concurring in part and concurring in the judgment.
I join Parts I and III of the court’s opinion and concur in the result reached in Part II of that opinion.
Defendants found to have violated § 10(b) of the Securities Exchange Act of 1934 are jointly and severally liable to the plaintiff, who may collect the entire judgment from less than all of the defendants, even one of them. The section has been construed, however, to give defendants a right of contribution. Musick, Peeler & Garrett v. Employers Ins., — U.S. -, 113 S.Ct. 2085, 124 L.Ed.2d 194 (1993). Thus, if a plaintiff collects the entire judgment from one of two defendants, that defendant may collect from the other that portion of the damages attributable to the other defendant’s relative fault; *930that contribution under § 10(b) is indeed calculated according to relative fault rather than some other formula is outlined in Part III below, but the court today clearly agrees with this construction of the § 10(b) contribution right. See Majority Op. at 925-926.
In this case, several defendants were sued. The plaintiff, appellee here, settled with some, but not all of the defendants. The settlement agreement provided that any judgment entered against the nonsettling defendants would be reduced by the amount of the settlement and that any order approving the settlement must bar the nonsettlors from seeking contribution from the settlors. The district court approved the settlement, and its order contained the bar order which was a condition of the settlement. The issue in this ease is whether the order impermissibly interfered with the contribution rights of the nonsettling defendants. I agree with the court today that it did, but not for the reasons the court offers. As I see it, the order interfered with nonsettling defendant’s contribution rights because the pro tanto credit it incorporated did not, and could not, necessarily reduce any judgment that might be entered against the nonsettlors to the amount flowing from the fault of the nonsett-lors and, hence, may have forced them to pay damages more equitably attributable to the settlors.
I
McDermott, Inc. v. AmClyde, — U.S. -, 114 S.Ct. 1461, 128 L.Ed.2d 148 (1994), convincingly explains why the pro tanto credit with a bar order infringes upon nonsettling defendants’ relative fault contribution rights. In McDermott, the parties asked the Court to decide how liability should be distributed in admiralty cases after a partial settlement. The Court considered three alternatives identified by the American Law Institute:
“(1) The money paid extinguishes any claim that the injured party has against the party released and the amount of his remaining claim against the other tortfea-sor is reached by crediting the amount received; but the transaction does not affect a claim for contribution by another tortfeasor who has paid more than his equitable share of the obligation....
“(2) The money paid extinguishes both any claims on the part of the injured party and any claim for contribution by another tort-feasor who has paid more than his equitable share of the obligation and seeks contribution. ...
“(3) The money paid extinguishes any claim that the injured party has against the released tortfeasor and also diminishes the claim that the injured party has against the other tortfeasors by the amount of the equitable share of the obligation of the released tortfeasor.”
— U.S. at -, 114 S.Ct. at 1465 (footnote omitted).1 The first ALI option essentially involves providing the nonsettling defendant a pro tanto judgment credit while leaving him free to pursue a contribution claim against the settling defendant. The second closely mirrors the regime the trial court approved in the § 10(b) action now before us: the nonsettling defendant is given a pro tan-to credit and he is barred from ever seeking further contribution against the settling defendant. The third option describes how fault is allocated when a proportionate share credit is employed.
In the process of analyzing which of these options should be utilized in admiralty cases, the Court acknowledged that admiralty defendants are jointly and severally liable to victorious plaintiffs. See McDermott, — U.S. at -, 114 S.Ct. at 1471. Joint and several liability, however, does create obvious inequities: one marginally culpable defendant may be held responsible for satisfying all of plaintiffs damages. To help ameliorate such inequities without sacrificing joint and *931several liability’s objective of fully compensating the plaintiff for his injury, the Court has generally permitted admiralty defendants to pursue contribution actions from one another. See Cooper Stevedoring Co. v. Fritz Kopke, Inc., 417 U.S. 106, 94 S.Ct. 2174, 40 L.Ed.2d 694 (1973). Since United States v. Reliable Transfer Co., 421 U.S. 397, 95 S.Ct. 1708, 44 L.Ed.2d 251 (1975), liability in admiralty actions has ordinarily been assigned to each responsible party according to his relative fault; consequently and unsurprisingly, contribution in admiralty is calculated according to relative fault principles and, thus, a defendant forced (by operation of joint and several liability) to pay plaintiff more than his equitable share of a damage award is allowed to seek monies from any defendant who has underpaid his equitable share. See, e.g., Leger v. Drilling Well Control, Inc., 592 F.2d 1246 (5th Cir.1979).
