Edward Perkowski and other officers of Carpenters Local 1243 named as defendants in this lawsuit appeal directly from the district court's order granting a motion to disqualify defendants' counsel. Several members of Carpenters Local 1243 brought suit on behalf of the Local against the union officers for breach of fiduciary duty. 29 U.S.C. Sec. 501(a) (1976). The plaintiffs moved to disqualify defendants' counsel on the ground of a potential conflict of interest. The district court granted the motion. We hold that an order disqualifying counsel in a civil case is immediately appealable under 28 U.S.C. Sec. 1291 (1976), and that the district court did not abuse its discretion in disqualifying defendants' counsel. In this opinion for publication, we discuss the single precedential issue of whether the disqualification order is immediately appealable under 28 U.S.C. Sec. 1291. The remaining issues are discussed in a simultaneously filed Memorandum that will not be published.

Generally, interlocutory orders are not appealable until after final judgment and in the absence of a "final order" from the district court we have no jurisdiction. 28 U.S.C. Sec. 1291. However, in Cohen v. Beneficial Loan Corp., 337 U.S. 541, 69 S. Ct. 1221, 93 L. Ed. 1528 (1949) the Supreme Court developed a collateral order exception to the final judgment rule of 28 U.S.C. Sec. 1291.1 In Coopers & Lybrand v. Livesay, 437 U.S. 463, 98 S. Ct. 2454, 57 L. Ed. 2d 351 (1978), the Supreme Court summarized the Cohen three part test for appealability under the collateral order exception as follows:

[T]he order must conclusively determine the disputed question, resolve an important issue completely separate from the merits of the action, and be effectively unreviewable on appeal from a final judgment.

Id. at 468, 98 S.Ct. at 2457 (citations and footnote omitted).

There is no doubt that the first two parts of the Cohen test are satisfied here. First, the order disqualifying counsel conclusively determines that question. After disqualification, counsel may not proceed with the case. The litigation continues on with new counsel. The order is not the kind which the court reconsiders from time to time during the course of the proceedings. Second, an order granting a disqualification motion serves to "resolve an important issue completely separate from the merits of the action." Coopers & Lybrand, 437 U.S. at 468, 98 S. Ct. at 2457. The issue whether counsel should be disqualified is completely separate from the merits of the suit. Determinations regarding the disqualification of counsel will ordinarily not require courts to become "enmeshed in the factual and legal issues comprising the plaintiff's cause of action," Coopers & Lybrand, 437 U.S. at 469, 98 S.Ct. at 2458, (quoting Mercantile National Bank v. Langdeau, 371 U.S. 555, 558, 83 S. Ct. 520, 522, 9 L. Ed. 2d 523 (1963)). The question is "too independent of the cause itself to require that appellate consideration be deferred until the whole case is adjudicated." Cohen, 337 U.S. at 546, 69 S.Ct. at 1225.

The more difficult question is presented by the third part of the test: whether an order disqualifying counsel can be effectively reviewed after final judgment.

In Firestone Tire & Rubber Co. v. Risjord, 449 U.S. 368, 376, 101 S. Ct. 669, 674, 66 L. Ed. 2d 571 (1981) the Supreme Court held that an order denying a motion to disqualify counsel in a civil case did not satisfy the Cohen test. Its rationale was that such an order may be effectively reviewed after final judgment.2 However, in Firestone, the Court specifically reserved the question whether an order granting a motion to disqualify counsel in a civil case is appealable under Cohen. 449 U.S. at 372 n. 8, 101 S. Ct. at 672 n. 8.

After fully considering Firestone, the Fifth Circuit determined that an order granting disqualification of counsel in a civil case is an appealable order under Cohen. Duncan v. Merrill, Lynch, Pierce, Fenner, & Smith, 646 F.2d 1020 (5th Cir.), cert. denied, 454 U.S. 895, 102 S. Ct. 394, 70 L. Ed. 2d 211 (1981). The Fifth Circuit reasoned that the order was "effectively unreviewable" because:

[T]he appellate court's power to vacate the judgment appealed from and order a new trial would seldom provide any solace to the party erroneously deprived of his chosen counsel. As the Second Circuit recently observed, "[i]f the order [granting disqualification] is erroneous, correcting it by an appeal at the end of the case might well require a party to show that he lost the case because he was improperly forced to change counsel. This would appear to be an almost impossible burden." (Citation omitted). Furthermore, if the party who suffers the erroneous grant of a disqualification motion prevails in the action, the disqualified counsel is not likely to find any remedy for the damage to his reputation or to his pocketbook. Thus, ... the harm caused by postponing review of an order granting disqualification of counsel would in most instances be irreparable.

