Three cases come to us presenting a common and significant question of procedural law: Should a judgment in favor of a passenger in an action against the operators of two colliding vehicles give rise to an estoppel, which would bar a subsequent action by one of the drivers against the other for his own personal injuries or property damage? To put the issue in terms of legal precedent, should Glaser v. Huette (232 App. Div. 119, affd. 256 N. Y. 686) be overruled? We conclude -that the need for a “ prompt and nonrepetitious judicial system ” and our recent decisions have so undermined the rationale of the Glaser holding that it is no longer viable as a precedent. (Cummings v. Dresher, 18 N Y 2d 105, 107.) With its disappearance, New York law will have arrived at a modern and stable statement of the law of res judicata. Glaser has been replaced by the sound principle that, where it can be fairly said that a party has had a full opportunity to litigate a particular issue, he cannot reasonably demand a second one (Liberty Mut. Ins. Co. v. Colon & Co., 260 N. Y. 305; Good Health Dairy Prods. Corp. v. Emery, 275 N. Y. 14; United Mut. Fire Ins. Co. v. Saeli, 272 App. Div. 951, affd. 297 N. Y. 611; Cohen v. Dana, 275 App. Div. 723, affd. 300 N. Y. 608; Israel v. Wood Dolson Co., 1 N Y 2d 116; Hinchey v. Sellers, 7 N Y 2d 287; Cummings v. Dresher, supra; B. R. De Witt, Inc. v. Hall, 19 NY 2d 141).
Traditionally, collateral estoppel was primarily limited to cases involving indemnity, employment or agency and was permitted in those cases only to avoid the absurd result of having the indemnitor exonerated, while the indemnitee was held liable.
*70We have, however, already discarded, as irrelevant to a proper consideration of the issues in this area, the fact that there may or may not have been any significant jural relationship between the party seeking to invoke the doctrine and the prior victor (Cummings v. Dresher, supra; Israel v. Wood Dolson Co., supra; B.R.De Witt, Inc. v. Hall, supra). Similarly, in our most recent decision in De Witt we stated that the doctrine of mutuality “ is a dead letter ” (19 1ST Y 2d, p. 147) and we removed the limitation that a prior judgment may only be used defensively.
Nevertheless, absent this most recent decision, Glaser would still have to be overturned (see concurring opn. of Halpern, J., in Ordway v. White, 14 A D 2d 498). For in a long series of cases, starting with Eissing Chem. Co. v. People’s Nat. Bank of Brooklyn (205 App. Div. 89, affd. 237 N. Y. 532); Good Health Dairy Prods. Corp. v. Emery (supra); Cohen v. Dana (supra) and, especially, Israel v. Wood Dolson Co. (supra), we have permitted a defensive use of collateral estoppel so long as there was an identity of issues, and the party against whom the estoppel was being asserted had a full opportunity to contest the issue. “ [T]he fact that a party has not had his day in court on an issue as against a particular litigant is not decisive in determining whether the defense of res judicata is applicable ” (Israel v. Wood Dolson Co., 1 N Y 2d, supra, p. 119; emphasis in the original).
In Good Health Dairy Prods. Corp. v. Emery (supra) we said (275 N. Y., p. 18): ‘ ‘ Behind the phrase res judicata lies a rule of reason and practical necessity. One who has had his day in court should not be permitted to litigate the question anew. Although normally it is necessary that mutuality of estoppel exist, an exception is at times made where the party against whom the plea is raised was a party to the prior action and ‘ had full opportunity to litigate the issue of its responsibility.’ (See Liberty Mutual Ins. Co. v. Colon & Co., 260 N. Y. 305, 312.) Under such circumstances the judgment is held to be conclusive upon those who were parties to the action in which the1 judgment was rendered. Where a full opportunity has been afforded to a party to the prior action and he has failed to prove his freedom from liability, or to establish liability or culpability on the part of another, there is no reason for permitting him to retry these issues. ’ ’
*71Although we have not previously said so, it is now evident that New York has adopted the full and fair opportunity test ■in applying the doctrine of collateral estoppel. (Zdanok v. Glidden Co., 327 F. 2d 944, 956 [2d Cir., Friendly, J.], cert. den. 377 U. S. 934; Graves v. Associated Transp., 344 F. 2d 894, 900 [4th Cir.]; Teitelbaum Furs v. Dominion Ins. Co., 58 Cal. 2d 601 [following through on Chief Justice (then Justice) Traynor’s seminal decision in Bernhard v. Bank of America, 19 Cal. 2d 807]; Ordway v. White, 14 A D 2d 498, 500-501 [Halpern, J., concurring], supra; see, also, Currie, Mutuality of Collateral Estoppel: Limits of the Bernhard Doctrine, 9 Stan. L. Rev. 281; Currie, Civil Procedure: The Tempest Brews, 53 Calif. L. Rev. 25, 31; Polasky, Collateral Estoppel —■ Effects of Prior Litigation, 39 Iowa L. Rev. 217, 250.) New York Law has now reached the point where there are but two necessary requirements for the invocation of the doctrine of collateral estoppel. There must be an identity of issue which has necessarily been decided in the prior action and is decisive of the present action, and, second, there must have been a full and fair opportunity to contest the decision now said to be controlling.
