OPINION AND ORDER

MUKASEY, District Judge.

Plaintiff Carolynn Anne Stanley was injured while aboard a vessel she alleges was negligently designed and manufactured by defendant Bertram-Trojan, Inc. Defendant has impleaded the vessel's owner, Chris Blackwell, seeking indemnification or contribution. Blackwell moves for summary judgment dismissing the third-party complaint. For the reasons set out below, the motion is granted, and the third-party complaint is dismissed.

I

On June 20, 1988, Stanley sustained an injury to her leg aboard the vessel "SeaBee." During a pleasure cruise, a hatch door located on the floor of the vessel was dislodged, exposing an access with a sharp fiberglass border. The injury occurred when plaintiff fell through the access and the sharp edge severed nerves and muscle tissue in her calf. Plaintiff alleges that defendant negligently designed and manufactured the hatch door and access.

She filed this action in June 1989 in the Eastern District of New York against Bertram-Trojan, Island Records Company, and Chris Blackwell, the owner of the vessel. Following transfer of the action to this Court on motion of Island Records, plaintiff, by an "Order and Stipulation of Dismissal With Prejudice" agreed to dismiss her claims against Blackwell and Island Records. Bertram-Trojan then impleaded Blackwell, alleging that plaintiff's injury was due to Blackwell's negligent maintenance of the vessel. This court has jurisdiction pursuant to 28 U.S.C. § 1333. Blackwell now seeks dismissal of Bertram-Trojan's alternative claims for indemnity or contribution.

II

The doctrine of indemnity entitles a tortfeasor cast in judgment to shift the entire loss to a joint tortfeasor. W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 51, at 341 (5th ed. 1984). Indemnification is proper when: (1) an express agreement creates the right; (2) the right is inferred from the relationship between the parties; or (3) a tort has occurred, and there is great disparity in the fault of the parties. Araujo v. Woods Hole, Martha's Vineyard, Nantucket S.S. Authority, 693 F.2d 1, 2 (1st Cir. 1982). Because there is no express agreement between the parties and no prior relationship that may give rise to a right to indemnification, defendant's claim must be one for tort indemnity.

Tort indemnity is based on the difference between the "kinds" of fault for which the tortfeasors are responsible. Slattery v. Marra Bros. Inc., 186 F.2d 134, 138 (2d Cir.) (L. Hand, J.), cert. denied, 341 U.S. 915, 71 S.Ct. 736, 95 L.Ed. 1351 (1951). For instance, indemnification is warranted if the indemnitee's negligence is "passive," and the indemnitor's negligence is "active." Zapico v. Bucyrus-Erie Co., 579 F.2d 714, 719 (2d Cir. 1978) (Friendly, J.). Active negligence is the creation of an unreasonable risk; passive negligence is the failure to discover or remedy a risk created by a joint tortfeasor. Wedlock v. Gulf Mississippi Marine Corp., 554 F.2d 240, 243 (5th Cir. 1977); see Doca v. Marina Mercante Nicaraguense, S.A., 634 F.2d 30, 34 (2d Cir. 1980), cert. denied sub nom., Pittston Stevedoring Corp. v. Doca, 451 U.S. 971, 101 S.Ct. 2049, 68 L.Ed.2d 351 (1981). Under such circumstances "'both are liable to the same person for a single joint wrong . . . [and] the temptation is strong if the faults differ greatly in gravity, to throw the whole loss upon the more guilty of the two.'" Zapico, 579 F.2d at 718 (quoting Slattery, 186 F.2d at 138).

Plaintiff's claims in the underlying action are as follows:

(a) Defendant was negligent in (i) failing to provide a latch to hold a hatch door in place, (ii) manufacturing a hatch with sharp edges, and (iii) failing to warn of potential dangers posed by the hatch. (Amended Complaint ¶ 10-13);

(b) Defendant breached a warranty of fitness and merchantability. (Amended Complaint ¶ 15-19);

(c) The vessel was inherently dangerous; therefore defendant is strictly liable. (Amended Complaint ¶ 21-22).

