UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT
filed: August 16, 1989; As Corrected, October 6, 1989.
SAMUEL B. MCCARTER, FIRST SENECA BANK & TRUST CO., AND ALLAN LEVINE, CO-EXECUTORS OF THE ESTATE OF DR. LEO LEVINE AND OLIVE M. HECK, APPELLANTS
MERLE B. MITCHAM, BUTCHER & SINGER, INC., AND THOMAS GABRESKI, APPELLEES
On Appeal From the United States District Court For the Western District of Pennsylvania (D.C. Civ. No. 87-34 Erie).
Opinion OF THE COURT
BECKER, Circuit Judge.
This appeal from a judgment of the district court, dismissing on grounds of res judicata and the statute of limitations, a securities fraud suit, brought under the civil RICO statute and federal securities law, presents three interesting questions. First, we must determine whether a dismissal for failure to file a complaint within the time limit set by the state court constitutes a sanction and therefore a decision on the merits, in which case it is a basis for claim preclusion, or a dismissal for failure to prosecute in which case it is not. Second, we must decide whether exclusive federal jurisdiction or concurrent jurisdiction exists over RICO claims. Under Pennsylvania law, if federal jurisdiction is exclusive, the state decision cannot act as a bar to litigation in the federal court. Third, we must consider whether this Court's decision in In re Data Access Systems Securities Litigation, 843 F.2d 1537 (3d Cir. 1988) (in banc), cert. denied, 488 U.S. 849, 109 S. Ct. 131, 102 L. Ed. 2d 103 (1988), which set a new standard for the determination of statutes of limitations for federal securities claims, should be applied retroactively.
For the reasons that follow, we hold that the dismissal was a sanction, that civil RICO jurisdiction is concurrent, and that Data Access should be applied retroactively given these facts. We therefore affirm.
I. FACTS AND PROCEDURAL HISTORY
On November 17, 1983, plaintiffs Samuel B. McCarter, First Seneca Bank & Trust Co., as executor for Dr. Leo Levine, and Olive M. Heck commenced an action by praecipe for writ of summons in the Court of Common Pleas of Warren County (Pennsylvania) against defendants Merle M. Mitcham, Butcher & Singer, Inc., who were their stockbrokers, and Thomas Gabreski, who was a stock purchaser. The action was intended to recover losses from an alleged scheme to defraud in a stock deal in which plaintiffs claim to have been induced to sell securities at a price below fair market value. On August 16, 1984, approximately nine months following the filing of their state court action, the prothonotary sent notice to the plaintiffs pursuant to local (Warren County) Rule L 307, informing them that their suit would be dismissed for a failure to prosecute if it was not placed on the trial list within 240 days after commencement of the suit.*fn1 The court subsequently ordered submission of a complaint by December 16, 1984, and discovery to be completed by January 15, 1985.
On January 21, 1985, after no complaint was filed by the date set by the court, defendants moved to dismiss for failure to file the complaint. On January 25, 1985, ten days after the discovery deadline, plaintiffs moved for an extension of time on the ground that their attorney was seriously ill. On April 25, 1985, plaintiffs filed a complaint, asserting claims under the Pennsylvania Securities Act of 1972, Pa.Stat.Ann. tit. 70, §§ 1-101 to 1-704 (Purdon Supp. 1988), and state common law. See App. at 66-90.
On June 6, 1985, defendants filed preliminary objections alleging that the complaint was not timely filed. On September 24, 1985, the Court of Common Pleas ruled that the plaintiffs' attorney's illness was insufficient to justify the delay in filing the complaint and dismissed the action with prejudice. No. A.D. 540 of 1983 (C.P. Warren Sept. 24, 1985). On December 15, 1986, the Superior Court of Pennsylvania affirmed. McCarter v. Mitcham, 362 Pa.Super. 644, 520 A.2d 1220 (1986).
