OPINION OF THE COURT
ADAMS, Circuit Judge.This appeal presents an issue important to judicial administration, namely, whether a district court may order an attorney to pay to the government the cost of impanelling a jury for one day as a sanction for the attorney’s abuse of the judicial process.
I
The underlying action in this case was a personal injury suit brought in a Pennsylvania state court in 1982 by plaintiffs Irvin and Yvonne Eash. Upon motion by the defendants, the case was removed to the District Court for the Western District of Pennsylvania, pursuant to 28 U.S.C. § 1441 (1982). Settlement negotiations took place between the parties in the spring of 1983, leading to a stipulation to dismiss the case. Consequently, on August 1,1983, the district court dismissed the action.
On August 12, 1983, without giving notice to the parties or conducting a hearing, the district court entered an order requiring defendants’ counsel to pay $390 to the Clerk of Court. A copy of the order appears in the appendix. The case was scheduled for trial before a jury on May 23, 1983, and according to the district court judge, plaintiffs’ attorney made repeated attempts to communicate with defendants’ counsel regarding settlement possibilities during the week of May 16, 1983. Receiving no response, plaintiffs’ attorney prepared for trial and came to court on May 23. At that point, defendants’ counsel proposed a settlement figure that plaintiffs accepted. The district court believed, although this is disputed, that because defendants’ attorney was scheduled for trial in state court that same day, the settlement avoided a scheduling conflict.
The district court concluded that under the circumstances “settlement on the eve of trial was not justified. Defendants’ attorney was given adequate notice by plaintiffs’ counsel and by court personnel to attempt to reach an agreement.” App. at 8A. The court therefore imposed a sanction of $390 on defendants’ attorney, calculated as follows: $30, the per diem fee for each juror, multiplied by 13, the minimum number of persons necessary to select a jury. Subsequently the court denied a petition for reconsideration submitted by defendants’ counsel, in which counsel disputed the factual basis of the order.
After a timely appeal was filed, the case was listed for resolution by a panel of this Court.1 Because of the importance of the questions presented, however, the Court in banc has reviewed the case.
II
We first address appellants’ contention that the district court was without jurisdiction to enter the order imposing the sanction because the order was entered eleven days after the dismissal of the case. It is clear that there are various contexts in which the district court retains jurisdiction over particular matters after a judgment has been entered. For example, the court retains jurisdiction to fix costs under Fed. R.Civ.P. 54(d) and Fed.R.Civ.P. 58, even after dismissal of the underlying action. See 10 C. Wright & A. Miller, Federal Practice and Procedure § 2668 (1983); cf. Richards v. Government of Virgin Islands, 579 F.2d 830, 833 (3d Cir.1978). In addition, courts generally have recognized that there is retained jurisdiction to assess attorney’s fees pursuant to 28 U.S.C. § 1927 (1982), within a reasonable time after a judgment on the merits is entered. See, e.g., Overnite Transportation Co. v. *560Chicago Industrial Tire Co., 697 F.2d 789, 793 (7th Cir.1983); Obin v. District of the International Assoc. of Machinists & Aerospace Workers, 651 F.2d 574 (8th Cir. 1981).
Assuming the district court had the authority to impose the sanction in this matter, we find no jurisdictional obstacle to the entry of the order in question eleven days after the underlying case was disposed of on the merits.
Ill
Counsel also challenges the district court’s power to impose the cost of impanelling a jury as a sanction for his misconduct. The district court did not specify the authority upon which it relied; however, there appear to be two possible sources: 28 U.S.C. § 1927 (1982) or the court’s inherent power.
As amended in 1980, 28 U.S.C. § 1927 states:
Any attorney or other person admitted to conduct cases in any court of the United States or any Territory thereof who so multiplies the proceedings in any case unreasonably and vexatiously may be required by the court to satisfy personally the excess costs, expenses, and attorneys’ fees reasonably incurred because of such conduct.
A threshold question is whether the terms “costs” and “expenses” contained in the statute include the per diem juror fees paid by the government. If not, the statute would appear to provide no authority, for the order in question.
In Roadway Express, Inc. v. Piper, 447 U.S. 752, 100 S.Ct. 2455, 65 L.Ed.2d 488 (1980), the Supreme Court held that “costs” under § 1927 are limited to those costs permitted under 28 U.S.C. § 1920 (1982), which enumerates the items that ordinarily may be taxed to a losing party.2 The Court reasoned that § 1927 should be read in pari materia with § 1920 because the two sections are part of an integrated legislative effort, first approved in 1853, to allow the award of excess costs against lawyers who vexatiously multiply litigation. 447 U.S. at 760, 100 S.Ct. at 2461. The Supreme Court concluded that the
most reasonable construction is that the [original Act containing the present § 1920] defined those costs that may be recovered from counsel. Congress, of course, may amend these provisions that derive from the 1853 Act. In the absence of express modification of those provisions by Congress, however, we should not look beyond the Act for the definition of costs under § 1927.
Id. at 760, 100 S.Ct. at 2461 (footnote omitted).3 Neither § 1920 nor § 1927 contains reference to the costs of impanelling a jury, costs which customarily are borne by the government. Only the opposing litigants’ costs and expenses incurred by virtue of an attorney’s misconduct are within the ambit of the statutes. Thus, the Supreme Court’s decision in Roadway precludes reliance on § 1927 as granting district courts the authority to impose the sanction employed in this case. See also United States v. Blodgett, 709 F.2d 608 (9th Cir.1983); United States v. Ross, 535 F.2d 346 (6th Cir.1976) (§ 1927 does not authorize assessment of juror costs).4
IY
A.