These claims allege "active" misconduct by defendant. If plaintiff prevails on any of her claims, the result must be predicated on a finding that defendant acted negligently and was not merely passive in failing to discover another's negligence. Thus, even if it is found that Blackwell negligently maintained the vessel as alleged in the third-party complaint, both tortfeasors will have been found actively negligent and their relative fault would not "differ greatly in gravity." Where joint tortfeasors are actively negligent the law refuses to recognize any disparity in fault that would justify shifting the entire loss onto one of the tortfeasors. Gordon H. Mooney, Ltd. v. Farrell Lines, Inc., 616 F.2d 619, 625 (2d Cir.) ("because [defendant's] negligence contributed to the loss, [defendant] cannot obtain indemnity"), cert. denied, Maislin Transport of Delaware v. Farrell Lines, Inc., 449 U.S. 875, 101 S.Ct. 217, 66 L.Ed.2d 96 (1980); In re Complaint of American Export Lines, Inc., 568 F. Supp. 956, 964 (S.D.N Y 1983) ("A party may sue in admiralty for tort indemnity if he is vicariously liable for the culpable conduct of another, but he may not recover on this theory if he himself has been guilty of negligence."). Third-party defendant Blackwell's motion for summary judgment on defendant's indemnity claim therefore must be granted and that claim dismissed.

III

Contribution distributes loss among joint tortfeasors by requiring each to pay a proportionate share of the loss to one who has discharged the entire joint liability. W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 52, at 341 (5th ed. 1984). The doctrine is employed when there is concurrent fault, Cooper Stevedoring Co. v. Fritz Kopke Inc., 417 U.S. 106, 115, 94 S.Ct. 2174, 2179, 40 L.Ed.2d 694 (1974), and "'one of the joint tortfeasors has paid more than his fair share of the common liability.'" In re "Agent Orange" Product Liability Litigation, 818 F.2d 204, 207 (2d Cir. 1987) (quoting Northwest Airlines, Inc. v. Transport Workers Union of America, 451 U.S. 77, 87-88, 101 S.Ct. 1571, 1579, 67 L.Ed.2d 750 (1981)). This motion entails an application of the doctrine that has divided the circuits. Specifically, I must determine whether under federal admiralty law, a settlement between plaintiff and one tortfeasor bars an action for contribution by a joint tortfeasor against the settling tortfeasor.

This country's treatment of the subject begins in the mid-nineteenth century with The Schooner Catharine v. Dickinson, 58 U.S. (17 How.) 170, 15 L.Ed. 233 (1855). Drawing on the English maritime law of the time, Catharine held that damages arising from a maritime collision should be equally divided among joint tortfeasors, regardless of their degree of fault. This rule governed maritime actions until 1975 when the Supreme Court recognized that the United States was the only major maritime nation that continued to follow the mutual fault rule and that that rule yielded unjust results in most cases. The Court overruled Catharine and held:

[W]hen two or more parties have contributed by their fault to cause property damage in a maritime collision or stranding, liability for such damage is to be allocated among the parties proportionately to the comparative degree of their fault.

Another course correction followed in Edmonds v. Compagnie Generale Transatlantique, 443 U.S. 256, 99 S.Ct. 2753, 61 L.Ed.2d 521 (1979). There, the principle of joint and several liability was applied to admiralty cases. As a result, a plaintiff in admiralty may choose to sue one joint tortfeasor and recover all damages sustained. If the tortfeasor against which judgment is obtained is held liable for more than its proportionate share, contribution permits that tortfeasor to claim against the others and provides the mechanism whereby damages are adjusted to reflect fault. Edmonds, 443 U.S. at 260 n. 8, 99 S.Ct. at 2756 n. 8; see Cooper Stevedoring, 417 U.S. at 110, 94 S.Ct. at 2176; Self, 832 F.2d at 1546. There is some tension, however, between Edmonds and the proportional fault rule of Reliable Transfer. Edmonds suggests that as between the injured party and the tortfeasors, the tortfeasors must bear the loss, even if the end result is that one tortfeasor is liable for more than its fair share as measured by relative fault. "Contribution remedies the unjust enrichment of the concurrent tortfeasor, and while it may sometimes limit the ultimate loss of the tortfeasor chosen by the plaintiff, it does not justify allocating more of the loss to the innocent [plaintiff], who was not unjustly enriched." Edmonds, 443 U.S. at 271-72 n. 30, 99 S.Ct. at 2762 n. 30 (citations omitted) (emphasis added); Ebanks v. Great Lakes Dredge Dock Co., 688 F.2d 716, 721 (11th Cir. 1982), cert. denied, 460 U.S. 1083, 103 S.Ct. 1774, 76 L.Ed.2d 346 (1983).