On February 17, 1987, plaintiffs filed a complaint in the district court for the Western District of Pennsylvania asserting essentially identical claims, but pleading them as claims arising under the Racketeer Influenced and Corrupt Organizations Act, 18 U.S.C. §§ 1961-1968 (1982); section 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b) (1982) (as well as claims under Rule 10b-5(a); 17 C.F.R. § 240.10b-5 (1988)); and state law (as pendent claims). See App. at 4-26. On March 12, 1987, defendants moved to dismiss, Fed.R.Civ.P. 12 (b) (6), and/or for summary judgment, Fed.R.Civ.P. 56, asserting that all of plaintiffs' claims were barred by res judicata, that the section 10(b) claims were barred by the statute of limitations, and that the civil RICO claim failed to state a claim for relief. On May 6, 1987, the district court granted defendants' motion to dismiss on the grounds that all of the claims were barred by res judicata and that the section 10b claims had not been filed within the applicable statute of limitations. See McCarter v. Mitcham, 693 F. Supp. 349, 350 (W.D.Pa. 1987).
On May 15, 1987, plaintiffs moved for reconsideration. On July 9, 1987, the district court stayed its action pending resolution of the appeal of the state court action by the Pennsylvania Supreme Court, but vacated the stay on April 22, 1988, after that court dismissed the appeal as improvidently granted. McCarter v. Mitcham, 517 Pa. 486, 538 A.2d 1335 (1988). On August 31, 1988, the district court denied the motion for reconsideration and reaffirmed its prior decision to dismiss, holding that the section 10(b) claims were time-barred and that all other claims were barred by Pennsylvania res judicata law. 693 F. Supp. at 351-53. Appellants appeal this ruling.*fn2
II. RES JUDICATA
A. Pennsylvania Claim Preclusion Law
Under 28 U.S.C. § 1738 (1982), a state court's judgment must be given the same effect in federal court that it would have been given in state court. Marrese v. American Academy of Orthopaedic Surgeons, 470 U.S. 373, 84 L. Ed. 2d 274, 105 S. Ct. 1327 (1985); Kremer v. Chemical Construction Co., 456 U.S. 461, 72 L. Ed. 2d 262, 102 S. Ct. 1883 (1982); Nanavati v. Burdette Tomlin Memorial Hospital, 857 F.2d 96 (3d Cir. 1988), cert. denied, 489 U.S. 1078, 109 S. Ct. 1528, 103 L. Ed. 2d 834 (1989). The Pennsylvania law of res judicata applies a four-part test to determine whether a claim is barred:
The application of res judicata requires a concurrence of four (4) conditions: (1) an identity of the thing sued upon; (2) an identity of the cause of action; (3) an identity of the persons and parties to the action; and (4) an identity of the quality or capacity of the parties suing or sued.
Dunham v. Temple University, 288 Pa.Super. 522, 534, 432 A.2d 993, 999 (1981).
For relitigation to be precluded, Pennsylvania law requires that the prior determination be "on the merits." See Ross v. Bowlby, 353 Pa.Super. 59, 509 A.2d 332 (1986) (disposition of case not on the merits does not bar relitigation); Consolidation Coal Co. v. Dist. 5, UMW, 336 Pa.Super. 354, 485 A.2d 1118 (1984) (res judicata applies where prior action was decided on the merits). Where the prior dismissal was based on a judgment of non pros, Pennsylvania courts will not give preclusive effect to the judgment. As the Pennsylvania Superior Court has stated:
the legal effect of the entry of a judgment of non pros is not such as to preclude a plaintiff who suffers such a judgment from instituting another suit on the same cause of action provided, however, that the second suit is brought within the period of the statute of limitations [and the plaintiff is willing to pay for costs of the former suit].
