dissenting:
We should grant the writ of mandamus and direct Judge Pollack to recuse himself from consideration of these cases.
Events which no one could have foreseen, and which no one arranged, have created a reasonable doubt of Judge Pollack’s impartiality in supervising these cases. Drexel inadvertently has been brought into a relationship with the financial interests of the wife of the judge who happens to be responsible for deciding litigation of great public importance brought against Drexel and four of its principals by the Securities and Exchange Commission (SEC) and individual plaintiffs. Judge Pollack has refused to recuse himself. The sole question before us is whether, in view of the wide publicity accorded these so-called insider trading cases, Judge Pollack’s “impartiality might reasonably be questioned”, 28 U.S.C. § 455(a), by a reasonable person knowing the relevant facts.
To me, the inescapable relevant fact is that Drexel has been, and is now, retained by the firm which is under contract with Mrs. Pollack and members of her family to arrange financing for the cash purchase of Palais Royal, their family business, from which members of the family will receive over $84 million and Mrs. Pollack herself, and as trustee, will receive $30 million in cash. It is clear to me that a reasonable person knowing these ultimate conceded facts would reasonably question Judge Pollack’s ability to supervise such litigation impartially. Such a reasonable person’s next question would be “Why hasn’t the judge stepped aside?” Moreover, Judge Pollack’s expressed resentment to the suggestion of recusal and his castigation of Drexel and its counsel has confirmed these doubts. He cannot act impartially in these cases.
There is absolutely no evidence that Drexel agreed to act for Bain in order to manufacture a situation in which Judge Pollack’s continued participation in ongoing litigation could be challenged. Drexel became involved in the Palais Royal deal in the same manner in which it might become involved in any of its deals. First, the stockholders of Palais Royal, consisting of Mrs. Pollack and family members, agreed on June 29, 1988 to sell the company to Bain Venture Capital (Bain) for over $84 million, on condition that Bain obtain financing for the purchase. Thereafter, Bain arranged with Drexel to obtain the necessary financing for the purchase of Palais Royal and another business with which Palais Royal is to be merged, in what is known as a leveraged buyout (LBO). It is undisputed that Bain sought out Drexel, a leading company in the successful and profitable managing of LBOs; Drexel did not seek to create a disqualification issue.
Moreover, there is no evidence that Drex-el is not committed to this transaction. Although the majority suggests that Drexel may not be contractually obligated to Bain, the evidence before us leads to the opposite conclusion. The majority’s attempt to deemphasize the nexus between Palais Royal and the instant litigation by hypothesizing that a court conceivably might find Drexel uncommitted to Bain is to brush aside both the uncontroverted evidence and the reasonableness standard of § 455(a). Drexel already has agreed to supply in excess of $200,000,000; it has completed an extensive “due diligence” investigation. It also has an option to purchase 15% of the *1318acquiring company. Although the June agreement fixes no closing date, the parties expect to close the deal sometime in November 1988.
The record is similarly devoid of any evidence which suggests that Drexel waited until an opportune moment to “discover” the conflict. On September 7, 1988, the SEC filed suit against Drexel and four of its principals seeking injunctive relief and disgorgement of profits. On the SEC’s suggestion that this new action was related to other civil suits already supervised by Judge Pollack, it was assigned to him. Two days later, on September 9, Bain advised Cahill Gordon & Reindel (Cahill Gordon), Drexel’s counsel, that the Moselle Pollack who had signed the agreement with Bain and who had a $30 million interest in the Palais Royal buyout was the wife of Judge Pollack. Drexel’s attorney informed Judge Pollack by telephone on the very next day of Mrs. Pollack’s interest in the deal.
When it became known to counsel for Drexel that the Moselle Pollack who would benefit from the buyout was the wife of Judge Pollack, the judge conducting the cases involving Drexel, it was their duty to bring the matter to the judge’s attention without delay. The duty to inform a judge of a potential conflict rests upon whichever party becomes aware of such a situation. Any feelings which either party may have about the judge’s favoring one or the other are wholly immaterial. To save embarrassment, time and expense, and the validity of any judgment, counsel have a duty to act promptly.
According to the undisputed evidence in the record before us, the Cahill Gordon attorneys who were representing Drexel in these cases did not know until September 9 that Moselle Pollack was the wife of Judge Pollack. Different attorneys in the firm had been representing Mrs. Pollack in her negotiations for the sale of Palais Royal, and nothing in the record indicates that anyone at Cahill Gordon knew of the link before September 9. If, as the SEC suggests, Drexel was anxious to get Judge Pollack out of these cases, it would seem that someone at Cahill Gordon would have acted as soon as the firm became aware that the judge was Moselle Pollack’s husband. In sum, there is no reason to believe that Cahill Gordon could have acted any sooner than it did.
At this point in the proceedings, Judge Pollack’s resentment at the idea of recusal became apparent. After conference with counsel in his chambers on September 13, Judge Pollack refused to consider recusal. He called the suggestion a “cockamamie story” and threatened counsel with sanctions for acting on “insufficient information.”