Edmonds, however, left uncertain how courts should deal with settling tortfeasors. As the Ninth Circuit has stated, there are three possible ways to allocate liability when one or more tortfeasors settle:

(1) allowing an action for contribution against a settling tortfeasor by any other tortfeasor who has paid more that his equitable share of plaintiff's claim; (2) imposing a bar to contribution claims against a settling tortfeasor, perhaps in conjunction with a requirement that the settlement be in "good faith"; and (3) reducing the claim of the plaintiff by the pro rata share of a settling tortfeasor's liability for damages, which has the effect of eliminating any reason to sue a settling tortfeasor for contribution.

The first approach apportions losses equitably among the tortfeasors because the availability of suits for contribution ensures that each tortfeasor, even one who elects to settle, will bear the loss in proportion to its fault. However, a tortfeasor has no incentive to enter an individual settlement if it will remain vulnerable to suit based on plaintiff's claim. This approach, therefore, disserves the strong policy in favor of settlement. See Williams v. First Nat'l Bank, 216 U.S. 582, 595, 30 S.Ct. 441, 445, 54 L.Ed. 625 (1910); Marek v. Chesny, 473 U.S. 1, 10, 105 S.Ct. 3012, 3017, 87 L.Ed.2d 1 (1985) (discussing express policy in favor of settlement contained in Fed.R.Civ.P. 68). As a result, this approach has not been applied in admiralty cases. See Miller, 887 F.2d at 906.

Imposing a bar to contribution actions, the second approach discussed above, encourages settlement because the settling tortfeasor is no longer subject to suits arising out of plaintiff's claim. However, adoption of a settlement bar may exacerbate the tension between Edmonds and Reliable Transfer. In Reliable Transfer, as previously discussed, the Court held that loss is to be shared among tortfeasors in proportion to their fault in order to achieve a "'fair and equitable' allocation of damages." Reliable Transfer, 421 U.S. at 411, 95 S.Ct. at 1715-1716. In contrast, a settlement bar in conjunction with the rule of Edmonds would work to permit situations where loss is not borne according to fault. A tortfeasor could settle, and a joint tortfeasor, against whom a suit is prosecuted, could still be held liable under Edmonds for 100% of the loss. Imposition of a settlement bar to contribution would preclude any redistribution of liability and, in addition, would enable a plaintiff to receive a windfall in the amount of the total of the damage award plus settlement proceeds, less actual injury. Furthermore, the bar creates the incentive for collusion between the plaintiff and one or more of the tortfeasors to take advantage of a joint tortfeasor with a deeper pocket. See Miller 887 F.2d at 905.

The third approach, reducing plaintiff's claim against the non-settling tortfeasor by the pro rata share of the settling tortfeasor's liability based on fault, also has its shortcomings. This approach frees the settling tortfeasor from additional litigation and achieves a fair apportionment of damages among joint tortfeasors. To that extent, the policies in favor of settlement and a just division of damages are furthered. However, the approach may work against the interest of the injured party. For instance, a plaintiff might not settle for the full amount of a tortfeasor's insurance coverage if that amount is less than the tortfeasor's pro rata share of the damages. This is because at trial the non-settling tortfeasor's liability will be reduced by the full amount of the settling tortfeasor's share. Therefore, the plaintiff is encouraged to litigate until all tortfeasors are willing to bear their share of the loss. See Restatement (Second) of Torts § 886A cmt. m(3) (1979). In addition, determining a settling tortfeasor's share of fault presents problems of proof. In the situation where one or more tortfeasors settle, the question arises as to who must bear the burden of establishing the respective tortfeasors' fault. Regardless of who must bear the burden, it is likely that in order to reach an accurate determination discovery must be had against the settling tortfeasor. Again, this diminishes somewhat the incentive to settle because some of the expense and annoyance of litigation will continue to haunt the settling tortfeasor.