Recognizing that Reliable Transfer attempts to assure an equitable apportionment of liability between all responsible wrongdoers, the Court in McDermott turned to adjudge which of the three ALI options would be most consistent with its aim. Its discussion of the second option — a pro tanto credit without an opportunity to pursue contribution — is especially pertinent to our own analysis since, again, this is essentially the option the settlement agreement contemplated and the trial court approved in the securities action before us. The Court rejected this option in McDermott, holding it to be in irremediable conflict with Reliable Transfer. Provided with no opportunity to seek contribution from settling defendants, and afforded a judgment credit equal only to the actual dollar amount of the settlor agreement with plaintiff (and not his equitable share of the damages assessed at trial), a nonsettling defendant in such a regime is, the Court noted, often left liable for paying damages not in any way equitably attributable to him. The Court explained its point by reference to an example:
Suppose, for example, that a plaintiff sues two defendants, each equally responsible, and settles with one for $250,000. At trial, the non-settling defendant is found liable, and plaintiffs damages are assessed at $1 million. Under the pro tanto rule, the nonsettling defendant would be liable for 75% of the damages ($750,000, which is $1 million minus $250,000). The litigating defendant is thus responsible for far more than its proportionate share of the damages.
McDermott, — U.S. at - n. 14, 114 S.Ct. at 1467 n. 14. In sum, then, rather than helping assure an equitable distribution of liability (within an admittedly joint and several liability scheme), the Court found ALI Option 2 fundamentally at odds with this objective. Instead of attempting a distribution of fault in accord with relative fault, it found such an option would often lead in the opposite direction, “to inequitable apportion-ments of liability, contrary to Reliable Transfer.” Id. at -, 114 S.Ct. at 1468.2
Accepting that the § 10(b) contribution right attempts to ensure that liability in securities cases is distributed according to relative fault much as Reliable Transfer attempts to ensure that result in admiralty, it necessarily follows that it would be improper for a court to employ a pro tanto credit with a bar order in a § 10(b) action just as the combination is impermissible in admiralty cases. Just as ALI Option 2 leads to results “contrary to Reliable Transfer,” — U.S. at -, 114 S.Ct. at 1468, so too the settlement agreement before us is at odds with a contribution right that aims to ensure the same fundamental objective as Reliable Transfer.3
*932Appellees and the trial court themselves concede that employing an untamed pro tan-to credit with a bar order is likely to leave nonsettling defendants liable for more than their jury-determined share of liability. Rather than abandoning the option, however, they propose a remedy. They suggest that, before the bar order is entered, a “good faith hearing” be held. At such a hearing a judge would review the proposed settlement to assure that it represents a fair forecast of the nonsettling defendant’s relative share of the judgment; this would help keep to a minimum the number of instance in which a nonsettlor is left with the tab for a portion of the judgment properly attributable to settling defendants.
This proposed remedy is, however, manifestly inadequate. As the Court noted in McDermott, good faith hearings simply “cannot fully remove the potential for inequitable allocation of liability. First, to serve their protective function effectively, such hearings would have to be minitrials on the merits, but in practice they are often quite cursory. More fundamentally, even if the judge at a good-faith hearing were able to make a perfect forecast of the allocation of liability at trial, there might still be substantial unfairness when the plaintiffs success at trial is uncertain.” McDermott, — U.S. at -, 114 S.Ct. at 1468 (footnotes omitted).4
In something of a last gasp, appellees argue that, even if it does compromise appellants’ § 10(b) contribution right, using a pro tanto rule in conjunction with a bar order is so successful at encouraging settlement that it should still be adopted. In effect, appel-lees suggest that we should drop our insis-fence on the full vindication of the nonsett-lor’s contribution rights in order to promote the maximum number of settlements and, with them, to enhance judicial efficiency. This is another unpersuasive suggestion. To begin, I am not convinced we are empowered to sacrifice nonsettling defendants’ recognized statutory right to contribution under § 10(b) merely to advance the efficiency interest in promoting additional partial settlements. Even if we were so empowered, McDermott raises substantial questions regarding whether limiting nonsettlors’ contribution claims is, in fact, any better at promoting settlements than providing a proportionate share judgment credit. — U.S. at -, 114 S.Ct. at 1468-69.
II
Unlike the court today, I think analogy to McDermott suffices to demonstrate why the trial court must be reversed. The entry of a pro tanto credit with a bar order creates a significant inconsistency with nonsettling defendants’ relative fault contribution rights under § 10(b) just as it does with Reliable Transfer. We have no need to explore the All Writs Act or other potential sources of authority for the bar order entered here: whatever authority may or may not have existed for the order, McDermott illustrates clearly why its combination with a pro tanto credit cannot be sustained. In sum, then, a far more economical route to the same result the court reaches does exist and it is one that has the virtue of responding directly to the arguments raised by the parties and trial court before us.5
*933Having overruled the trial court’s approval of the settlement agreement, I agree with the court that we have no need or authority-today to direct the entry of an alternative judgment credit. The settlement agreement in this ease was conditioned on the approval of a pro tanto credit with a bar order; given that these conditions cannot be provided, a partial settlement no longer exists and the case must be remanded for further proceedings.