646 F.2d at 1027.

We find the reasoning in Duncan to be highly persuasive. However, before determining whether to join in the Fifth Circuit's view, we must carefully consider what we have said in two recent cases in which we addressed closely related issues.

In United States v. Greger, 657 F.2d 1109 (9th Cir.1981), we considered the appealability of an order disqualifying defense counsel in a criminal case. We found the order to be effectively reviewable after final judgment and for that reason held that it did not fall within the collateral order exception. Id. at 1112. Our decision relied heavily on the special policies that militate against piecemeal appeals in criminal cases. Id. The Greger holding was specifically limited to criminal cases and did not address the issue presented here. Id. at 1113.

Shortly after Greger was decided, we again considered the appealability of an order disqualifying counsel, but this time in a civil case. In In re Coordinated Pretrial Proceedings in Petroleum Products Antitrust Litigation, 658 F.2d 1355 (9th Cir.1981), cert. denied, --- U.S. ----, 102 S. Ct. 1615, 71 L. Ed. 2d 850 (1982), we held that an order disqualifying a non-party's counsel in a civil case was appealable under the collateral order exception. We distinguished the Greger holding by reasoning that:

[D]istinct reasons for a different result apply in this civil case. The special policies which counsel against piecemeal appeals in criminal cases have no counterpart here. Cf. Greger, supra, at 1112. An appeal of a disqualification order in a civil case does not necessarily disrupt the litigation because activity must be postponed in any case while new counsel is obtained. Furthermore, in a civil case an order disqualifying attorneys for nonparties is effectively unreviewable by appeal from a final judgment. Finally, in a civil case, the presumption of prejudice which would mandate a new trial if error were established under the criminal law does not exist. Cf. Greger, supra, at 1113; Slappy v. Morris, 649 F.2d 718 at 723 n. 4 (9th Cir.1981).

Id. at 1358.

The reasoning of Petroleum Products is equally applicable here. In a civil case it is at least as important for a party to be able to obtain immediate review of an order disqualifying his counsel as it is for a non-party.3

We hold that an order disqualifying a party's counsel in a civil case is "effectively unreviewable" after final judgment. The burden a party would be required to meet in order to obtain a reversal could not, practicably, be met after final judgment in most civil cases. The fact that counsel was improperly disqualified would not, in itself, be sufficient to warrant a new trial since, as we said in Petroleum Products, there is no presumption of prejudice which attaches to such disqualification.

There are valid reasons why, if an appeal were delayed until after final judgment, the judgment should not be reversed in the absence of a showing of prejudice. A per se rule requiring reversal whenever the disqualification of counsel in a civil case is held to be erroneous would be unfair to litigants and contrary to the interests of orderly judicial administration. The adoption of such a rule would allow the party whose counsel was disqualified erroneously a free run at a verdict. He would have everything to gain and nothing to lose at the first trial. He could await its outcome and, if he did not like the result, could belatedly assert his right to a trial with the counsel he originally chose. On the other hand, if the results were acceptable, he could stand pat. His opponent, however, would be in a no-win situation. If the opponent obtained a favorable result at the first trial, the party whose counsel was erroneously disqualified would have the verdict overturned. If the opponent lost the first time, the party whose counsel was erroneously disqualified would forego his right to automatic reversal and accept his victory.

While a per se rule of reversal might conceivably prove a deterrent to unjustified challenges of parties' counsel, we do not believe such a rule is either necessary or desirable. Most unjustified challenges are in any event disposed of expeditiously by the district court. Since the district court's order rejecting a challenge is non-appealable, Firestone, 449 U.S. at 376, 101 S.Ct. at 674, unjustified challenges do not present any serious problem insofar as the orderly disposition of cases is concerned; thus there is no need for a rule which would have a deterrent effect.

Although a showing of prejudice would be necessary to support a post-judgment reversal based on an erroneous disqualification of counsel, a post-judgment inquiry into prejudice would impose an unrealistic burden on the courts. From a practical standpoint, we believe the disqualification issue would be "effectively unreviewable" on appeal. The court would not be considering whether the new counsel was incompetent (cf. Cooper v. Fitzharris, 586 F.2d 1325 (9th Cir.1978) (en banc) cert. denied, 440 U.S. 974, 99 S. Ct. 1542, 59 L. Ed. 2d 793 (1979)) or negligent, but whether and how the original counsel might have handled the matter differently, i.e., taken more or less discovery, called more or fewer witnesses, asked or refrained from asking particular questions, introduced or refrained from introducing particular evidence, raised different objections, adopted different trial strategies, made other or more persuasive arguments to the jury, and what the effect on the verdict might have been. The argument in most cases would quickly degenerate into speculation and hindsight, and a proper decision would be extremely difficult, if not impossible, to arrive at.