In choosing among these three imperfect solutions, the circuits agree that because of the strong policy in favor of settlement a tortfeasor with which a good faith settlement is reached is immune from suits for contribution. See Miller, 887 F.2d at 906. The uncertain issue is whether recovery against a non-settling tortfeasor can be limited by the settling tortfeasor's share of damages based on proportionate degrees of fault. In other words, it is unclear whether the second approach is to be applied alone or in conjunction with the third approach. The First, Fifth and Eleventh Circuits have imposed a settlement bar and prohibited reduction of the non-settling tortfeasor's liability, regardless of whether this permits inequitable distribution of liability or recovery in excess of actual injury. The Eighth Circuit permits reduction of liability based on relative fault.

The Fifth Circuit adopted the pro rata approach in Leger v. Drilling Well Control, supra. There, the non-settling defendants were held liable to the plaintiff to the extent of their respective percentages of fault. Leger, 592 F.2d at 1249. The Eleventh Circuit, however, declined to follow the Leger rule. In Ebanks v. Great Lakes Dredge Dock Corp., supra, plaintiffs were injured when a barge owned by their employer, Great Lakes, collided with a tanker owned by Chevron. Plaintiffs settled with Chevron. At trial with Great Lakes the jury decided that Chevron was 100% responsible for the disaster, and, as a result, there was no award of damages. Reversing, the court stated:

Since the plaintiff is entitled to recover . . . against either of several tortfeasors, without regard to the percentage of fault, it was error for the trial court to distract the juror's [ sic] attention by requiring it to allocate the degree of fault between the defendant and a non-party.

Regarding reallocation of liability among tortfeasors the Ebanks court went on to say that the issue "is to be tried at a different time between two live opponents." Id. On the facts of Ebanks, this statement is troubling because it implies that a suit for contribution may be allowed against a settling tortfeasor. Drake Towing Co. v. Meisner Marine Constr. Co., 765 F.2d 1060, 1067-68 (11th Cir. 1985), followed Ebanks, and again implied that a suit for contribution would be allowed against a settling tortfeasor. The court finally clarified its position in Self v. Great Lakes Dredge Dock Co., supra, by both prohibiting reduction of liability to reflect fault and holding that contribution cannot be had against a settling tortfeasor. 832 F.2d at 1547. Thus, the rule in the Eleventh Circuit is that non-settling tortfeasors are jointly and severally liable to the plaintiff, and actions for contribution against settling tortfeasors are prohibited whether or not the loss is distributed inequitably or the plaintiff recovers more than the actual injury warrants. The First Circuit has also adopted this approach, Joia v. Jo-Ja Service Corp., 817 F.2d 908, 915-17 (1st Cir. 1987), cert. denied sub nom., Boat Niagra Falls, Inc. v. Joia, 484 U.S. 1008, 108 S.Ct. 703, 98 L.Ed.2d 654 (1988), and in the Fifth Circuit allegiance to Leger is fading in favor of the Eleventh Circuit's rule. See Hernandez v. M/V Rajaan, 841 F.2d 582, 591 (5th Cir.), cert. denied, 488 U.S. 981, 109 S.Ct. 530, 102 L.Ed.2d 562 (1988).

By contrast, the Eighth Circuit has recently adopted the pro rata approach. Associated Electric Cooperative Inc. v. Mid-America Transportation Co., 931 F.2d 1266 (8th Cir. 1991) (citing Leger). In Associated Electric the court stated that the proportional fault approach "deters collusive settlements [and] . . . does not discourage defendants from settling." 931 F.2d at 1271. I agree that the proportional fault approach provides the best solution to the settling tortfeasor problem.