The court does press on, though, to offer some guidance to the parties and the court below by commenting on other possible settlement arrangements and whether they would or would not offend nonsettling defendants’ § 10(b) contribution rights. The court, for instance, mentions the option of supplying nonsettlors a pro tanto credit without also employing a bar order (essentially ALI Option 1). See Majority Op. at 923, 924. The Supreme Court in McDermott, however, frowned on this alternative, noting that
[i]t discourages settlement, because settlement can only disadvantage the settling defendant. If a defendant makes a favorable settlement, in which it pays less than the amount a court later determines is its share of liability, the other defendant (or defendants) can sue the settling defendant for contribution. The settling defendant thereby loses the benefit of its favorable settlement.
— U.S. at -, 114 S.Ct. at 1467 (footnote omitted).
Though the court today discusses the possibility of the plaintiff indemnifying the settling defendant against contribution actions in ordér to make settlement more attractive under Option 1, see Majority Op. at 924, the Supreme Court in McDermott clearly thought indemnity would not make the option more attractive: “The plaintiff can mitigate the adverse effect on settlement by promising to indemnify the settling defendant against contribution.... This indemnity, while removing the disincentive to settlement, adds yet another potential burden on the courts, an indemnity action between the settling defendant and plaintiff.” — U.S. at -, 114 S.Ct. at 1467.6
At places in its opinion today the court also suggests that a proportional fault judgment credit might be employed. See, e.g., Majority Op. at 926 (“if a court or jury properly decides the settling defendants’ share of the fault and somehow credits that amount to the nonsettling defendants, ... [a bar order] probably would [be] authorize[d].”). After making seemingly favorable comments about the use of a proportionate share judgment credit, however, the court insists that any proportionate fault reduction be “properly” determined and then proceeds to suggest that this may be an impossible task: “[t]he district court may not have had jurisdiction to [allocate a portion of § 10(b) damages to settling defendants], especially since the non-settling defendants had not filed cross-claims for contribution.” Majority Op. at 926-927.
It is difficult to understand why the court questions the power of a trial judge to cany out a settlement agreement that provides for a proportionate share credit. To begin, it must be admitted that such a credit is fully consistent with the purposes behind nonset-tling defendants’ relative fault contribution rights under § 10(b), much as it is with Reliable Transfer’s goals: far from foisting on the nonsettling defendant damages for which he is not responsible, this option assures nonsettlors they will ordinarily pay just their equitable share of the judgment. See McDermott, — U.S. at —-—, 114 *934S.Ct. at 1466-67.7 One of our sister circuits has already made just this point in the process of approving the proportionate share judgment credit for use in § 10(b) cases. See Franklin v. Kaypro Corp., 884 F.2d 1222 (9th Cir.1989).8
As for the absence of settling defendants who have been dismissed, they are no doubt protected from further liability since plaintiff has dismissed his action against them and nonsettling defendants receive a credit for the nonsettlors’ full share of any judgment awarded, thus obviating any need for them to pursue any contribution action. Furthermore, it is truly the plaintiff, not the settlors, who will have a stake in the relative fault assigned to the settling defendants (since under a proportionate share regime the plaintiff is responsible for any fault attributed by the jury to the settling defendants that exceeds the amount of the settlement agreement, see supra n. 7); and, there is no doubt that the plaintiff will litigate the issue fully with the help, if need be, of the settlors as witnesses. In sum, then, it is the will of the settling parties that the respective share of the damages caused by the nonsettlors and the settlors will be determined at trial, without the presence of the settlors, and I see no reason to interfere with such an arrangement.9
With regard to the court’s suggestion that nonsettlors must file contribution claims against settlors before a court might have “jurisdiction” to effectuate a proportionate share credit, I am mystified. As I have already argued, the settlors have no interest in the amount of fault assigned them by the jury; they are already relieved from any responsibility beyond the settlement agreement. The degree of the settlors’ fault is, however, highly relevant to the remaining parties’ dispute over the size of the judgment credit plaintiff will have to afford nonsettling defendants. I see no reason why a court would be powerless to consider the question as part of its resolution of the judgment credit required and the ultimate balance of obligations between the remaining parties. To require, as the court apparently would, a formal contribution action be filed — to mandate the presence of settling defendants— would only be a waste of time and effort, truly a judicial exercise in elevating form over substance.10
If any doubt lingers on this point, precedent should dispel it. The Court in McDer-mott held that a trial judge has the power to enforce a proportionate share credit regardless whether a contribution claim is filed in admiralty actions. See — U.S. at -, 114 S.Ct. at 1469. The Uniform Comparative Fault Act authorizes trial judges in states that adopt its measures to do the same. See 12 U.L.A. 50, 57 (1993 Supp.). The Ninth Circuit, too, has specifically ruled that a trial court is empowered to enforce a proportionate share credit without the presence of settling defendants in § 10(b) actions. See *935Franklin, 884 F.2d at 1231.11 Other courts have also apparently not been troubled with this practice, applying proportionate share credits without ever mentioning any Article III concerns, which, of course, every federal court has a duty to raise and consider for itself at every stage of a litigation. See, e.g., In re Brooklyn Navy Yard Asbestos Litig., 971 F.2d 831 (2d Cir.1992); Austin v. Raymark Indus., Inc., 841 F.2d 1184 (1st Cir.1988).