The amicus suggests that the district court’s order is most properly viewed as an *561exercise of the court’s inherent power. That courts have inherent powers — powers vested in the courts upon their creation, see Michaelson v. United States, 266 U.S. 42, 66, 45 S.Ct. 18, 20, 69 L.Ed. 162 (1924); Ex parte Robinson, 86 U.S. (19 Wall.) 505, 510, 22 L.Ed. 205 (1874); Anderson v. Dunn, 19 U.S. (6 Wheat.) 204, 227, 5 L.Ed. 242 (1821), and not derived from any statute, see Link v. Wabash Railroad Co., 370 U.S. 626, 630, 82 S.Ct. 1386, 1388, 8 L.Ed.2d 734 (1962); United States v. Hudson, 11 U.S. (7 Cranch) 32, 34, 3 L.Ed. 259 (1812) — is not disputed. Inherent power has been frequently invoked by the courts to regulate the conduct of the members of the bar as well as to provide tools for docket management. Courts have thus relied on the concept of inherent power to impose several species of sanctions on those who abuse the judicial process. For example, federal courts may dismiss a case for failure to prosecute, Link v. Wabash Railroad Co., 370 U.S. at 629-30, 82 S.Ct. at 1388-89; see also National Hockey League v. Metropolitan Hockey Club, Inc., 427 U.S. 639, 96 S.Ct. 2778, 49 L.Ed.2d 747 (1976) (per curiam); Titus v. Mercedes Benz of North America, 695 F.2d 746 (3d Cir.1982). Similarly, the contempt power is rooted in the inherent power of the judiciary. E.g., Levine v. United States, 362 U.S. 610, 615, 80 S.Ct. 1038,1042, 4 L.Ed.2d 989 (1960). Commentators have also noted occasions in which, under its inherent power, a court has disbarred, suspended from practice, or reprimanded attorneys for abuse of the judicial process. See Comment, Financial Penalties Imposed Directly Against Attorneys in Litigation Without Resort to the Contempt Power, 26 UCLA L.Rev. 855, 856 (1979); Comment, Involuntary Dismissal for Disobedience or Delay: The Plaintiff’s Plight, 34 U.Chi. L.Rev. 922, 937 n. 96 (1967); accord Ex parte Wall, 107 U.S. 265, 288-89, 2 S.Ct. 569, 588-89, 27 L.Ed. 552 (1883); cf. Spevack v. Klein, 385 U.S. 511, 524, 87 S.Ct. 625, 633, 17 L.Ed.2d 574 (1967) (Harlan, J., dissenting) (courts “have endeavored to regulate the qualification and practice of lawyers, always in hope that this might better assure the integrity and evenhandedness of the administration of justice ... since the 17th century”). This Court has stated that in the absence of a statute, the taxation of costs in the appellate court “is a matter, inherently and necessarily within its general powers.” See Island Development Co. v. McGeorge, 37 F.2d 345, 345 (3d Cir.1930); see also Levin & Amsterdam, Legislative Control Over Judicial Rule-Making: A Problem of Constitutional Revision, 107 U.Pa.L.Rev. 1, 16 (1958). Courts, pursuant to inherent powers, have declared attorneys who choose to be absent from docket call “ready for trial,” even though this may lead ineluctably to the entry of a default judgment. Williams v. New Orleans Public Service, Inc., 728 F.2d 730, 732 (5th Cir.1984); see also Schlesinger v. Teitelbaum, 475 F.2d 137, 142 (3d Cir.), cert. denied, 414 U.S. 1111, 94 S.Ct. 840, 38 L.Ed.2d 738 (1973) (an inherent power to set counsel fees in cases involving persons of presumed incapacity).
Despite historical reliance on inherent powers, including Supreme Court jurisprudence dating back to 1812, the notion of inherent power has been described as nebulous,5 and its bounds as “shadowy.” See Rosenberg, Sanctions to Effectuate Pretrial Discovery, 58 Colum.L.Rev. 480, 485 (1958). The conceptual and definitional problems regarding inherent power that have bedeviled commentators for years, see Burbank, Sanctions in the Proposed Amendments to the Federal Rules of Civil Procedure: Some Questions About Power, 11 Hofstra L.Rev. 997, 1004 (1983); see also Sanctions Imposable, supra n. 5, stem from several factors. First, perhaps because federal courts infrequently resort to their inherent powers or because such reliance most often is not challenged, very few federal cases discuss in detail the topic of inherent powers. Cf. Note, Power of *562Federal Courts to Discipline Attorneys for Delay in Pretrial Procedure, 38 Notre Dame Law. 158, 161 (1963).6 More importantly, those cases that have employed inherent power appear to use that generic term to describe several distinguishable court powers. Cf Williams, The Source of Authority for Rules of Court Affecting Procedure, 22 Wash.U.L.Q. 459, 473-74 (1937); Frankfurter & Landis, Power of Congress over Procedure in Criminal Contempts in “Inferior” Federal Courts — A Study in Separation of Powers, 37 Harv.L.Rev. 1010, 1023 (1924). To compound this lack of specificity, courts have relied occasionally on precedents involving one form of power to support the court’s use of another. See Burbank, supra, 11 Hofstra L.Rev. at 1005.
These observations suggest that it is not always possible to categorize inherent power decisions. Nevertheless, it appears that the term inherent power has been employed in three general fashions. The first stems from the fact that once Congress has created lower federal courts and demarcated their jurisdiction, the courts are vested with judicial powers pursuant to Article III. This use of inherent power, which might be termed irreducible inherent authority, encompasses an extremely narrow range of authority involving activity so fundamental to the essence of a court as a constitutional tribunal that to divest the court of absolute command within this sphere is really to render practically meaningless the terms “court” and “judicial power.” See Levin & Amsterdam, supra, 107 U.Pa.L.Rev. at 30-32. In this limited domain of judicial autonomy, courts may act notwithstanding contrary legislative direction. These inherent powers are grounded in the separation of powers concept, because to deny this power “and yet to conceive of courts is a self-contradiction.” Frankfurter & Landis, supra, 37 Harv.L.Rev. at 1023; see Levin & Amsterdam, supra, 107 U.Pa.L.Rev. at 33; see also United States v. Klein, 80 U.S. (13 Wall.) 128, 147, 20 L.Ed. 519 (1872).
Boundaries for this sphere of minimal judicial integrity are not possible to locate with exactitude.7 Certainly the power must be exercised with great restraint and caution. Whatever the proper limit of this form of inherent power may be, it is not the power to which the amicus has pointed as authority for the sanction imposed by the district court here.
