritime law, a conceptual body whose cardinal mark is uniformity. These values of uniformity, with their companion quality of predictability, a prized value in the extensive underwriting of marine risks, are best preserved by declining to recognize a new and distinct doctrine without assuring the completeness of its fit. We are persuaded that the fit within general maritime principles of a doctrine of strict liability for defective products without comparative fault would be uneven at best.
The Death on the High Seas Act, which encompasses claims for personal injuries caused by defects in products, illustrates the problems of not recognizing comparative fault in maritime products liability cases. Under DOHSA, the court is directed to “take into consideration the degree of negligence attributable to the decedent and reduce the recovery accordingly.” 46 U.S.C. § 766. If Lewis’s argument were accepted, when a worker’s death on the high seas was caused by a defective product, the recovery would be reduced on account of the worker’s negligence, but not when he was only injured. Moreover, because DOHSA applies to accidents occurring beyond a marine league from shore plaintiffs would be treated differently depending upon where a fatal accident occurred.
Other examples of its poor fit come quickly to mind in multi-party litigation so common to the admiralty practice. When a negligent plaintiff, negligent defendants, and the manufacturer of a defective product are all held jointly responsible for injuries, plaintiff’s negligence would diminish his potential recovery from the negligent defendants but not from the manufacturer. If the liability was joint and several, plaintiff could recover the entire amount of his damages from the manufacturer. From the plaintiff’s perspective, assuming the solvency of the manufacturer, it is as if there were no doctrine of comparative fault with respect to the negligent defendants as well. From the manufacturer’s perspective, contribution might be available, but somebody would bear more than his share of the damages. In other words, erosion of the comparative fault principle, once started in the products liability field, will cut at the legs of negligence as well.
[1429] The traditional doctrine of seaworthiness will also likely be affected. If a vessel is unseaworthy because a product was defective, we will be forced to decide whether to hold the manufacturer of the product to a stricter standard of liability than the vessel owner, traditionally a near insurer in cases of unseaworthy vessels. Even more taxing will be the categorization process as seamen attempt to escape the comparative fault of the traditional theory of unseaworthiness and label their case products cases. Ultimately, there would be the inquiry of whether a vessel is not itself a product. It takes little imagination to see, indeed predict, that should we reject comparative fault, many maritime torts of our circuit will become product cases with the companion problem that the courts of this circuit would be favored over more convenient courts by seamen with a choice of forum.
While the issue seems to be open in most circuits, our decision to apply comparative fault to a strict liability case controlled by the general maritime law is supported by the only other circuit court to expressly consider the issue. In Pan-Alaska Fisheries, Inc. v. Marine Construction & Design Co., 565 F.2d 1129 (9th Cir.1977), the Ninth Circuit, impressed by the recent endorsement of comparative fault by state courts and by the prevalence of comparative fault in maritime law, held that the damages suffered by a plaintiff in a maritime strict products liability case could be reduced by his fault’s contribution to his injury.
IV
Lewis offers two primary reasons why strict liability and comparative fault are unsuitable partners. Lewis first argues that comparing a defendant’s strict liability with a plaintiff's negligence is an “apples and oranges” effort. The argument is that while negligence focuses on a plaintiff’s personal conduct, the focus of the strict products liability action is on the condition of the product and not on the conduct of the defendant. The argument continues that this difference hinders apportionment of fault in that it requires a necessarily crude and essentially arbitrary allocation, given the task of comparing incomparable ideas.
The second and related argument against comparative fault is that it requires a trier of fact to hypothesize the fault of the defendant in an unstructured way in frustration of the allocating objective of enterprise liability. That objective is to place upon the manufacturer the burden of accidental injuries caused by its products, an objective accomplished in part by a rejection of the defense of contributory negligence. See Restatement (Second) of Torts § 402A, Comments c and n (1965). The rationale is that the manufacturer is in a better position than the user to absorb the economic loss by spreading it throughout the chain of distribution. Eventually, the cost is passed on to society in the form of an increased cost of the product. The effect of reducing a plaintiff's recovery by the amount of his fault, the argument goes, will be to reduce or remove the manufacturer’s incentive to produce safe products.