In the face of this authority, the court suggests only that the “jurisdictional” issue is less acute in admiralty cases than securities controversies. It contends that “[ljiability in Rule 10b-5 cases is strictly joint and several and is never allocated among individual defendants in deciding the plaintiffs claim.... The [settling] defendants’ relative fault is therefore irrelevant to the plaintiffs claim.” Majority Op. at 927. By comparison, “admiralty law is comparative, which allows courts to allocate fault among defendants in deciding the plaintiffs claim-” Id. at 927-928.
This argument is unsatisfying. To begin, it fails to account for the fact that substantial authority outside admiralty has suggested that courts are empowered to decide settlors’ fault in their absence as part of applying a proportionate share credit. The court creates a plain split with Franklin with regard to a trial court’s power to enforce a proportionate share credit without a contribution claim, a case specifically decided in the § 10(b) context, but it does so without comment. It necessarily intimates that the Uniform Comparative Fault Act is wrong as well, but never addresses the point directly. Furthermore, even with regard to its attempt to distinguish admiralty from securities cases, the court fails. Liability in admiralty is fully joint and several, just as it is in tort actions under the Uniform Comparative Fault Act, see 12 U.L.A. at 50 (1993 Supp.), and securities actions under § 10(b). See, e.g., McDermott, — U.S. at -, 114 S.Ct. at 1471. It is simply that in each regime— admiralty, tort law under the Uniform Comparative Fault Act, and § 10(b) — the joint and several liability rule has been tempered with an assurance (stemming from Reliable Transfer in admiralty and from the relative fault contribution right of both the Uniform Comparative Fault Act and § 10(b)) that fault will be distributed according to relative fault when possible.12 Deciding settling defendants’ proportionate share of the liability in the primary action between plaintiff and nonsettlors (regardless of the existence of a contribution action) follows sensibly in each regime from this shared ambition and itself does no damage to the full vindication of joint and several liability principles.13 Not one of the cases cited by the court for support, see Majority Op. at 927, ever holds that deciding absent settlors’ share of the fault would be in any way inimical to the full satisfaction of joint and several liability rules in a § 10(b) action.14
*936Finally, though some portions of the court’s opinion today might be read to suggest otherwise, I see no reason why a settlement agreement providing a proportionate share judgment credit but conditioned on the entry of a bar order could not be approved. The bar order in such a scenario would surely be unnecessary given that a proportionate share credit provides a nonsettling defendant all that his relative fault contribution claim does — given that such a credit makes any further contribution actions superfluous “because the nonsettling defendants pay no more than their share of the judgment.” McDermott, — U.S. at -, 114 S.Ct. at 1466. If, however, the parties wish to add such an order to their agreement, there would be ample authority for its issuance. Surely the federal courts are empowered to construe the contribution rights guaranteed by § 10(b), to decide whether a particular judgment credit satisfies those rights, and, if it does, to ensure that nonsettlors recognize that the credit extinguishes any contribution rights they may have had against the set-tlors. All of this seems to me to involve no more than a problem of statutory construction and a decision to employ the court’s traditional equitable powers to issue injunctions to protect its decrees against relit-igation.15 Moreover, while I agree with the court that the federal policy interest in promoting settlement cannot override nonset-tling defendants’ right to contribution, that interest may supply some additional basis for permitting the entry of a bar order when, coupled with a proportionate share credit, it would in no way trench upon nonsettlors’ protected contribution right, but would merely provide an additional assurance to the settlor that he will never have to defend a contribution action. Certainly our sister circuit in Franklin, 884 F.2d at 1229, held this to be the ease and I see no reason to express disagreement with its decision in this case.16
*937III
One matter requiring explanation remains. I have argued that the settlement agreement must be rejected in large part because a pro tanto credit with a bar order creates an inconsistency with the nonsettling defendants’ relative fault contribution right; I have noted, too, that the proportionate share judgment credit creates no such inconsistencies with the § 10(b) right. The critical premise to both conclusions is, however, that the § 10(b) contribution right is indeed one calculated according to relative fault.
The court today accepts the premise— readily admitting that contribution is calculated according to relative fault under § 10(b), see Majority Op. at 925-26 — and one might be tempted to leave the matter at that. It seems to me, however, that the point deserves more attention than either the court or I have, to this point, provided. It does so because, though the Supreme Court recognized that a right to contribution does exist under § 10(b) in Musick, Peeler, it has to date left as an open question how that contribution is to be calculated under that section. And, the issue has not yet been spoken to by this circuit. Thus, because it has not yet been decided, and because so much of my argument depends on the matter, I pause here to offer some explanation for my view of the § 10(b) right.
The place to start is, of course, with Mu-sick, Peeler itself. Though the Court there declined to discuss in any detail the nature of the contribution right it found in § 10(b), it did indicate that it was willing to recognize a contribution right under that section in large measure because such a right is explicitly mentioned §§ 9 and 18 of Securities Exchange Act and these provisions are highly analogous to § 10(b) in both structure and design. See 113 S.Ct. at 2091. To ascertain the nature of the § 10(b) contribution right, thus, it is sensible to turn to examine these kindred provisions.