The second, and most common, use of the term “inherent power” encompasses those powers sometimes said to arise from the nature of the court, see Ex parte Terry, 128 U.S. 289, 303, 9 S.Ct. 77, 79, 32 L.Ed. 405 (1888); United States v. Hudson, 11 U.S. (7 Cranch) at 34, 3 L.Ed. 259, but more often thought to be the powers “necessary to the exercise of all others.” E.g., Roadway, 447 U.S. at 764, 100 S.Ct. at 2463 (quoting Hudson). Here courts are referring to powers implied from strict functional necessity. In Roadway the Supreme Court termed the contempt sanction “the most prominent” of these powers.8 Histor*563ically, it has viewed this particular power as “essential to the administration of justice,” Michaelson, 266 U.S. at 65, 45 S.Ct. at 20, and “absolutely essential” for the functioning of the judiciary. Levine v. United States, 362 U.S. 610, 616, 80 S.Ct. 1038, 1042, 4 L.Ed.2d 989 (1959); see also Cooke v. United States, 267 U.S. 517, 539, 45 S.Ct. 390, 396, 69 L.Ed. 767 (1925); Myers v. United States, 264 U.S. 95, 103, 44 S.Ct. 272, 273, 68 L.Ed. 577 (1924).
Because this second form of inherent power arises from necessity, on several occasions the Supreme Court has stated that while the authority “may be regulated within limits not precisely defined,” it can “neither be abrogated nor rendered practically inoperative.” Michaelson, 266 U.S. at 66, 45 S.Ct. at 20. Indeed, in one of its earliest decisions regarding the contempt power, the Supreme Court observed that congressional enactment of a contempt statute should “be considered, only as an instance of abundant caution, or a legislative declaration, that the power of punishing for contempt shall not extend beyond its known and acknowledged limits____” Anderson, 19 U.S. (6 Wheat.) at 226, 5 L.Ed. 242; see also Cammer v. United States, 350 U.S. 399, 404, 76 S.Ct. 456, 458, 100 L.Ed. 474 (1956); United States v. Hall, 198 F.2d 726, 728 (2d Cir.1952), cert. denied, 345 U.S. 905, 73 S.Ct. 641, 97 L.Ed. 1341 (1953) (statute merely “defines the acts toward which it may be directed”).
The third form of authority subsumed under the general term inherent power implicates powers necessary only in the practical sense of being useful. An early example is Ex parte Peterson, 253 U.S. 300, 40 S.Ct. 543, 64 L.Ed. 919 (1920), in which the Supreme Court determined that “the court possesses the inherent power to supply itself” with an “auditor” to aid in its decisionmaking. Id. at 312, 40 S.Ct. at 547. “Courts have (at least in the absence of legislation to the contrary) inherent power to provide themselves with appropriate instruments required for the performance of their duties,” and to appoint “persons unconnected with the court to aid judges in the performance of specific judicial duties.” Id., see also Ruiz v. Estelle, 679 F.2d 1115, 1161 (5th Cir.1982), cert. denied, 460 U.S. 1042, 103 S.Ct. 1438, 75 L.Ed.2d 795 (1983); Schwimmer v. United States, 232 F.2d 855, 865 (8th Cir.), cert. denied, 352 U.S. 833, 77 S.Ct. 48, 1 L.Ed.2d 52 (1956).9
This third category of inherent power has sometimes been said to be “rooted in the notion that a federal court, sitting in equity, possesses all of the common law equity tools of a Chancery Court (subject, of course, to congressional limitation) to process litigation to a just and equitable conclusion.” ITT Community Development Corp. v. Barton, 569 F.2d 1351, 1359 (5th Cir.1978); cf. Hall v. Cole, 412 U.S. 1, 5, 93 S.Ct. 1943, 1946, 36 L.Ed.2d 702 (1973) (courts possess “inherent equitable power”); Johnston v. Marsh, 227 F.2d 528, 531 (3d Cir.1955) (“Our Federal judiciary has consistently recognized that at common law this inherent power existed.”). In other cases this power is said to derive from necessity; for example, the court termed “essential” the appointment of an auditor in Peterson, 253 U.S. at 312, 40 S.Ct. at 547. Yet it is clear that such power is necessary only in the sense of being highly useful in the pursuit of a just result. See Note, Compulsory Reference in Actions at Law, 34 Harv.L.Rev. 321, 324 (1921).
As suggested by the above quotation from Peterson, courts may exercise this kind of inherent power only in the absence of contrary legislative direction. See Williams, supra, 22 Wash.U.L.Q. at 473; see also Alyeska Pipeline Service v. Wilderness Society, 421 U.S. 240, 259, 95 S.Ct. 1612, 1622, 44 L.Ed.2d 141 (1975) (“These exceptions are unquestionably assertions of inherent power in the courts to allow attorneys’ fees in particular situations, unless forbidden by Congress____”).
*564The third form of power has also been invoked as “the underlying federal basis that permits the court to elect to use” a state mechanism for certification of a question of doubtful state law, Weaver v. Marine Bank, 683 F.2d 744 (3d Cir.1982); in order to grant bail in a situation not dealt with by statute, Johnston v. Marsh, 227 F.2d at 531; see also Wright v. Henkel, 190 U.S. 40, 63, 23 S.Ct. 781, 787, 47 L.Ed. 948 (1903); In re Gannon, 27 F.2d 362, 363 (E.D.Pa.1928); and to dismiss a suit pursuant to the doctrine of forum non conveniens, Gulf Oil Corp. v. Gilbert, 330 U.S. 501, 502, 507, 67 S.Ct. 839, 840, 842, 91 L.Ed. 1055 (1947).
Courts rarely have explained exactly what kind of authority they mean to invoke when using an inherent power to sanction an attorney. See Dowling, The Inherent Power of the Judiciary, 21 A.B.A.J. 635 (1935). Nevertheless, it seems quite clear that at least in the absence of contrary legislation, courts under their inherent powers have developed a wide range of tools to promote efficiency in their courtrooms and to achieve justice in their results. E.g., Link, 370 U.S. at 630-31, 82 S.Ct. at 1388-89 (where the Supreme Court stated that the authority to dismiss a case is an inherent “control necessarily vested in courts to manage their own affairs so as to achieve the orderly and expeditious disposition of cases”); see also Moulton v. Commissioner of Internal Revenue, 733 F.2d 734, 735 (10th Cir.1984); Barnd v. City of Tacoma, 664 F.2d 1339, 1342 (9th Cir.1982); Penthouse International Ltd. v. Playboy Enterprises, Inc., 663 F.2d 371, 386 (2d Cir.1981).
B.