At a practical level, Lewis’s argument that negligence cannot be compared to strict liability fault overlooks the fact that such comparisons were already and inevitably required in this case. Here, apart from questions of Lewis’s own contribution to his injuries, the trial judge compared Joy Manufacturing’s strict liability fault with the negligent fault of Edwards Rental and Rebel Rentals. In short, Lewis’s proffered “conceptual problem has never bothered admiralty courts in applying the rule.” Owen & Moore, “Comparative Negligence in Maritime Personal Injury Cases,” 43 La.L.Rev. 942, 948 (1983).
Nor have the arguments persuaded common law jurisdictions for despite these conceptual attacks, see, e.g., Kinard v. Coats Co., 553 P.2d 835 (Colo.App.1976), the majority of state courts and federal courts sitting in diversity that have faced this issue have held that comparative fault, as adopted by the legislature or the courts, should be applied to actions founded on strict products liability. Their reasoning supports the application of comparative [1430] fault in maritime cases based on strict products liability.
Alaska was one of the first states to apply its judge-made comparative fault doctrine in a strict products liability case. In Butaud v. Suburban Marine & Sporting Goods, Inc., 555 P.2d 42 (Alaska 1976), the Alaska Supreme Court analyzed the purposes of the two doctrines and emphasized the anomalous situation that would exist “in a products liability case to have damages mitigated if the plaintiff sues in negligence, but allow him to recover full damages if he sues in strict liability, particularly where the complaint contains alternate counts for recovery in negligence, strict liability, and/or breach of warranty.” Id. at 46 (footnote omitted). The court concluded that “the public policy reasons for strict product liability do not seem to be incompatible with comparative negligence. The manufacturer is still accountable for all the harm from a defective product, except that part caused by the consumer’s own conduct.” Id.
This court, applying Mississippi law, reached a similar result in Edwards v. Sears, Roebuck & Co., 512 F.2d 276 (5th Cir.1976). The trial court there had instructed the jury to compare the conduct of the defendants and the plaintiffs, to decide if both contributed to the cause of the accident, and to then reduce plaintiff’s recovery to the extent his negligence contributed to the accident. Holding that “the trial court took the correct path through the thicket of strict liability and contributory negligence,” we remarked that “a noted commentator has suggested that the proper interaction between strict liability and contributory negligence ‘should be apparent on reflection. It is to apply a system of comparative fault of the pure type and apply it to strict liability as well as negligence.’ ” Id. at 290 (quoting Wade, “Strict Tort Liability,” 44 Miss.L.J. 825, 850 (1973)). Similarly, in West v. Caterpillar Tractor Co., Inc., 547 F.2d 885 (5th Cir.1977), we held, in response to the Florida Supreme Court’s answer to a certified question, that “strict liability ... lies in bystander actions, and that want of ordinary due care — in its comparative negligence form — is a defense....”2 We thus affirmed a 35 percent reduction in plaintiff’s judgment to reflect his fault.
The reasoning in Butaud, Edwards, and West was amplified by the California Supreme Court in Daly v. General Motors Corp., 20 Cal.3d 725, 144 Cal.Rptr. 380, 575 P.2d 1162 (1978). Extending its judge-made “pure” comparative fault system to strict liability cases, the court rejected the same arguments now made by Lewis and answered the suggestion that the two concepts cannot be merged:
The inherent difficulty in the “apples and oranges” argument is its insistence on fixed and precise definitional treatment of legal concepts. In the evolving areas of both products liability and tort defenses, however, there has developed such conceptual overlapping and interweaving in order to attain substantial justice.... We think, accordingly, the conclusion may fairly be drawn that the terms “comparative negligence,” “contributory negligence” and “assumption of risk” do not, standing alone, lend themselves to the exact measurements of a micrometer-caliper, or to such precise definition as to divert us .from otherwise strong and consistent countervailing policy considerations. Fixed semantic consistency at this point is less important than the attainment of a just and equitable result.