It is in light of these new views that we proceed to re-examine the Glaser decision. The plaintiffs and claimant* here argue — and, indeed, Glaser rested on the proposition — that the parties in the second action were not adversaries in the first action. As noted above, our decisions in recent years (see cases cited supra, pp. 69-70) have made this distinction an insignificant if not irrelevant one. Even if it were still an acceptable position, it would have no validity under the factual pattern present in these cases.
In the years since Glaser was decided, whatever merit there might have been to the argument — and there was little —■ has been completely dissipated. Section 211-a of the Civil Practice Act (now CPLR 1401) which became law shortly before the Glaser decision provided for contribution between joint tortfeasors. As a result, while each defendant driver seeks complete exoneration from liability to the passenger, he also desires to hold in the other defendant. Moreover, in preparing for the trial, each defendant now has full discovery against his *72codefendant whether or not there is a claim between them (CPLR 3101; Lombardo v. Pecora, 23 A D 2d 460). At the trial the codefendants have the same rights of cross-examination with respect to each other’s witnesses as they have with respect to the passenger’s. In every respect they are antagonists, even to the éxtent that evidence introduced by one codefendant may be relied upon by the other. (See, e.g., Civ. Prac. Act, § 212, subd. 3.) In fact, it may rightly be said that in many cases the battle between the codefendants is more strenuous than is their attack against their supposedly main adversary, the plaintiff. The argument that it is unfair to apply the earlier judgment in the subsequent action between the codefendants, on the ground that the parties were not true adversaries, is wholly without merit.
A decision whether or not the plaintiff drivers had a full and fair opportunity to establish their nonnegligence in the prior action requires an exploration of the various elements which make up the realities of litigation. A comprehensive list of the various factors which should enter into a determination whether a party has had his day in court would include such considerations as the size of the claim, the forum of the prior litigation, the use of initiative, the extent of the litigation, the competence and experience of counsel, the availability of new evidence, indications of a compromise verdict, differences in the applicable law and foreseeability of future litigation.
The present plaintiffs were full participants in the earlier cases. All of these actions involved substantial sums. They did not involve claims for property damage amounting to but a few hundred dollars so that there would be no assurance that there was a vigorous fight on the issue of liability. Moreover, each of the plaintiffs had a full opportunity to tell his story at the first trial in order to remove himself from liability or to cast it elsewhere. ■ None claims a lack of adequate representation, and none has shown any prejudice because of the forum of the earlier action. Likewise, no prejudice is claimed because in the earlier action the passenger had any tactical advantages. Finally, there is no assertion of any significant new evidence, which would almost certainly change the earlier result. It is, therefore, utterly fair to apply the doctrine of collateral estoppel *73here and to the typical case where driver sues driver following a verdict by a passenger against both.
When we turn from the specifics of these cases, we find other more general arguments which have been advanced against permitting the use of collateral estoppel in cases such as the type involved here. The contentions raise problems which are not unreal, but to which there is a simple answer. It is said, for example, that -the first verdict in favor of the passenger may have been influenced by sympathy since the jury almost certainly knows that the defendants had at least some insurance. Yet where the driver interposes a counterclaim or there is a consolidation of actions, this danger is greatly mitigated and wholly eliminated if the driver’s injuries are more severe than the passenger’s.
In this connection, claimant Wever points out that he was not in a position to interpose a counterclaim as he was required to seek relief in arbitration pursuant to the endorsement on his' insurance policy because his accident involved an uninsured automobile. This claim is not without some merit, but Wever has offered nothing which would indicate that he was actually prejudiced by his inability to join his claim to the prior action. He has not shown that the first verdict was influenced by excessive sympathy for the injured passenger or was a compromise verdict.
No one would contend that the doctrine of collateral estoppel should be applied rigidly. In De Witt (19 N Y 2d 141, supra), however, we indicated that the burden rests on the defendant to show that collateral estoppel should not be applied because he did not have a full and fair opportunity (supra, p. 148), just as the burden of showing that the issue was identical and necessarily decided rests upon the moving party. This apportionment of the burdens is both fair and necessary. Otherwise much of the value of collateral estoppel will be lost.