The First, Fifth and Eleventh Circuit cases adopting the settlement bar approach to the exclusion of the pro rata approach, claim to be based on the Supreme Court's Edmonds decision. These circuits discern "the philosophy governing Edmonds" to require that " any inequity which results from the implementation of a . . . damage award should be borne by the tortfeasors rather than by the [plaintiff]." Self, 832 F.2d at 1546 (citing Joia, 817 F.2d at 917) (emphasis added). Because the Court in Edmonds allocated to the tortfeasors the expense of apportioning liability, these circuits conclude that the Edmonds Court intended to shield personal injury plaintiffs from all similar "burdens." See Edmonds, 443 U.S. at 268, 99 S.Ct. at 2760. These burdens include having a jury "distract[ed]" by an instruction regarding pro rata liability, Ebanks, 688 F.2d at 722, and having to present evidence regarding the settling tortfeasor's fault. See Drake, 765 F.2d at 1067. However, permitting both inequitable apportionment of damages and recovery in excess of injury does not follow from any "philosophy" of the Edmonds decision that I perceive, and certainly is not required by the Edmonds holding.

The Court in Edmonds did not deal with a settling tortfeasor. See 443 U.S. at 268, 99 S.Ct. at 2760. Thus, Edmonds, although it holds that defendants were responsible for the apportionment of liability, did not directly address problems posed by inequitable apportionment of damages, and did not address at all the policy favoring settlement of claims or the problems posed by recovery in excess of injury. Moreover, the Court in Edmonds was applying the Longshoremen's and Harbor Workers' Compensation Act, 33 U.S.C. § 901 et seq., which includes a workers' compensation scheme that potentially limited plaintiff's recovery to an amount less than actual injury.443 U.S. at 270, 99 S.Ct. at 2761. Here, there is no statute that can limit recovery. In fact, in Edmonds the Court was faced with a scheme under which "inequity appear[ed] inevitable." Id. Certainly, in such a situation, it is reasonable to conclude that inequities should be borne by the tortfeasors. In the settling tortfeasor situation, however, inequity is not inevitable.

Nevertheless, it is true that Edmonds applied a policy of solicitude towards plaintiffs. However, the First and Eleventh Circuits' application of this policy appears to my eye to extend Edmonds so far beyond its intended scope as to generate an unnecessary conflict with Reliable Transfer. By contrast, as set forth below, it is consistent with both Edmonds and Reliable Transfer to reduce the claim against the non-settling tortfeasor.

First, the objection to the pro rata approach raised by the American Law Institute in the Restatement (Second) of Torts — that plaintiffs would be less likely to accept a settlement from a joint tortfeasor due to uncertainty regarding a possible reduction in total damages — is simply a restatement of the disincentive faced by any plaintiff who considers settlement. A plaintiff considering settlement must always consider the possibility that settlement will yield less than litigating a cause to completion, and then weigh that possibility against the costs of continuing to litigate, including the possibility of winding up with nothing at the end. The joint tortfeasor situation presents merely a slight variation of this often intractable but quite commonplace dilemma. It presents neither an additional disincentive to settlement nor a significant burden on a plaintiff.

Moreover, in electing to settle, a plaintiff presumably is acting voluntarily and expects to benefit by the decision to discontinue the litigation against one or more tortfeasors. Most settlements involve expedited receipt of funds and reduced cost of litigation. It is so obvious as to be virtually a tautology that settlement involves surrendering voluntarily the hope of a superb result in return for the certainty of a tolerable one; that's why they call it settlement. When settlement is voluntarily chosen, limiting the non-settling tortfeasors' liability to a level commensurate with their degree of fault does not unfairly burden a plaintiff. Plaintiff simply must choose between potentially beneficial courses of action. There is always the option of not settling with anyone, pursuing suit against only one of the parties, as in Edmonds, and enjoying the benefit of joint and several liability if it can be achieved. If both settlement and joint and several liability are permitted, plaintiff may settle with some of the tortfeasors and subsequently pursue litigation seeking 100% recovery against the non-settling tortfeasors. The only burden placed on plaintiff by requiring a choice would be plaintiff's inability to recover an amount in excess of what is warranted by plaintiff's injury.

The circuits which have excluded the pro rata approach do not explain why joint and several liability is necessary in situations where one or more tortfeasors have entered some type of agreement with plaintiff regarding discharge of claims. By choosing settlement, it seems plaintiff at least implicitly acknowledges that joint and several liability is unnecessary under the circumstances because the settling tortfeasor's liability has been limited. It is unnecessary for defendants to apportion damages in a situation where plaintiff has already chosen to do so. There is no more reason to allow joint and several liability when plaintiff has settled with a tortfeasor than there is reason to allow plaintiff to pursue joint and several liability against two non-settling tortfeasors in two separate actions.