Haefner v. Sprague, 343 Pa.Super. 342, 347, 494 A.2d 1115, 1118 (1985).*fn3
Although neither the Superior Court nor the Supreme Court of Pennsylvania have spoken to the issue, Pennsylvania claim preclusion law also appears to require that the original court have subject matter jurisdiction over the cause of action in order for that cause of action to be later barred in a second lawsuit in a court with subject matter jurisdiction over the cause of action. Pennsylvania lower courts have so held. See, e.g., Philadelphia v. Stradford Arms, Inc., 1 Pa.Commw. 190, 274 A.2d 277 (1971) (claim preclusion will only apply when first tribunal has equivalent subject matter jurisdiction as second) Furthermore, courts of other states have nearly universally required subject matter jurisdiction in the first action as a prerequisite to barring the second action. See Nanavati, 857 F.2d at 111.
B. Was the State Court Dismissal on the Merits?
The first major issue in this case is whether the state court dismissal with prejudice was a decision "on the merits" as that term is applied in Pennsylvania jurisprudence or whether instead it is analogous to a dismissal for failure to prosecute. As discussed above, an entry of judgment of non pros pursuant to the local rules for mere inactivity for an extended period of time would not bar relitigation under Pennsylvania preclusion law.
Applying Pennsylvania law, this Court has drawn a distinction between dismissal as a result of the simple failure to prosecute and dismissal as a sanction for failure to obey a court order. Dismissals of the latter type most likely have preclusive effect. See Wade v. City of Pittsburgh, 765 F.2d 405, 408-09 (3d Cir. 1985). This interpretation is suggested by the Superior Court in Dunham, which seems to say that, where a party fails to follow court orders and as a result its case is dismissed, res judicata applies even though there was no judicial decision finally addressing and adjudicating the merits of the case. See Dunham, 288 Pa.Super. at 533, 432 A.2d at 999. Moreover, the Superior Court in Consolidation Coal stated that where the prior action was dismissed with prejudice, further litigation may be barred even though there was, literally speaking, no decision "on the merits." See Consolidation Coal, 336 Pa.Super. at 365, 485 A.2d at 1123. Thus, while no Pennsylvania court has squarely held that failure to follow a court order is a decision on the merits, it appears that the Pennsylvania courts interpret the phrase "on the merits" expansively, and would prohibit relitigation in some cases even when the original court has not actually entered a final decision on the merits of the case as such.
In this case, the state court, rather than permitting the prothonotary to dismiss the case automatically for failure to prosecute pursuant to Rule L 307, dismissed the case with prejudice following a failure by the plaintiffs to file the complaint within the extended time limits set by the court. The dismissal was not entered automatically by the prothonotary upon failure to file the complaint, but rather was issued by the court with a well-considered opinion and order dismissing the case with prejudice. The court stated in its opinion that "defendant's motion is not for procedural non prosequitur." Com Pleas Op. at 5. We believe that, under these circumstances, the court's dismissal was not for failure to prosecute, but rather as a sanction for the untoward delay in filing of the complaint.
We conclude therefore that the order to dismiss with prejudice was an order "on the merits" within the meaning set forth by the Pennsylvania courts. Hence, the prior litigation can be preclusive of any further litigation as long as the four prongs of the Pennsylvania res judicata inquiry are met and the Pennsylvania court had subject matter jurisdiction.