Nearly a quarter century ago, and more than fifteen years prior to the Supreme Court’s decision in Roadway, this Court sitting in banc held that a district court lacked inherent authority to impose a “fine” on an attorney who had failed to file a pretrial memorandum in a timely fashion. Gamble v. Pope & Talbot, Inc., 307 F.2d 729 (3d Cir.), cert. denied, 371 U.S. 888, 83 S.Ct. 187, 9 L.Ed.2d 123 (1962). The plurality of the Court believed that the effect of the fine, whatever “it be called ... was to punish defendant’s attorney for contempt in failing to file the defense pretrial memorandum within time.” Id. at 731. The district court in Gamble did not, however, designate the lawyer’s act as contempt nor did it invoke the formal contempt proceeding. On review, this Court noted that there was nothing in the Federal Rules of Civil Procedure which explicitly authorized the sanction that had been imposed. Id. While the plurality conceded that district courts have substantial local rulemaking power, it determined that such a “basic disciplinary innovation” as the sanction in question required a national “uniform approach.” Id. at 732. The plurality believed that only a statute or Federal Rule could accomplish this. Significantly, while stating that no inherent authority for the fine existed, neither the plurality opinion nor the concurring statement of Judge Hastie discussed the doctrine of inherent powers, and neither opinion cited any case involving inherent power.
Judge Goodrich, dissenting, asserted that a judge “undoubtedly has inherent power to impose sanctions for the disciplining of lawyers who, in matters not amounting to contempt, do not obey rules.” Id. at 737 (Goodrich, J., dissenting). In a separate dissent Chief Judge Biggs disputed the majority’s perception that the fine carried any criminal connotation. He emphasized that unlike many of the penalties that courts impose for an attorney’s failure to perform his or her duty, the monetary sanction did not punish clients for the shortcomings of their counsel. Id. at 734.10
*565The decision in Gamble was promptly and roundly criticized. E.g., Note, Power of Federal Courts, 38 Notre Dame Law. 158; Comment, Dismissal for Failure to Attend a Pretrial Conference and the Use of Sanctions at Preparatory Stages of Litigation, 72 Yale L.J. 819 (1963). And the case has continued to come under attack. See, e.g., Renfrew, Discovery Sanctions, A Judicial Perspective, 67 Calif.L. Rev. 264, 270 (1979); Comment, Sanctions Imposed by Courts on Attorneys Who Abuse the Judicial Process, 44 U.Chi.L. Rev. 619, 635 (1977). Commentators have highlighted the curious and contradictory result created by the juxtaposition of Link, decided in 1962, and Gamble, decided one year later. Under Link, courts may invoke their inherent power to dismiss a plaintiffs entire case when a lawyer has failed without adequate excuse to appear at a scheduled pretrial conference, Link, 370 U.S. at 628-29, 82 S.Ct. at 1387-88, but under Gamble courts may not sanction financially a blameworthy attorney who filed a pretrial memorandum ten months late, Gamble, 307 F.2d at 730. See, e.g., Comment, Financial Penalties, 26 UCLA L.Rev. at 877.
Recently a number of courts have rejected the result and reasoning of Gamble as unduly narrow. See, e.g., Miranda v. Southern Pacific Transp. Co., 710 F.2d 516, 520 (9th Cir.1983) (“For the reasons stated by Judge Biggs, we decline to follow Gamble.”); Martinez v. Thrifty Drug & Discount Co., 593 F.2d 992, 993 (10th Cir. 1979) (explicitly rejecting Gamble); In re Sutter, 543 F.2d 1030, 1037 (2d Cir.1976) (“We ... decline to follow the Third Circuit.”); see also Richman v. General Motors Corp., 437 F.2d 196, 200 (1st Cir.1971). A major factor in the decisions, that have refused to follow Gamble has been the startling increase in the number and complexity of cases filed in the federal courts. The dramatic rise in litigation in the last decade has led trial judges to conclude that indulgent toleration of lawyers’ misconduct is simply a luxury the federal court system no longer can afford. See Renfrew, supra, 67 Calif.L.Rev. at 275-76. Chief Justice Burger has recently observed that “a small handful [of lawyers] must not be permitted to abuse the system and preempt its time and machinery for purposes not intended, thus delaying and denying” access to courts to others in need of the courts’ limited resources. Burger, Abuses of Discovery, Trial (Sept.1984). For example, the Second Circuit in Sutter upheld the imposition of juror costs based in part upon “intensified concerns” over “the increasing backlog of calendars.” 543 F.2d at 1037. The Martinez court concluded that a sanction based on the cost of impanelling a jury had the “object and purpose of administering the court in an efficient manner.” 593 F.2d at 994.
C.
With the foregoing discussion of inherent powers and the evolution of practice regarding attorney sanctions as background, we turn to the challenged order at issue here. Appellants, relying on the continuing vitality of the reasoning in Gamble, urge that a sanction in the form of juror costs improperly and “informally inflict[ed] a criminal” punishment, like contempt, without resort to the contempt statute. Gamble, 307 F.2d at 733. To a considerable extent, Judge Sloviter’s dissent also rests on this view.
In Gamble this Court essentially was concerned with the imposition of a fine unrelated to any actual consequence of counsel’s conduct and that knew no bounds other than each individual judge’s notion of an appropriate penalty warranted by a counsel’s misdeeds. The present case is significantly different in that the district court tied its sanction to specific costs that bore a direct relationship to the alleged misconduct and thus offered a nexus and a limit.
Moreover, and perhaps most significantly, the Gamble court’s reasoning runs counter to the Supreme Court’s observations that a court’s broad power to discipline attorneys as officers of the court for misconduct not properly categorized as con*566tempt is substantially different from the contempt power. The Supreme Court has noted, for example, that the “ ‘power to disbar an attorney proceeds upon very different grounds’ from those which support a court’s power to punish for contempt.” Cammer v. United States, 350 U.S. at 408 n. 7, 76 S.Ct. at 460 n. 7 (quoting Ex parte Robinson, 86 U.S. (19 Wall.) 505, 512, 22 L.Ed. 205 (1873)). Similarly, we agree with the Ninth Circuit that a reasonable “monetary sanction for failure to carry out [an attorney’s] special responsibility [to the court] ... differs from the more severe infraction of contempt for which attorneys and members of the general public can become liable.” Miranda, 710 F.2d at 521. The former is based simply on an unjustified failure to discharge an administrative responsibility as an officer of the court.