Sections 9 and 18 provide a defendant the opportunity to “recover contribution as in cases of contract from any person who, if joined in the original suit, would have been liable to make the same payment.” 15 U.S.C. §§ 78i(e) and 78r(b). The language “as in eases of contract” has led some to surmise that the contribution right under these sections is, in essence, a pro rata right — allowing for the apportionment of damages in equal portions amongst all defendants regardless of their relative fault. Professor Ruder, for one, has argued that
[t]he phrase “as in cases of contract” ... aids in answering the question “According to what method is the contribution to be allowed?” Theoretically contribution could be required either on a pro-rata basis of on some basis involving a degree of fault analysis. Since the Securities Acts incorporate the contract standard for contribution between tort-feasors, the pro-rata method used in common law contract cases should apply.
D. Ruder, Multiple Defendants in Securities Law Fraud Cases: Aiding and Abetting, Conspiracy, In Pari Delicto, Indemnification, and Contribution, 120 U.Pa.L.Rev. 597, 650 (1972). Other advocates of the pro rata rule have pressed for its adoption less out of concern for congressional intent and more out of an interest in promoting efficiency: the pro rata rule is, they promise, the easiest contribution standard for courts to administer. See, e.g., W. Douglas & G. Bates, The Federal Securities Act of 1933, 43 Yale L.J. 171, 178-180 (1933).
If contribution rights were to be determined on a pro rata basis under §§ 9 and 18 (and, thus, one might suggest, under § 10(b)), one might well argue that a pro rata judgment credit — one that assures a litigating defendant he will never be held liable in any ultimate judgment for a settling defendant’s pro rata share of the damages— *938would be entirely fair to a settling defendant and permit the trial court’s approval of a partial settlement. There would be no basis for suggesting, as I have above, that a proportionate share approach would assure consistency with the § 10(b) contribution right.17
That said, like the only federal appellate court to have studied the issue in any depth, see Smith v. Mulvaney, 827 F.2d 558, 560-61 (9th Cir.1987), I do think contribution under § 10(b) is measured by relative fault principles.18 In response to Professor Ruder’s suggestion that the “as in cases of contract” language found in §§ 9 and 18 necessitates the use of a pro rata rule, I agree with Mulvaney, 827 F.2d at 561, that this reads far too much into an exceptionally murky phrase. To begin, “as in cases of contract” does not as a matter of plain language or obvious implication direct pro rata to be employed. There is, moreover, no legislative history suggesting Congress intended the phrase to endorse a pro rata rule or any other particular contribution scheme. See Mulvaney, 827 F.2d at 561. All extant historical evidence suggests, in fact, that the Act borrowed this phrase from the English Companies Act — and that the phrase was not directly aimed at assuring the application of a pro rata rule even in England, but only at avoiding peculiar restrictions on the contribution right in English common law at the time. Thus, if anything, the phrase appears to be now almost a vestigial remnant of another time and place, not some obvious endorsement of the pro rata rule.19 Even if the language could be properly taken as directing us to the contribution rule in common law contract actions, finally, it would not be a foregone conclusion that pro rata rule must be employed. Pro rata is not always applied in common law contract actions; it sometimes gives way to a relative fault analysis when defendants are responsible for unequal portions of the liability.20 As to the contention that the pro rata rule is preferable because of its ease in administration, Mulva-ney and many recent commentators in the field have suggested that the relative fault method is only marginally more difficult for a court to employ. See Mulvaney, 827 F.2d at 561.
In the end, I think the contribution right under § 10(b) is most plausibly viewed as a relative fault right largely because of the design and intention behind the contribution doctrine itself. Contribution was originally developed in courts of equity out of the concern that one defendant ought not to be unjustly benefitted at the expense of another. See id. at 561. See also Gould v. American-*939Hawaiian S.S. Co., 387 F.Supp. 163 (D.Del.1974).21 Thus, even as between wrongdoers, our courts do aim to provide some semblance of fairness. The relative fault rule gives full vent to this interest — allowing any defendant who has paid more than his equitable share to seek payment from those who have underpaid theirs. A pro rata contribution scheme, by contrast, allows one defendant to benefit unjustly at another’s expense, except in those rare eases where each defendant is responsible for an equal share of any damage done.
That the pro rata rule is far less fair between defendants (and, thus, less harmonious with the purposes underlying contribution) receives support from a study of the Court’s opinion in Reliable Transfer. There the Court overturned a century-old precedent under which joint tortfeasors in admiralty collision cases were liable for equal, or pro rata, shares of the damage and replaced it with a relative fault rule; in doing so the Court recognized that “[i]t is no longer apparent, if it ever was, that th[e] Solomonic division of damages serves to achieve even rough justice.” — U.S. at -, 113 S.Ct. at 405. The Court noted, too, that virtually every maritime country in the world had already discarded the pro rata rule as unfair and had adopted a relative fault approach in its stead, id. at -, 113 S.Ct. at 403-404, and pointed out that comparative negligence had long been the admiralty rule in personal injury actions, id. at -, -, 113 S.Ct. at 407, 409.