It is also urged against the approach we have here adopted that attorneys who represent the drivers in the earlier actions are representatives of liability insurance carriers and are not-counsel of the drivers’ choice. As noted before, it is not difficult to conceive of a situation where a carrier may have devoted almost all of its efforts, in defending the action by the passenger, *74to obtaining a partner to share in paying a recovery by the passenger. Other possible sources of inadequate representation may arise if in the first action the passenger had suffered relatively minor injuries or the claim had involved a relatively small sum of money. The carrier may not then have made a substantial investment of time and effort in defending the action. But here again, these dangers can be obviated if the driver interposes a counterclaim in the first action or consolidates his action with the passenger’s. At that point he will have an attorney of his own choosing who will be protecting only his interests in the litigation.
It can be seen that the arguments against applying collateral estoppel may be effectively answered while, on the other hand, the arguments for applying it are powerful indeed. It will reduce the number of inconsistent results which are always a blemish on a judicial system. The dissent concedes the logical defects of the Glaser rule, but forcefully argues that it had the justifiable result of permitting recoveries by at least one of the two drivers in collision cases. This, however, is an argument for a change in the principle of fault liability or for establishing a system of comparative negligence which permits for cross recoveries. As long as we abide by the present framework, it is difficult to tolerate a condition where, on relatively the same set of facts, one fact-finder, be it a court or jury, may hold the driver liable, while the other exonerates him. Finally, in this day of great delays in accident litigation, a single trial of all claims growing out of the same accident will constitute a great saving of judicial manpower and time, without any true unfairness to the parties. (See Semmel, Collateral Estoppel, Mutuality and Joinder of Parties, 68 Col. L. Rev. 1457.)
Each of the plaintiffs also argues that there has been no showing of identity of issues here and that the jury in the actions brought by the passengers may have held the plaintiffs here liable on account of acts which might be considered negligent vis-á-vis the passengers, but not insofar as the other drivers are concerned. A simple response to this contention is that, in the usual collision accident, this would be most unlikely and the courts may presume such is not the case. Moreover, a closer examination of two of the cases here shows this not to be the case at all. Take for example, the Bartolone case, where it is *75argued that the jury might have held Miss Bartolone liable because she failed to warn her passengers of the impending danger. The first Bartolone case (Di Salvo v. Bartolone) reached this court in the form of a motion for leave to appeal to this court from a unanimous affirmance against her by the Appellate Division (27 AD 2d 932). She (or the carrier to be more precise) argued that there was no substantial evidence to hold her liable. Leave was denied (20 N Y 2d 642). An examination of the papers submitted on that motion convinces us that there is no question that it was Miss Bartolone’s driving upon which her liability was based. She must necessarily be guilty of contributory negligence here.
Likewise, Justice Eager’s opinion in the Schwarts case points out that the liability of the driver Schwartz in the action by the passengers rested upon a finding of negligence in the operation of the vehicle he was driving (Schwarts v. Public Administrator, 30 A D 2d 193, 197). The argument of the dissent in the Schwarts case that the liability might have some other basis is purely speculative. The prior verdict rested upon a finding of negligence in the operation of the vehicle, and that issue was necessarily determined in the prior action. Consequently, it must necessarily preclude the bringing of the present action.
The record of the first trial in the Wever case is not before us. Still it is fair to assume that, in the typical automobile negligence case, a finding of negligence on the part of the driver rests or arises out of his negligent driving. In our view, identity of issues was shown in all three cases.
One final point should be made. Plaintiff Bartolone urges that, if we do effect a change in the law, we should give our decision prospective effect only. Bartolone’s counsel claims that, in reliance on pre-existing law, he made a conscious choice not to interpose a counterclaim in the first action. There is always a problem for lawyers and their clients when a rule of procedural law is changed. But there is always a certain amount of uncertainty about the law, and, especially, there can be no strong claim of reliance here since the law of New York has been in evolution for many years. It is, therefore, not unfair to apply this new rule to these cases, especially where the plaintiffs have made no showing of prejudice.
*76Accordingly, the order of the Appellate Division in Schwartz v. Public Administrator of Bronx should be affirmed; the order of the Appellate Division in Bartolone v. Niagara Car & Truck Rentals should be reversed,' the certified question answered in the negative, and defendants’ motion for leave to serve a supplemental and amended answer should be granted; and the order of the Appellate Division in Matter of Wever v. MVAIC should be reversed and appellant’s motion for a stay of arbitration granted, with costs to the prevailing party in each case.
Hereafter the .term “ plaintiffs ” will be used to refer to the plaintiffs in the Schwartz and Bartolomé cases, and the claimant in Matter of Wever.