Relying in part on the Supreme Court’s holding in Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 72 S.Ct. 863, 96 L.Ed. 1153 (1952), that the President was without inherent power to order the Secretary of Commerce to take possession of and actually operate most of the nation’s steel mills, one dissenting opinion argues that the district court’s assertion of authority to impose a sanction of $390 upon an attorney who has abused the judicial process somehow violates the doctrine of separation of powers. The Supreme Court has consistently recognized the federal court’s powers — notwithstanding the absence of Congressional action — to sanction errant attorneys financially both for contempt and for conduct not rising to the level of contempt. See, e.g., Roadway, 447 U.S. at 765, 100 S.Ct. at 2463. Thus the dissent is relegated to arguing that the specific type of monetary sanction imposed here, one “unrelated to the other parties’ costs,” is akin to a criminal “fine” and “a fine can be imposed only to punish conduct that has been prohibited by the legislature.” No authority has been offered for the proposition that financial sanctions by a federal court for abuse of the judicial process by an attorney before that court must be directly related to the other litigants’ costs. Instead the dissent employs as an analog cases forbidding federal courts from creating federal common law crimes or from expanding the reach of criminal statutes. We agree with the Ninth Circuit that the imposition of juror costs when an attorney abuses the judicial process in a manner that forces the unnecessary calling of a jury does not carry any criminal connotation. Miranda, 710 F.2d at 521; see Comment, Financial Penalties, 26 UCLA L.Rev. at 890; see also Dowling, supra, 21 A.B.A.J. at 637.
The impact of the contrary position, moreover, would apparently be to invalidate many of the financial sanctions imposed by this Court and other courts of appeals, cf., e.g., Miranda, 710 F.2d at 521, for myriad violations of court rules, deadlines, or orders, such as the late-filing of a brief, or for conduct unbecoming a member of the bar. Although these financial penalties, payable to the government, are often based to some extent on Fed.R.App.P. 46(c), neither that Rule nor any Congressional enactment explicitly provides for such financial sanctions, nor expressly prohibits the conduct sanctioned by the court — a step Judge Sloviter’s dissent argues is constitutionally mandated.
More fundamentally, we agree with those courts that have found the result in Gamble to be excessively restrictive in practice. While the judicious use of sanctions is a practical necessity in the management of caseloads, Comment, Financial Penalties, 26 UCLA L.Rev. at 875, the more traditional penalties based on a court’s inherent powers, and the specialized sanctions of the Federal Rules of Civil Procedure, are at times inadequate to regulate the wide range of attorney misconduct. Cf. Miller, The Adversary System: Dinosaur or Phoenix, 69 Minn.L.Rev. 1, 25 (1984). For example, the court’s power to dismiss a plaintiff’s action because of the misdeeds of counsel, while unquestioned in principle after Link, must be exercised with great restraint because it subverts the sound public policy of deciding cases on their merits and punishes the client for the counsel’s conduct. E.g., Hritz v. Woma *567Corp., 732 F.2d 1178, 1181 (3d Cir.1984) (“we have repeatedly stated our preference that cases be disposed of on the merits whenever practicable”). The imposition of a modest monetary sanction on counsel is obviously considerably less severe than outright dismissal of an action, and is perhaps more appropriate in that the penalty is directed at the lawyer responsible for the infraction, rather than the litigant who may be completely innocent. Id.; see also Miranda, 710 F.2d at 521; Gamble, 307 F.2d at 734 (Biggs, C.J., dissenting); Miller, supra, 69 Minn.L.Rev. at 25, 27. The dissent, by denying the district courts this flexibility, would seem only to magnify the risk of a harsh and potentially inequitable response to attorney misconduct. The contempt sanction is similarly viewed as a drastic step. Contempt, concerned theoretically with the order, dignity, and decorum of a court, or with willful obstruction of justice, see Comment, Financial Penalties, 26 UCLA L.Rev. at 879, may be inappropriate in many situations of attorney misdeeds. Miranda, 710 F.2d at 521; Comment, Dismissal for Disobedience, 34 U.Chi.L.Rev. at 937; see also Sanctions Imposable, supra note 5, at 180-81 n. 470.
Judge Sloviter questions any reliance on “usefulness” as a factor relevant to the analysis, but we cannot overlook the language of the Supreme Court in Link, 370 U.S. at 630-31, 82 S.Ct. at 1388-89 stating that the “inherent power” to sanction an attorney was “governed not by rule or statute but by the control necessarily vested in courts to manage their own affairs so as to achieve the orderly and expeditious disposition of cases.” If a court’s inherent powers include the ability to do whatever is reasonably necessary to deter abuse of the judicial process, cf. National Hockey League, 427 U.S. at 643, 96 S.Ct. at 2781, courts must be able to impose reasonable sanctions for conduct by lawyers that falls short of contempt of court. Sutter, 543 F.2d at 1037; Renfrew, supra, 67 Calif.L. Rev. at 270; Comment, Dismissal for Disobedience, 34 U.Chi.L.Rev. at 937. A court’s inherent power to manage its caseload, control its docket, and regulate the conduct of attorneys before it, provides authority to fashion tools that aid the court in getting on with the business of deciding cases. A recent study of sanctions in federal courts concluded that “the imposition of financial penalties is the only sanction both mild enough and flexible enough to use in day-to-day enforcement of orderly and expeditious litigation.” Sanctions Imposable, supra note 5, at 86.