Others have also come to accept a relative fault approach as the fairer approach. While pro rata was once ascendant in state common law jurisprudence, relative fault has in recent years emerged as the dominant measure of contribution in large part because of a growing recognition of its superiority in promoting equitable results. See L. Kornhauser & R. Revesz, Settlements Under Joint and Several Liability, 68 N.Y.U.L.Rev. 427, 437 (1993). See also Uniform Comparative Fault Act § 4(a), 12 U.L.A. at 54; Restatement (Second) of Torts § 886A(2). Securities law commentators likewise overwhelmingly support the use of the relative fault rule in § 10(b) cases out of concerns for fairness. See, e.g^ L. Loss, ALI, Fed.Sec.Code 1418(f)(2) (Draft No. 2, 1976); M. Adamski, Contribution and Settlement in Multiparty Actions Under Rule 10b-5, 66 Iowa L.Rev. 553, 557 (1981); J. Fischer, Contribution in 10b-5 Actions, 33 Bus.Law. 1821, 1829 (1978); J. Freund & H. Hacker, Cutting Up the Humble Pie: A Practical Approach to Apportioning Litigation Risks Among Underwriters, 48 St. John’s L.Rev. 461, 473 (1974).22
* * *
While joining Parts I and III of the court’s opinion, for the reasons expressed above I concur only in the result reached in Part II.
. As the Court noted in McDermott, see - U.S. at - n. 8, 114 S.Ct. at 1465 n. 8, these three options correspond to three model Acts drafted at different points in time by the National Conference of Commissioners on Uniform State Laws. ALI Option 1 corresponds to the Uniform Contribution Among Tortfeasors Act, drafted in 1939, 12 U.L.A. 57-59 (1975); ALI Option 2 represents the Revised Uniform Contribution Among Tortfeasors Act of 1955, id. at 63-107; ALI Option 3 tracks the Uniform Comparative Fault Act adopted in 1977, 12 U.L.A. 45-61 (1993 Supp.).
. In case one were inclined to think low settlements few and far between and, thus, Option 2's inconsistency with Reliable Transfer only a rare and minor problem, McDermott has noted that "deviations from the equitable apportionment of damages will be common” under ALI Option 2 for "settlements seldom reflect an entirely accurate prediction of the outcome of a trial." - U.S. at -, 114 S.Ct. at 1467. The Court commented, too, that settlements will fail to represent a settling defendant's full share of the liability because a discount reflecting the uncertainties of trial will usually be built into the settlement figure. — and because a plaintiff may be willing to accept a low amount in settlement with one defendant in order to develop a “war chest” to finance his litigation against the remaining defendants. Id.
. It is true that the Revised Uniform Contribution Among Tortfeasors Act (which ALI Option 2 describes) envisions barring nonsettlors from *932seeking contribution by statutory means, whereas in the case before us nonsettlors are barred from seeking contribution by an injunction. This distinction, however, makes no difference. Nothing in McDermott turned upon the means by which nonsettlors were barred from seeking contribution; likewise we need today say nothing about the particular means by which appellants are prevented from exercising their contribution rights. The critical point both in McDermott and the instant case is that the provision of a pro tanto credit alone without any additional right to contribution (however the negation of that right is achieved) is inimical to nonsettlors' rights.
. It is worth noting that, in the process of making its point about the inadequacy of good-faith hearings, the Court specifically made negative reference to the trial court’s endorsement of such hearings in this case. See McDermott, — U.S. at - n. 18, 114 S.Ct. at 1468 n. 18.
. The court avers at one point that analogy to McDermott will not suffice to justify overruling the trial court's decision since, in contrast with this case, "no bar order was at issue" and no settlement agreement was involved in McDer-mott. See Majority Op. at 923. As discussed above, McDermott specifically considered and rejected ALI Option 2 which incorporates a pro tanto credit while operating to bar nonsettlors from receiving anything further from settling defendants. Furthermore, as also suggested above, there is absolutely no basis in McDermott for *933believing that a trial court could any more approve a settlement agreement incorporating ALI Option 2 or some variant on it than it could impose such an option itself when left room to do so by the parties' settlement arrangements.
. The Court's reasons for finding a pro tanto credit without bar order unattractive even when coupled with the provision of indemnity by plaintiff are obviously in no way dependent upon the fact that McDermott arose in admiralty. Indeed, it is worth noting that the National Conference of Commissioners on Uniform State Laws, the very body that suggested Option 1 in 1939 for use in tort actions generally, came to reject it in its 1955 Revised Uniform Contribution Among Tort-feasors Act, for precisely the same reasons the Court rejected it in McDermott — viz. due to its tendency to discourage settlement and burden the courts with additional indemnity-oriented litigation. See 12 U.L.A. at 98-99 (commissioners’ comment).