On September 20, the petitioners filed a formal motion to recuse, returnable October 4, which Judge Pollack adjourned to October 11th. As the judge continued to rule on pretrial matters in the private civil cases, the petitioners sought a writ of mandamus on September 30. We denied the petition as premature in view of the hearing scheduled for October 11.
Meanwhile, on October 10, the day before he was to hear the recusal motion, Judge Pollack, on his own motion, issued an order directing Paul, Weiss, Rifkind, Wharton & Garrison (Paul, Weiss), which represents Michael R. Milken, a top Drexel official involved in the SEC litigation and the. petition for recusal, and two of its partners, Arthur Liman and Martin Flu-menbaum, to show cause why “they should not be required to identify and withdraw the papers on said disqualification motion which they have drafted, filed and served herein.” With the order, Judge Pollack served copies of documents from his family’s files to support his charge that, by joining in the recusal motion on behalf of the individual principals of Drexel, the Paul, Weiss firm was undertaking “conflicting” representation and acting adversely toward the Pollack family members, whom they had represented in other matters. At the October 11 hearing, Judge Pollack refused to hear argument on his order and referred the matter to Chief Judge Brieant “for disciplinary resolution, if required.”
*1319It appears from this response to the re-cusal motion that the judge viewed the suggestion that he recuse himself from the Drexel litigation as adverse to the interests of his family — as if his continued participation in this litigation would be favorable to those family interests.
Judge Pollack’s hostility towards the moving parties was repeated in his opinion of October 17. Judge Pollack emphasized the lack of “privity” between Mrs. Pollack and Drexel, writing that Mrs. Pollack’s transaction “is not a Drexel-financed transaction.” He also made much of the argument that Drexel is not essential to the deal and that Bain could easily retain one of Drexel’s competitors to provide the required financing. In finding no reasonable basis for doubting his impartiality, Judge Pollack wrote that the “speculations” in which Drexel’s lawyers engaged to “concoct[ ]” perceptions of impropriety seem so “far-fetched” as reactions of a reasonable person as to be “ludicrous”.
My colleagues, in denying the petitioners’ renewed application for mandamus, also rely upon the lack of “privity” between Mrs. Pollack and Drexel. Their argument rests on the fact that, instead of Drexel paying Mrs. Pollack with its own check for $30 million, Drexel will arrange the payment to a Bain subsidiary created for that purpose, which in turn will pay the $30 million to Mrs. Pollack. This can fool no one regarding who is responsible for raising the money and causing it to get to Mrs. Pollack. According to present plans, Mrs. Pollack will not get $30 million but for Drexel.
In this case, to impute a lack of privity is to exalt the form of the transaction over its substance, and to ignore the reasonableness standard of § 455(a), which requires us to judge the situation from the viewpoint of the reasonable person, and not from a purely legalistic perspective. Drex-el is integrally involved in this transaction and has been so for several months. It is largely because of Drexel’s efforts that Mrs. Pollack will receive $30 million. That is the common sense of it; and it most certainly is the public perception that Mrs. Pollack is to be paid $30 million thanks to Drexel.
Of course, there might have been others who would be interested in working on the deal, but that is irrelevant. If Bain could have made other arrangements with a firm as well qualified as Drexel and on equally advantageous terms, it probably would have done so during the weeks when it was working on the deal and before it chose Drexel. In the absence of any showing of' artifice on Drexel’s part, the question is whether Judge Pollack reasonably appears not to be impartial. It is beside the point to assert that there is no showing that Drexel is indispensable.
The nature and extent of the interest of Mrs. Pollack and her family in the deal from which the family members will receive no less than $84 million is obviously of concern to Judge Pollack. His interest is considerably stronger than was the fiduciary interest of Judge Robert Collins, as á trustee of Loyola University, which the Supreme Court recently considered in Liljeberg v. Health Services Acquisition Corp., — U.S.-, 108 S.Ct. 2194, 100 L.Ed.2d 855 (1988). In Liljeberg, Loyola had contracted to sell a tract of its land for about $6.5 million to John Liljeberg, Jr. for the construction of a hospital, subject to a buyback clause which would be triggered if Liljeberg did not secure a construction contract within one year and Loyola did not receive certain benefits from the anticipated rezoning of Loyola’s land adjacent to the hospital. At the same time, Judge Collins was presiding over an action in which a third party, with whom Liljeberg was contemplating building a hospital on a different site, was challenging Liljeberg’s ownership of a state-issued “certificate of need” for a hospital. Had Liljeberg not won ownership of the certificate, which is required for the successful operation of a hospital, it is likely that the buyback provision would have been triggered and that Loyola would have lost the benefits associated with the hospital’s construction.
Unlike Judge Pollack, who has been aware of his conflicting interest since the third day after the SEC filed suit, Judge *1320Collins did not actually know about his fiduciary interest in the litigation until after he had filed his opinion. The Supreme Court nevertheless held that the judge’s attendance at meetings of the Board of Trustees at which the prospective sale was discussed raised reasonable doubts about his impartiality and vacated Judge Collins’ judgment.