Referring to language in Link that the inherent power to dismiss a case was of “ancient origin,” Judge Sloviter suggests that a lack of similar history fatally undermines the sanction used in the present case. It is true that historically most monetary sanctions imposed directly on attorneys, outside of the contempt context, have been related in some way to the expenses unnecessarily incurred by the opposing party. Cf. United States v. Blodgett, 709 F.2d at 611. Nevertheless we do not believe that a long-standing tradition need under-gird every particular sanction imposed pursuant to the district court’s inherent powers. See Note, Power of Federal Courts, 38 Notre Dame Law. at 165 (“courts have exercised vastly similar powers ... for centuries,” and the failure of many “courts to use this particular penalty does not prove the absence of the power to levy it”). The historical discussion in Link constituted a description; it did not establish a prerequisite to the employment of a sanction. Perhaps more telling is Roadway, in which the Supreme Court held that, despite the absence of Congressional authorization, in a proper ease attorneys’ fees may be assessed directly against a lawyer for abuse of the judicial process. The decision also reaffirmed a court’s power to dismiss a case when a plaintiff fails to pursue the litigation diligently, and observed that the power of a court over the members of its bar must be at least as great as its authority over litigants. Roadway, 447 U.S. at 766, 100 S.Ct. at 2464. In upholding the taxing of counsel fees directly against an attorney, the Court was not dissuaded by the fact that this particular sanction did not have a long legal ancestry; in fact it ran *568counter to the “American rule” that each side pay its own fees. Rather, the Supreme Court emphasized the well acknowledged inherent power of a court to levy a reasonable sanction in response to abusive litigation practices. Id. at 765, 100 S.Ct. at 2463. Significantly, the Supreme Court cited Chief Judge Biggs’ dissent in Gamble with approval in its discussion of the inherent power of a court to impose reasonable sanctions upon those admitted to its bar. Id. at 766 n. 12, 100 S.Ct. at 2464 n. 12 (citing Gamble, 307 F.2d at 735-36 (Biggs, C.J., dissenting)).
Roadway also would appear to refute the argument that the mere existence of the cost statutes, §§ 1920 & 1927, prevents reliance by courts on their inherent powers to support the imposition of a reasonable monetary sanction closely tied to the attorney’s misconduct. Although there may be some inherent powers that a court may not exercise in the face of contrary legislation, e.g., Alyeska, 421 U.S. at 259, 95 S.Ct. at 1622, it would be an unwarranted extension of these precedents to argue that through §§ 1920 and 1927 Congress has pre-empted entirely the field of monetary sanctions against errant attorneys. In Roadway, the Court found that nothing in the statutes or their history suggests that Congress intended to preclude the use of a monetary sanction not expressly provided for in § 1920 and § 1927. Indeed, the Court acknowledged that although assessment of attorney’s fees directly on an errant attorney was not a cost enumerated by Congress, the sanction was fully within the court’s inherent powers. Cf. Link, 370 U.S. 626, 82 S.Ct. at 1386 (existence of Federal Rule permitting district court to dismiss case upon motion of party did not supersede court’s inherent power to dismiss sua sponte); Note, Power of Federal Courts, 38 Notre Dame Law. at 169.
The present appeal does not raise one of the precise issues proffered in Gamble, i.e., the propriety of a monetary sanction wholly unrelated to any costs incurred by a litigant, the court, or the government as a result of attorney misconduct. We do not resolve here the narrow question of what degree of nexus between sanction and misconduct is necessary. However, our fresh evaluation of the importance and necessity of some kind of sanction as one of the reasonable and flexible instruments for curbing abuse of the judicial process suggests that Gamble should no longer control this conceptual area. In light of the above described development of the law of sanctions, the persuasive reasoning of commentators, the combined wisdom of several other courts of appeals, and the fact that the Supreme Court significantly undercut Gamble in Roadway, we now most respectfully overrule Gamble.
We do not imply, in upholding the power of the district court to impose a sanction on an attorney in a proper case, that the imposition of a sanction was proper in the present case, see infra Part YI, nor do we express any view on what forms of attorney misconduct would justify the imposition of the sanction. Certainly, however, the district court must exercise discretion and sound judgment in dealing with the myriad methods with which lawyers may abuse the judicial process. Nor is anything in this opinion meant to suggest that settlement on the eve of trial is in and of itself improper. Frequently a settlement may be in the best interest of not only clients and their attorneys, but of the judicial system and society as a whole. The suggestion in the present proceeding, however, is that because of misconduct by an attorney a settlement took place at an unjustifiably late date thereby occasioning the waste of scarce judicial resources and the unnecessary expense of calling a jury.
V
The district court in the case at hand did not appear to base its sanction upon any local rule. The Supreme Court has stated, however, that a court’s inherent authority over members of its bar is not limited to those specific facets of conduct covered by a local rule. In upholding a court’s inherent power to dismiss a lawsuit, the Supreme Court in Link held:
*569Petitioner’s contention that the District Court could not act in the conceded absence of any local rule covering the situation here is obviously unsound. Federal Rule of Civil Procedure 83 expressly provides that “in all cases not provided for by rule, the district courts may regulate their practice in any manner not inconsistent with these rules.”
Link, 370 U.S. at 633 n. 8, 82 S.Ct. at 1390 n. 8, see also Sutter, 543 F.2d at 1037. While we believe that the district court in imposing the sanction at issue here need not have relied on a local rule, we nevertheless suggest that if the district courts believe that there are occasions in which the imposition of such a sanction would be just, wise, and efficacious, a local rule on the imposition of such a sanction might well be salutary.
The rulemaking power of the district courts is now codified at 28 U.S.C. § 2071 (1982), which provides that the district courts may make rules prescribing the conduct of court business. The only statutory requirement is that the local rules promulgated be consistent with acts of Congress and the rules prescribed by the Supreme Court. The Federal Rules of Civil Procedure permit the district courts to make and amend rules governing their practice not at variance with the other Federal Rules. We agree with the Second, Ninth, and Tenth Circuits that the district courts have the power, absent a statute or rule promulgated by the Supreme Court to the contrary, to make local rules that impose reasonable sanctions where an attorney conducts himself in a manner unbecoming a member of the bar, fails to comply with any rule of court, including local rules, or takes actions in bad faith. See Miranda, 710 F.2d at 521-22; Martinez, 593 F.2d at 994; Sutter, 543 F.2d at 1037-38; see also Sanctions Imposable, supra note 5, at 265; Renfrew, supra, 67 Calif.L. Rev. at 270; Comment, Sanctions Imposed by Courts, 44 U.Chi.L.Rev. at 635; accord Fed.R.App.P. 46(b) (conduct unbecoming a member of the bar).