Thus, implementation of the pro rata approach harmonizes Edmonds and Reliable Transfer. No additional burdens are placed on plaintiff, who simply has the choice of benefits. Thus, the philosophy of Edmonds is respected. At the same time, liability is apportioned fairly according to Reliable Transfer. There is no danger of an inequitable distribution of damages nor recovery by plaintiff in excess of injury. Edmonds requires solicitude for the personal injury plaintiff so as to assure as close to full recovery as possible, not prodigality so as to permit a windfall.

Of course, as noted in Drake Towing, supra, there must be some determination of the settling tortfeasor's degree of fault. See 765 F.2d at 1067. Evidence must be presented, and presumably this means that plaintiff must take up the task of minimizing the settling tortfeasor's fault, just as the remaining defendants will take up the burden of maximizing the settling tortfeasor's fault. That places no unfair burden on the plaintiff any more than it places an unfair burden on the remaining defendants. Again, plaintiff has voluntarily chosen settlement. Always available is joint and several liability, and leaving the burden of apportioning damages to the defendants. That option is all Edmonds requires. In cases such as this, plaintiff has chosen to accept some benefit in exchange for releasing a tortfeasor from suit. Thus, it does not seem unfair to require plaintiff to present evidence regarding relative fault. Moreover, it is unlikely that plaintiff will have a problem obtaining necessary evidence. As indicated by the settlement, plaintiff has some working relationship with the settling tortfeasor, and, in any event, the subpoena remains a potent discovery tool.

Although the need to litigate apportionment may present a slight disincentive to settlement, both to a plaintiff who would wish not to litigate the question and to a settling defendant who would wish not to provide evidence, the benefits of settlement generally outweigh any such disincentive. Plaintiff will have to litigate the apportionment question whether or not settlement is reached, and it may be easier and less expensive to obtain evidence from a cooperative party than from an adversary. Similarly, the settling defendant avoids trial and significant legal expense, and certainly would rather provide information to a party that has agreed to withdraw its claims than to an active adversary, as would occur in discovery absent settlement.

Finally, it is worth noting that apportioning liability according to degree of fault is consistent with the approach many states have taken. In New York, for instance, settlement with one tortfeasor reduces any claim against a non-settling defendant by the greater of the amount stipulated in the settlement or the settling tortfeasor's equitable share of the damages. N.Y. General Obligations Law § 15-108 (McKinney 1991). Moreover, the Uniform Comparative Fault Act, drafted in 1977, favors the pro rata approach as consistent with the modern system of comparative negligence. 12 U.L.A.Supp. 40, 52-53 (1989); see Miller, 887 F.2d at 905.

In admiralty cases the best rule is that actions for contribution are prohibited against a tortfeasor which has reached a good faith settlement with plaintiff. Any claim against a non-settling party is then reduced by the greater of the amount of settlement or the settling tortfeasors' equitable share of the damages. See Associated Electric, 931 F.2d at 1271. This provides no major disincentive to settlement, apportions damages fairly, and, in addition, respects the Edmonds policy of solicitude towards plaintiffs.

The only question left to decide is whether plaintiff Stanley and third-party defendant Blackwell reached a good faith settlement. The "Order and Stipulation of Dismissal With Prejudice," which I signed on November 7, 1990, states that plaintiff "dismisses with prejudice the actions against Island Records, Inc. and Chris Blackwell." A dismissal with prejudice constitutes a final judgment on the merits and bars any future suits based on the same cause of action. Nemaizer v. Baker, 793 F.2d 58, 60-61 (2d Cir. 1986) (Stipulation reading "this action is dismissed with prejudice" has res judicata effect.). Defendant has raised no hint of impropriety by the settling parties. Thus, all plaintiff's claims against Blackwell have been settled in good faith.

Betram-Trojan's third-party complaint is dismissed. Damages, if any, that are awarded plaintiff following trial will be limited to defendant's pro rata share of damages based on its proportional fault, if any.

SO ORDERED.