. "Consider an example in which ... the two defendants are each equally at fault, and the plaintiff's damages are $100. If the plaintiff settles with one defendant for $20, under the [proportionate] share set-off rule it can recover only $50 from the other defendant, thus suffering a shortfall of $30. What happens, however, if the settlement is for $70?" Recovery is allowed under a pure proportionate share approach. L. Kornhauser & R. Revesz, Settlements Under Joint and Several Liability, 68 N.Y.U.L.Rev. 427, 439 (1993).
. "Settling defendants pay an amount to which they voluntarily agree.... Nonsettling defendants never pay more than they would if all parties had gone to trial. This comports with the equitable purpose of contribution [determined by reference to relative fault]." Franklin, 884 F.2d at 1231.
. Though the use of a proportionate share judgment credit clearly envisions having the juiy determine an absent settling defendant's share of the fault, the jury may not, of course, ascribe fault to an absent nonsettling defendant. See, e.g., Uniform Comparative Fault Act (12 U.L.A.) § 2 (1993 Supp.). Present nonsettling defendants will, however, have every incentive to join any potentially liable absent nonsettling defendant that plaintiff has not already named in his action. See id.
. Contrary to the court's charge, see Majority Op. at 927, I hardly suggest here that a federal court can accept jurisdiction, willy-nilly, over any question put to it by two parties seeking an answer. Rather, my point is that I see requiring the filing of a contribution action as a mere formalism that Article III surely would not require. As will become evident in a moment, this is far from a novel suggestion.
. “Obviously, there will be a certain amount of 'finger-pointing' at the 'empty chair.’ ... However, settling defendants will be protected ... and the financial motives of both plaintiffs and nonsettling defendants to vigorously press their arguments at trial will be unchanged." Franklin, 884 F.2d at 1231.
. If one nonsettling defendant's relative share is uncollectible, the joint and several liability rule might well operate to require a reallocation of fault. See, e.g., Uniform Comparative Fault Act, 12 U.L.A. at 50-51 (1993 Supp.).
. Deciding the settling defendants’ share of the liability in the primary action'may result in plaintiff "underrecovering” or "overrecovering" his damages (placing on him as it does both the risk of a poor settlement and the benefit of an exceptionally good one). But, the Supreme Court in McDermott made plain that this in no way trenches upon joint and several liability princi-pies. See discussion infra, at n. 16. It is, after all, plaintiff's decision whether to settle or not to settle.
.By way of example, one of the cases the court relies upon, Borden, Inc. v. Florida East Coast Ry. Co., 772 F.2d 750 (11th Cir.1985), involved no settlement and held that the trial judge erred in apportioning damages according to relative fault no doubt because Florida state common law (which the court was interpreting) did not permit contribution according to relative fault, but only according to pro rata rules. See Lincenberg v. Issen, 318 So.2d 386, 393-94 (1975). Furthermore, Borden took no account of the Uniform Comparative Fault Act which plainly endorsed the idea that joint and several liability is not compromised by the entry of a proportionate share credit in tort. Another case the court relies upon, Ebanks v. Great Lakes Dredge & Dock Co., 688 F.2d 716 (11th Cir.1982), is an admiral*936ty case whose authority may be questionable after McDermott.
. The equitable jurisdiction of a federal court to enter an injunction preventing relitigation of its decrees in the same court is beyond cavil. See, e.g., Root v. Woolworth, 150 U.S. 401, 14 S.Ct. 136, 37 L.Ed. 1123 (1893). See generally Dugas v. American Surety Co., 300 U.S. 414, 57 S.Ct. 515, 81 L.Ed. 720 (1937); Local Loan Co. v. Hunt, 292 U.S. 234, 54 S.Ct. 695, 78 L.Ed. 1230 (1934). To the extent this same principle finds expression in the All Writs Act, that legislation would also support the issuance of a bar order.
. To this point, I have discussed only the court's rather unconventional objections to the application of the proportionate share credit in § 10(b) cases — objections, incidentally, that neither the parties nor the trial court pressed. There do exist, however, a pair of more conventional arguments against applying the credit, ones raised by appellees and the trial court in the course of arguing that the imperfections of the proportionate share rule suggest the wisdom of approving a settlement agreement with a pro tanto credit. The two arguments are essentially flip-sides of the same coin and neither is persuasive.