575 P.2d at 1167-68.
The court also rejected the arguments that applying comparative fault would [1431] erode the protection afforded by the strict liability doctrine and reduce the incentive to produce safer products. In response to the concern that comparative fault would diminish protection of consumers, the court emphasized that plaintiffs will continue to be relieved of proving that a manufacturer or distributor was negligent and that their recovery will be reduced only to the extent their lack of reasonable care contributed to the injury. Id. at 1168. The court also reasoned that manufacturers, who cannot assume the users of a defective product to be blameworthy, will not face reduced incentives because their continuing liability for a defective product “will be lessened only to the extent that the trier finds that the victim’s conduct contributed to his injury.” Id. at 1169. Moreover, the court noted that the extension of comparative principles to strict liability actually would produce the “felicitous result” of relieving the inequities associated with absolute defenses that provide windfalls to manufacturers. Id.
Following the reasoning of Daly, the Third Circuit in Murray v. Fairbanks Morse, 610 F.2d 149 (3d Cir.1979) (applying Virgin Islands law) also responded to the suggestion that a trier of fact will be unable to apportion fault between a negligent plaintiff and a strictly liable defendant. Though conceding that “there is no proven faulty conduct of the defendant to compare with the faulty conduct of the plaintiff,” the court noted that:
In apportioning damages we are really asking how much of the injury was caused by the defect in the product versus how much was caused by the plaintiff’s own actions.... Although fault, in the sense of the defendant’s defective product or the plaintiff’s failure to meet a standard of care, must exist before a comparison takes place, the comparison itself must focus on the role each played in bringing about the particular injury.
Id. at 159-60 (footnote omitted).
The accuracy of the court’s observation in Murray is seen when one looks at possible jury questions. Within its broad discretion in the manner of instructing the jury a district court might sequence Rule 49 interrogatories as follows: (1) was the product defective; (2) was it a cause of injury to plaintiff; (3) was the plaintiff at fault; (4) was plaintiff’s fault a cause of plaintiff’s injury; and (5) the percentage of plaintiff’s injury caused by plaintiff’s fault. The result then is that when the jury “compares fault” the focus is upon causation. It is inevitable that a comparison of the conduct of plaintiffs and defendants ultimately be in terms of causation. “Fault” that did not cause injury is not relevant.
An increasing number of courts have been persuaded by the policy considerations articulated in such cases as Daly and Murray. See, e.g., Trust Corp. of Montana v. Piper Aircraft Corp., 506 F.Supp. 1093 (D.Mont.1981) (applying Montana law); Star Furniture Co. v. Pulaski Furniture Co., 297 S.E.2d 854 (W.Va.1982); Kaneko v. Hilo Coast Processing, 654 P.2d 343 (Hawaii 1982); Kennedy v. City of Sawyer, 4 Kan. App.2d 545, 608 P.2d 1379, 1386 (Ct.App.), aff’d, 228 Kan. 439, 618 P.2d 788 (1980); Baccalleri v. Hyster Co., 287 Or. 3, 597 P.2d 351 (1979). Most recently, the Texas Supreme Court in Duncan v. Cessna Aircraft Co., 26 Tex.S.CU. 507 (Tex. July 16, 1983) applied comparative fault to strict products liability cases despite a comparative contribution statute otherwise held inapplicable to strict liability cases. In short, at the same time that much judicial learning is moving towards comparative fault in strict liability cases, Lewis would have us abandon those principles in the maritime law.
Finally, we inquire whether any strong policy of Louisiana, in whose territorial waters this accident occurred, would be frustrated by adopting comparative fault in maritime products cases. While Louisiana courts do not appear to have applied comparative fault principles to strict products liability cases, the state has no “significant policy” against doing so. Its legislature adopted a comparative fault statute that became effective on August 1, 1980. See Acts 1979, No. 431 (amending LSA-C.C. [1432] Art. 2323). Unlike some comparative fault statutes that expressly apply only to actions based upon negligence, see e.g., Kirkland v. General Motors Corp., 521 P.2d 1353 (Okl. 1974), Louisiana’s statute applies “[w]hen contributory negligence is applicable to a claim for damages.” It proportionately reduces recovery when “a person suffers injury, death or loss as the result partly of his own negligence and partly as a result of the fault of another person or persons....” LSA-C.C. Art. 2323. We have recently certified to the Louisiana Supreme Court the question whether Louisiana law recognizes contributory negligence as a defense to a products liability action, see Bell v. Jet Wheel Blast, 709 F.2d 6 (5th Cir.1983). See also Hyde v. Chevron U.S.A., Inc., 697 F.2d 614 (5th Cir.1983); Plant, “Comparative Negligence and Strict Tort Liability,” 40 La.L.Rev. 403 (1980). While we may be uncertain in this reading of Louisiana law, we are confident that recognition of comparative fault in products cases will not “frustrate” a dedicated policy of Louisiana.