Most able lawyers, given a chance to review the results of their initial efforts, could discover ways in which they might have improved on their performance. Armed with knowledge of the judge's and jury's reactions to his handling of the various difficult questions presented by the trial of a lawsuit, any respectable counsel could easily develop better techniques or approaches for dealing with them than he could without that knowledge. More important, almost any attorney, knowing the outcome of all the twists and turns of a hotly litigated matter, could tell anyone interested how he would have tried the case and won it had he been the trial lawyer rather than someone less talented or creative. Yet, if appeals are delayed until after trial, courts would be required to deal with precisely that sort of post hoc rationalization in every case. The tales and explanations of how original counsel could have won the case, had he been permitted to try it, would be legion and most would be at least superficially plausible. From a practical standpoint, courts would have no way of determining whether the original counsel would have acted as he says, after the fact, that he would. Even assuming the best of faith on the part of original counsel, it would be difficult for him to put the events which occurred at the trial out of mind when, following a verdict his client wishes to reject, he is compelled to tell the court how he would have handled the matter differently.

There is another problem which would make it extremely difficult in civil cases to determine whether the handling of a case by one counsel rather than another caused any prejudice to a party, or, to put it more directly, any difference in the result. Verdicts in civil cases frequently do not represent absolute victories for one party or the other. Where only damages are sought, it is rare that the verdict is returned in the full amount requested in the complaint. Similarly, where the relief sought is non-monetary or where the complaint seeks both monetary and other relief, a party will frequently obtain some, but not all, of the requested relief. Thus, in order to determine whether a party suffered prejudice as a result of a district judge's refusal to permit him to use the attorney he originally retained, a court would have to determine whether, but for that ruling, the relief awarded might have been greater or less. In many cases it would not be possible to make the necessary judgment based on any objective standards. A trial lawyer's ability to persuade a jury to return the largest possible verdict may have been one of the key factors in the client's original decision to retain him. A court would have a most difficult time attempting to determine by any objective test whether that lawyer's particular skills would have caused a jury to award greater damages than it returned after listening to a different attorney whom the client did not originally wish to retain.

The question of what constitutes prejudice would present other unusually troublesome problems for the courts. If a court concluded, through some mystical process, that the attorney originally selected might have obtained a higher verdict for a plaintiff or a lower verdict for a defendant, and the extent of that difference in result, how much higher or lower would that verdict have to be before sufficient prejudice would attach to warrant reversal? In short, in civil cases the courts would be plunged into a morass of speculation were review of the propriety of the granting of a motion to disqualify counsel delayed until after the trial had occurred.

In determining whether collateral orders are immediately appealable, we must also weigh the interests of "judicial economy" and the potential misallocation of judicial resources. Bradshaw v. Zoological Society of San Diego, 662 F.2d 1301, 1315 (9th Cir.1981). Our concern for avoiding a waste of judicial resources stems primarily from the effect that such waste has upon litigants, not only upon those involved in the particular case at issue but also upon all whose cases are awaiting judicial action. While delaying appeals might conceivably result in a judicial savings in some cases in which the district court's order is ultimately upheld or a post-trial appeal is not taken, this factor is outweighed by the inordinate loss of judicial time and resources which would occur in attempting to reconstruct hypothetical trials, and then conducting new trials, in cases where it is determined that the district court erred. Moreover, since a factual showing as to how the original counsel would have handled the case differently would be necessary before reversible error could be found, the district court would undoubtedly be required to conduct an onerous post-trial reconstruction in all cases in which a party whose originally chosen counsel was disqualified desires to file an appeal.4

For the above reasons, we agree with the Fifth Circuit that, in civil cases, an order disqualifying a party's counsel is "effectively unreviewable" after final judgment and is immediately appealable under 28 U.S.C. Sec. 1291 pursuant to the Cohen collateral order exception. Accordingly, we apply the rule of Petroleum Products to such cases and hold that we have jurisdiction over the appeal before us.

Having determined the jurisdictional question in the manner set forth in this opinion, we affirm the district court for the reasons given in our simultaneously filed memorandum disposition.

AFFIRMED.