First, one might argue that the proportionate share credit can operate to prevent a plaintiff from receiving a full recovery for his loss. And, indeed this can happen when a plaintiff's agreement with the settling defendants proves to be less than the settlors’ jury-determined equitable share of the fault and plaintiff is left unable to recoup the difference from anyone (under a pro tanto regime, the plaintiff would recover the difference from the nonsettling defendant). One might suggest, thus, that a proportionate share credit should be rejected on the grounds that denying plaintiff a full recovery improperly elevates Reliable Transfer and the § 10(b) relative fault contribution right’s goal of an equitable distribution of fault between defendants over a joint and several liability regime's most fundamental objective of assuring plaintiff full recovery for his injury. The Supreme Court in McDer-mott, however, explained the fallacy inhering in this argument when it wrote that "[j]oint and several liability applies when there has been a judgment against multiple defendants. It can result in one defendant's paying more than its apportioned share of liability when the plaintiff’s recovery from other defendants is limited by factors beyond the plaintiff's control, such as a defendant's insolvency. When the limitations on the plaintiff's recovery arise from outside forces, joint and several liability makes the other defendants, rather than an innocent plaintiff, responsible for the shortfall.... [Tjhe proportionate share rule announced in this opinion applies [only] when there has been a settlement. In such cases, the plaintiff's recovery against the settling defendant has been limited not by outside forces, but by its own agreement to settle. There is no reason to allocate any shortfall to the other defendants, who were not parties to the settlement.” - U.S. at ---, 114 S.Ct. at 1471-72 (footnotes omitted).
Secondly and contrarily, one might object to a proportionate share judgment credit on the grounds that it occasionally leads to a violation of the “one satisfaction rule” by affording the plaintiff "overcompensation” (i.e. more than one *937hundred percent of the jury-determined damages). With the proportionate share rule, if plaintiff's agreement with the settling defendants is for an amount greater than the settlors' relative share of the damages, plaintiff is entitled to keep the full amount of the settlement plus all due him from the nonsettlor, thus resulting in a recovery of more than one hundred percent of the jury award. Again, however, the Court in McDermott saw no merit to this argument, see - U.S. at ---, 114 S.Ct. at 1470-71, and its reasoning applies as persuasively outside the admiralty arena as it does within it. See Franklin, 884 F.2d at 1231.
. If § 10(b) were construed to extend a pro rata contribution right, it is obvious, however, that a pro tanto credit with a bar order would still likely create inconsistencies with that right.
. A few decisions predating Mulvaney did apparently consider the contribution right in securities fraud cases to be a pro rata right. See, e.g., Herzfeld v. Laventhol, Krekstein, Horwath & Horwath, 540 F.2d 27, 39 (2d Cir.1976); Globus, Inc. v. Law Research Serv., Inc., 318 F.Supp. 955, 958 (S.D.N.Y.1970), aff'd., 442 F.2d 1346 (2d Cir.1971), cert. denied sub nom. Law Research Serv., Inc. v. Blair & Co., 404 U.S. 941, 92 S.Ct. 286, 30 L.Ed.2d 254 (1971). There are, however, a few reasons why these decisions do not carry much precedential weight. [1] None paused to consider the nature of the contribution right involved, but only rotely applied the pro rata rule. [2] In some of these decisions the joint tortfeasors were held equally culpable. Thus, the courts in these cases never had any serious need to consider whether the pro rata rule was correct or not since the ultimate distribution of liability would not have changed even had the relative fault rule been applied. [3] Since Mulvaney was decided, I am unable to find a published case in which a federal court has chosen to apply a pro rata rule under §§ 9, 18 or 10(b). Even the Second Circuit, where the pro rata rule had once been applied, appears to have altered its position in recent years. In Singer v. Olympia Brewing Co., 878 F.2d 596 (2d Cir.1989), that court did not adopt a pro rata judgment credit for nonsettlors in § 10(b) actions as would have been the sensible course if it thought the contribution right a pro rata right; the court seemed to acknowledge, too, at least the possibility that "a plaintiff may be entitled to more damages from one defendant than from another....” Id. at 600.
. See J. Fischer, Contribution in 10b-5 Actions, 33 Bus.Law. 1821, 1829 (1978). After a detailed discussion of the English origins of the "as in cases of contract” language, Fischer concludes that Professor Ruder's thesis that the phrase requires application of a pro rata rule is in error and that "it appears that the phrase ... cannot be read too broadly. Indeed, its textual ambiguity and impreciseness makes its true application one of conjecture.” 33 Bus.Law. at 1830.
. See Gould v. American-Hawaiian S.S. Co., 387 F.Supp. 163, 171 (D.Del.1974).
. "Traditionally, equity has been the hallmark of contribution. The doctrine originated in courts of equity, and its rationale has not been altered by adoption in the common law. Jones v. Schramm, 436 F.2d 899 (D.C.Cir.1970); George's Radio, Inc. v. Capital Transit Co., 126 F.2d 219 (D.C.Cir.1942); 18 Am.Jur.2d Contribution §§ 4 and 5. The governing principle of contribution throughout has been that one of two or more joint wrongdoers should not be required to pay more than his share of the burden, or to put it another way, that no obligor should be unjustly benefited at the expense of another.” Id. at 170.
. It is worth noting, as well, that contribution is determined by reference to relative fault, not a strict pro rata rule, under other federal statutory schemes. See, e.g., Environmental Transp. Sys. v. Ensco, Inc., 969 F.2d 503 (7th Cir.1992) (holding that Congress intended courts to reference relative fault when assessing damages under CERC-LA).