V
It is relevant to an analysis3 of how a rule allocates liability for accident losses resulting from use of a product to consider: (1) short-term and long-term cost; (2) amount of use of the product in the economy or “activity”; and (3) cost of administering the rules of liability. It is relevant because fault has both an ethical and an efficiency dimension. The latter is expressed by asking which party can prevent the injury at the least costs.
The short-term costs are the immediate expenditures to avoid accidents as well as the immediate costs of accidents themselves. Of course the two primary actors influenced by the rule choice are the manufacturer and the user. The manufacturer will alter its product to avoid an accident if the manufacturer’s share of the expected cost of the accident (coverage cost times the probability it will occur) exceeds the cost of altering the product. A system of strict liability with comparative fault includes in the manufacturer’s share of the accident costs only those costs caused by product defects. In that case the manufacturer will have the correct economic incentive to adjust the design of the product to minimize accident costs caused by the design. A system of strict liability with no comparative fault would add to the manufacturer’s share those accident costs caused by negligent use and not by any product defect. This increase in the manufacturer’s share would result in an increased, and therefore inefficient, level of expenditures on preventive measures.
The situation with respect to the user’s expenditures is precisely complementary. The user will intentionally alter his use of the product only if his perceived cost of altering his use to avoid an accident is less than his expected cost from an accident resulting from his failure to alter his behavior. The inclusion of comparative fault will affect user behavior in a manner that results in a more efficient utilization of resources. Under simple strict liability, as proposed by the plaintiff, the user has no economic incentive to avoid an accident that he could avoid more cheaply than the manufacturer.
Besides affecting long-term research for safe products and the immediate decisions on how much to invest in preventive measures, rules of liability affect the level of product use. When the liability for blameless accidents is placed on the manufacturer, the price of products whose use results in high accident costs will go up relative to [1433] those whose use results in small accident costs. The use of the comparative fault standard reduces the risks of non-negligent users indirectly paying for negligent users. The comparative fault standard allows the price of the product to reflect the cost of its non-negligent use. Hence a comparative fault standard allows the economically efficient amount of the product to be used. If, for example, the use of a particular piece of equipment has resulted in several costly accidents due to the negligence of the user, the cost of the product will not be driven up because of the expected cost of accidents for which the producer will be or has been liable. The producer will not have to charge non-negligent consumers a premium to cover the liability from accidents by negligent users. The proper use of a safe product will not be stifled by negligent use.
The final economic consideration in choosing a rule of liability is the cost of administering the system. It might appear that strict liability without comparative fault would be less expensive to administer both because it simplifies the issues at litigation and because it removes uncertainty thereby facilitating settlements, which are cheaper than trials. But see United States v. Reliable Transfer Co., Inc., 421 U.S. 397, 408 n. 13, 95 S.Ct. 1708, 1714 n. 13, 44 L.Ed.2d 251 (1975). The matter, however, is more complex: by increasing the certainty of victory, if it does, strict liability may increase the plaintiff’s willingness to spend money on litigation and decrease his willingness to settle. There is no indication that strict liability with comparative fault would increase cost.
VI
We are persuaded that general considerations of fairness and efficiency support a comparative fault defense in products liability actions. In maritime suits, these considerations are bolstered by the historical reliance on comparative fault as integral to an essentially uniform and unitary body of law. We hold that it governs here, AFFIRM the district court’s application of maritime principles of comparative fault, and RETURN the case to the panel for its review of Lewis’s assertion that the level of found fault was not supported by the evidence.