The Liljeberg Court squarely held that whenever a question involving disqualification is presented, it is critical promptly to examine all possible bases for the judge’s disqualification. Liljeberg, supra, 108 S.Ct. at 2205. The Court found unconvincing the facts that Loyola was not a party to the suit, that the judge was unaware of the conflict during the pendancy of the case, and that the judge’s interest was only as a fiduciary rather than as a personal investor. What persuaded the Supreme Court to vacate the judgment after it was entered was the appearance of impropriety. As the Liljeberg Court said: “[Pjeople who have not served on the bench are often all too willing to indulge suspicions and doubts Concerning the integrity of judges. The very purpose of § 455(a) is to promote confidence in the judiciary by avoiding even the appearance of impropriety whenever possible.” 108 S.Ct. at 2204-05 (footnote omitted).
Such an appearance of impropriety is, in fact, usually avoided by our federal judges as a matter of course. Since well before the amendment of § 455 in 1974, it has been the custom of federal judges in the Second Circuit and elsewhere to recuse themselves whenever they or their spouses hold any stock or have any financial interest of any kind, however small, in any company or enterprise which is a party, or is linked with any party, in any matter coming before them. Each year every judge, including every senior judge who still sits, must file a statement of the stock holdings and other investments of the judge and the judge’s spouse. That statement also reflects the sales and acquisitions of such interests during the preceding year. The spirit of these practices has been to resolve any doubt in favor of recu-sal. See United States v. Murphy, 768 F.2d 1518, 1536-41 (7th Cir.1985).
Moreover, the American Bar Association Code of Judicial Conduct requires that a judge recuse himself whenever “he knows that he ... or his spouse ... has a financial interest in the subject matter in controversy or in a party to the proceeding, or any other interest that could be substantially affected by [its] outcome.” A.B.A. Code of Judicial Conduct Canon 3C(l)(c). Given the public attention that has been focused on the disclosures regarding insider trading and the public concern for the integrity of the marketplace, there is a heightened awareness of the importance of insuring the absolute impartiality of any judge who participates in the resolution of litigation concerning such activities. If Judge Pollack remains in charge of these cases, Mrs. Pollack’s financial interest in Drexel’s successful completion of the deal can only exacerbate these public concerns.
The cases relied upon in opposition to the petition for mandamus are inapposite. In Mavis v. Commercial Carriers, Inc., 408 F.Supp. 55 (C.D.Cal.1975), Judge Hauk held that recusal was not required where the judge owned some common stock in a subsidiary, twice removed, of a corporation that had done some business with the defendant company. In Mavis, the corporation indirectly owning the company in which the judge held stock had done business with the defendant in the past. The Palais Royal buyout, on the other hand, is happening now.
In re Placid Oil Co., 802 F.2d 783 (5th Cir.1986), held that the judge’s ownership of stock in a bank did not mandate his recusal from a case involving unrelated banks. The argument in support of his recusal, that the outcome of the litigation might affect the banking industry generally, was speculative and too remote from the judge’s investment in a different bank. 802 F.2d at 786-87. Mrs. Pollack’s imminent realization of $30 million from the deal which Drexel is financing is hardly so speculative or remote.
Finally, I turn to the appropriateness of mandamus in this situation. While manda*1321mus is indeed an extraordinary remedy, no cases have been cited that would bar the issuance of the writ in a situation as extraordinary as that before us. On the contrary, we have repeatedly held that mandamus is appropriate in recusal motions, believing that there are “few situations more appropriate for mandamus than a judge’s clearly wrongful refusal to disqualify himself.” In re International Business Machines Corp., 618 F.2d 923, 926 (2d Cir.1980) (quoting from Rosen v. Sugarman, 357 F.2d 794, 797 (2d Cir.1966)). See also In re United States, 666 F.2d 690, 694 (1st Cir.1981) (public confidence in the courts requires that questions of disqualification must be disposed of at the earliest possible opportunity). Moreover, we have held that the standard for review of a judge’s refusal to recuse himself is not an abuse of discretion standard. Rather, we must query whether the trial judge could exercise that discretion in the face of a “personal, extrajudicial bias which precludes dispassionate judgment.” In re International Business Machines, supra, at 926.
Fortunately, the proceedings in the suit brought by the SEC have just begun, and in the private civil cases discovery and pretrial proceedings are still underway. It is not unusual for cases to be transferred to other judges in the court as the need to do so arises. In addition to Judge Pollack, there are presently 23 active district court judges in the Southern District of New York and more than a dozen senior district judges who still sit. In the case before us, in which Drexel is underwriting a transaction in which Judge Pollack’s wife will receive $30 million, we must act to avoid compounding the harm already done to the public’s perception of the integrity of the judiciary.
A central pillar of our government is the people’s confidence that our courts are free of bias or favor and are administered by judges who are unquestionably impartial. Whenever any reasonable basis to doubt a judge’s impartiality exists, this public trust demands that we act swiftly and decisively.
We should grant the petition for mandamus and direct Judge Pollack to recuse himself.