The Gamble plurality, in addition to finding the sanction imposed there to be unauthorized by inherent power, thought the “fine” assessed to be a “basic disciplinary innovation” and thus beyond the scope of the “local rulemaking power.” 307 F.2d at 732. It is not clear, however, that the use of a reasonable, albeit nontraditional, sanction closely tied to the misconduct of the attorney fairly can be characterized as a “basic innovation.” This is true especially given the universally recognized need to deter abuse of the judicial process. The Gamble court apparently was relying on language in Miner v. Atlass, 363 U.S. 641, 80 S.Ct. 1300, 4 L.Ed.2d 1462 (1960), which held that district courts could not introduce discovery into admiralty proceedings pursuant to local rule. Discovery was termed “one of the most significant innovations” of the Federal Rules of Civil Procedure, id. at 649, 80 S.Ct. at 1305, and thus a local rule governing discovery in the admiralty context was regarded at that time as improper. The Supreme Court suggested, however, that a local rule might be more appropriate in situations involving “the necessary choice of a rule to deal with a problem which must have an answer, but need not have any particular one.” Id.
More recently the Supreme Court has clarified the role of local rules in achieving procedural change. In Colgrove v. Battin, 413 U.S. 149, 93 S.Ct. 2448, 37 L.Ed.2d 522 (1973), it upheld the validity of a local rule providing for a jury of six in a civil trial. The Court declared that the “requirement of a six-member jury is not a ‘basic procedural innovation.’ ” The Court went on to define the “ ‘basic procedural innovations’ to which Miner referred” as “those aspects of the litigatory process which bear upon the ultimate outcome of the litigation.” Id. at 164 n. 23, 93 S.Ct. at 2456 n. 23. A reasonable monetary sanction on an errant attorney is not a procedural innovation beyond the reach of a local rule since it is not outcome-determinative in the sense suggested by the Supreme Court.
Regarding the wisdom of a local rule on monetary sanctions, Gamble suggested that a national “uniform approach” was *570required. However, given that the district courts vary tremendously in size, volume of cases, calendar congestion, and types of cases, and that litigation tactics of attorneys may differ across the country, it is not readily apparent that a national rule is either required or desirable. Cf. Comment, Financial Penalties, 26 UCLA L.Rev. at 875. In the context of judicial disciplinary rules a recent commentator has noted that in “some instances the potential benefits of continuing experimentation are so obvious, and the costs arising from disuniformity so speculative, that requiring a uniform rule would be premature.” Burbank, Procedural Rulemaking Under the Judicial Councils Reform and Judicial Conduct and Disability Act of 1980, 131 U.Pa.L. Rev. 283, 326 (1982). Others have observed that desirable differences in local rules stem in part from varying “local expectations and practice.” Flanders, Local Rules in Federal District Courts: Usurpation, Legislation, or Information, 14 Loyola L.A.L.Rev. 213, 264 (1981).
Several courts have promulgated rules that provide for an assessment of juror and related costs, when a settlement occurs improperly shortly before trial.11 “Such local rules occupy a vital role in the district courts’ efforts to manage themselves and their dockets,” Flanders, supra, 14 Loyola L.A.L.Rev. at 263, and are essential tools in implementing court policy. Id. at 218. The local rule device fulfills important informational purposes, placing the bar on notice of a court’s policies. Id. at 263. Similarly, a local rule may well be the most effective means of ensuring that all members of the bar are aware that a particular practice is deemed improper, and thus subject to a sanction. Local rules may also alert rulemakers to the need for changes in national rules and supply an empirical basis for making such changes. Furthermore, a local rule may be a powerful implement for rationalizing diverse court practices and imposing uniformity within a given district. Id. at 268, 269.
VI
Finally, appellants assert that the imposition of a monetary sanction by the district court without affording the attorney prior notice and an opportunity to be heard violates due process. We agree.
In the absence of extraordinary circumstances, procedural due process requires notice and an opportunity to be heard before any governmental deprivation of a property interest. Boddie v. Connecticut, 401 U.S. 371, 379, 91 S.Ct. 780, 786, 28 L.Ed.2d 113 (1971). The form which those procedural protections must take is determined by an evaluation of all the circumstances and an accommodation of competing interests. See Goss v. Lopez, 419 U.S. 565, 579, 95 S.Ct. 729, 738, 42 L.Ed.2d 725 (1975); Renfrew, supra, 67 Calif.L.Rev. at 281. The individual’s right to fairness and accuracy must be respected, as must the court’s need to act quickly and decisively.
In considering the imposition of a penalty upon attorneys, we note that the Court has cautioned that like “other sanctions, attorney’s fees certainly should not be assessed lightly or without fair notice and an opportunity for a hearing on the record.” Roadway, 447 U.S. at 767, 100 S.Ct. at 2464. Similarly, a court may not disbar an attorney without notice and a hearing. Ex parte Bradley, 74 U.S. (7 Wall.) 364, 372-74, 19 L.Ed. 214 (1869). Courts of appeals may not impose disciplinary sanctions on attorneys until “after reasonable notice and an opportunity to show cause to the contrary, and after hearing, if requested.” Fed.R.App.Pro. 46(c). Although in Link the Supreme Court suggested that not every order entered without a preliminary adversary hearing offends due process. Link, 370 U.S. at 632, 82 S.Ct. at 1389, we believe that as a general practice a monetary detriment should not be imposed by a court without prior notice and some occasion to respond.
*571These procedural safeguards will ensure that the attorney has an adequate opportunity to explain the conduct deemed deficient. For example, in the present case, the attorney by affidavits disputes the factual predicate upon which the order was based. Furthermore, such procedures will afford the judge adequate time to evaluate the propriety of the particular sanction in light of the offending attorney’s explanation as well as to consider alternatives. See Hritz v. Woma Corp., 732 F.2d at 1182; Edgar v. Slaughter, 548 F.2d 770, 773 (8th Cir.1977); see also Miranda, 710 F.2d at 522-23. Moreover, by providing a record, a hearing will facilitate appellate review. Miranda, 710 F.2d at 522-23; Renfrew, supra, 67 Calif.L.Rev. at 281. In some cases, it may be that the record developed at the time of the alleged misconduct will, itself, satisfy this need as long as the attorney has been afforded an opportunity to adduce the relevant facts. Upon imposing such a sanction it would seem appropriate for the district court to make adequate written findings. Renfrew, supra, 67 Calif.L.Rev. at 281; cf. Wilson v. Volkswagon of America, Inc., 561 F.2d 494, 516 (4th Cir.1977), cert. denied, 434 U.S. 1020, 98 S.Ct. 744, 54 L.Ed.2d 768 (1978) (appellate review difficult without adequate findings).12
A final point worth noting, although not raised in the briefs, is that the due process calculus may also be affected by the “knowledge which the circumstances show [the offending] party may be taken to have of the consequences of his own conduct.” Link, 370 U.S. at 632, 82 S.Ct. at 1390. Thus, fundamental fairness may require some measure of prior notice to an attorney that the conduct that he or she contemplates undertaking is subject to discipline or sanction by a court. Consequently the absence, for example, of a statute, Federal Rule, ethical canon, local rule or custom, court order, or, perhaps most pertinent to the case at hand, court admonition, proscribing the act for which a sanction is imposed in a given case may raise questions as to the sanction’s validity in a particular case.
As noted in Part V of this opinion there does not appear to be a local rule covering the conduct of the attorney in question. Nor is it apparent that any analogous form of actual or constructive notice was given to the attorney whose settlement conduct was deemed sanctionable by the district court, although there is no way of ascertaining that fact from the record before us. Therefore, a remand of the matter for consideration of this as well as the previously discussed due process issues is required.
VII
The order of the district court will be vacated and the case remanded for action consistent with this opinion.
APPENDIX
The following is the text of the order challenged in this appeal:
ORDER
The above captioned case was set for trial on Monday, May 23, 1983. Plaintiff’s attorney made repeated attempts to contact defendant’s attorney in regards to settlement possibilities during the week of May 16, 1983. Receiving no response, plaintiff prepared for trial and appeared on May 23rd, ready to present his case.
At this point, defendant’s attorney made a proposal of settlement which was accepted by plaintiff. The defendant's attorney was scheduled for trial in state court the same day. The settlement avoided a possibly unfortunate conflict.
The settlement on the eve of trial was not justified. Defendant’s attorney was given adequate notice by plaintiff’s counsel and by Court personnel to attempt to reach an agreement.
*572IT IS HEREBY ORDERED that William Tighe, Esquire, is directed to pay to the Clerk of Court for the Western District of Pennsylvania the sum of $390.00, computed as follows: $30.00 (per diem fee for jurors) times 13, the minimum number of jurors necessary to select a jury of seven persons, one being an alternate, to be selected by lot. Said amount is payable by September 2, 1983.
. The court appointed Robert J. Bartow, Professor, Temple University School of Law, to serve as amicus curiae and to assert a position, in the interest of informed decision-making, adverse to that of appellants.
. Taxable costs under § 1920 include: (1) fees of the clerk and marshal; (2) fees of the court reporter for all or any part of the stenographer’s transcript necessarily obtained for use in the case; (3) fees and disbursements for printing and witnesses; (4) fees for exemplification and copies of papers necessarily obtained for use in the case; (5) docket fees under § 1923 of Title 28; (6) compensation of court-appointed experts, compensation of interpreters, and salaries, fees, expenses and costs of special interpretation services under § 1828 of Title 28.
. Congress amended § 1927 in 1980 to include the taxing of attorney’s fees.
. Thus, we need not address whether an attorney’s delay of settlement may constitute multiplying the proceedings and what mental state is required by the terms "unreasonably" and "vexatiously.’’
. See R. Rodes, K. Ripple & C. Mooney, Sanctions Imposable for Violations of the Federal Rules of Civil Procedure 179 n. 466 (Federal Judicial Center 1981) [hereinafter Sanctions Imposable ].
. There is a plethora of state court cases regarding the powers of the state judiciary.
. Levin & Amsterdam, supra, 107 U.Pa.L.Rev. at 33, note that courts have relied on this concept to void legislation requiring a written opinion in every case, Vaughan v. Harp, 49 Ark. 160, 4 S.W. 751 (1887); declaring within what time every case must be heard, Atchison, Topeka & Santa Fe Railway Co. v. Long, 122 Okla. 86, 251 P. 486 (1926); denying a court the power to issue its mandate until a prescribed period of time after the judgment has elapsed even if this renders the judgment meaningless, Burton v. Mayer, 274 Ky. 263, 118 S.W.2d 547 (1938); or providing for the automatic disqualification of judges simply upon the application of a party, State ex rel. Bushman v. Vandenberg, 203 Ore. 326, 329, 280 P.2d 344 (1955). We do not pass on the validity of these holdings but only note that courts have sought to carve out a sphere of minimal autonomy based upon separation of powers notions.
. Although now codified at 18 U.S.C. § 401 (1982) and in Fed.R.Crim.P. 42, the contempt power is rooted principally in the inherent power of the judiciary. See Levine v. United States, 362 U.S. 610, 615, 80 S.Ct. 1038, 1042, 4 L.Ed.2d 989 (1960); Ex parte Robinson, 86 U.S. (19 Wall.) at 510, 22 L.Ed. 205; Sanctions Imposable, supra note 5, at 74; Comment, Financial Penalties Imposed Directly Against Attorneys in Litigation Without Resort to the Contempt Power, 26 UCLA L.Rev. 855, 878 (1979).
. The long-standing practice of the federal courts of making such appointment has been codified in Fed.R.Civ.P. 53. See Reed v. Cleveland Board of Education, 607 F.2d 737, 743 n. 1 (6th Cir.1979).
. Traditionally courts have treated the attorney as the client’s agent so that the attorney's acts and omissions legally bind the client. Sanctions Imposable, supra note 5, at 70; see generally Link, 370 U.S. at 633-34, 82 S.Ct. at 1390-91. Because this approach ignores the realities of the lawyer-client relationship, see Comment, Involuntary Dismissal for Disobedience or Delay: The Plaintiff's Plight, 34 U.Chi.L.Rev. 922, 929 (1967), there has been a trend to seek to impose penalties upon only the offending lawyer.
. See, e.g., D.Col.R. 11; D.Del.R. 5.5(D); N.D. Ill.Civ.R. 11; W.D.La.R. 5(g); D.N.J.Gen.R. 20(G); D.N.M.R. 13(e); W.D.Tenn.R. 6(b); E.D. Va.R. 20(e).
. To permit the district court to impose monetary sanctions for attorney misconduct without affording notice and a hearing, as is required in the context of contempt, might tend to undercut the contempt sanction's procedural safeguards. See Fcd.R.Crim.P. 42(b).