C. Are the Civil RICO Claims Barred by the Pennsylvania Law of Res Judicata?
As explained above, Pennsylvania courts would most likely not apply claim preclusion where the original court lacked subject matter jurisdiction Plaintiffs argue that there is exclusive federal jurisdiction over civil RICO claims and therefore that their civil RICO claims cannot be barred. We conclude, however, that the civil RICO statute does not confer exclusive federal jurisdiction, but rather concurrent state and federal jurisdiction
It is not clear whether, in the context of determining the state's law of preclusion, the jurisdictional reach of a federal statute is established by the state or federal court determination of the issue. Yet, we need not resolve this issue in this case, because we reach the same conclusion as that reached by the highest court of Pennsylvania to address the issue, Village at Camelback v. Carr, 371 Pa.Super. 452, 538 A.2d 528 (1988), app. granted, 519 Pa. 668, 548 A.2d 257 (1988): the civil RICO statute does not create exclusive federal jurisdiction.*fn4
The courts of appeals are currently split on the question whether the civil RICO statute provides for exclusive federal jurisdiction. The Fourth and Ninth Circuits have held that there is concurrent jurisdiction. See Brandenburg v. Seidel, 859 F.2d 1179, 1193-95 (4th Cir. 1988); Lou v. Belzberg, 834 F.2d 730, 738-39 (9th Cir. 1987), cert. denied, 485 U.S. 993, 108 S. Ct. 1302, 99 L. Ed. 2d 512 (1988). The Fifth and Seventh Circuits have expressed doubt that jurisdiction over civil RICO claims is exclusively federal. See Crotty v. City of Chicago Heights, 857 F.2d 1170, 1172 n.6 (7th Cir. 1988); DuBroff v. DuBroff, 833 F.2d 557, 562 (5th Cir. 1987); County of Cook v. Midcon Corp., 773 F.2d 892, 905 n. 4 (7th Cir. 1985). Only the Sixth Circuit has held that there is exclusive federal jurisdiction under civil RICO. See Chivas Products Ltd. v. Owen, 864 F.2d 1280, 1286 (6th Cir. 1988). The Supreme Court has now granted certiorari to resolve this conflict. See Tafflin v. Levitt, 865 F.2d 595 (4th Cir. 1989) (concurrent jurisdiction over civil RICO claims), cert. granted, 490 U.S. 1089, 109 S. Ct. 2428, 104 L. Ed. 2d 985, 57 U.S.L.W. 3779 (U.S. 1989).
In view of the grant of certiorari, there would be nothing gained by an extensive discussion of this issue. Suffice it to say then that we are persuaded by Judge Krupansky's dissent in Chivas as well as Judge Beck's opinion in Village at Camelback and by those opinions of the courts of appeals that have determined that civil RICO jurisdiction is concurrent, Brandenburg and Belzberg. We hold therefore that there is not exclusive federal jurisdiction over civil RICO claims, but rather concurrent jurisdiction.
There can be no question that the four-part test, as set out in Dunham, see supra Slip Op. at 6, for determining whether a claim is barred is met here with respect to the RICO claims. Between the federal complaint and the state complaint, there is a complete identity of the thing sued upon, the cause of action, the persons and parties to the action, and the quality or capacity of the parties suing or being sued. We conclude that the Pennsylvania courts would have had jurisdiction over the civil RICO claims and therefore that plaintiffs' civil RICO claims are precluded by the Pennsylvania law of claim preclusion.*fn5
D. Are the Securities Exchange Act of 1934 Claims Barred by the Pennsylvania Law of Res Judicata?
As explained above, the Pennsylvania law of res judicata would not bar relitigation where the state court did not have jurisdiction over the claim. Plaintiffs assert, and Mitcham does not argue to the contrary, that the section 10(b) and Rule 10b-5 Securities Exchange Act of 1934 claims are subject to exclusive federal jurisdiction.
Indeed, the Securities Exchange Act expressly provides for exclusive federal jurisdiction. See 15 U.S.C. § 78aa (1982 & West Supp. 1989), and hence we must conclude that the Pennsylvania state courts would not have had jurisdiction over these claims. Thus, under Pennsylvania's law of claim preclusion, the federal securities law claims are not barred. We must consider, therefore, whether the Securities Exchange Act of 1934 claims are barred by the statute of limitations.
III. STATUTE OF LIMITATIONS FOR THE SECURITIES EXCHANGE ACT OF 1934 CLAIMS
The sales of stock in question occurred between March 1981 and July 1981. There is no dispute that the claim accrued in 1981. The federal complaint was filed on February 17, 1987, over 5 1/2, years later. The district court held that the Securities Exchange Act of 1934 claims are barred by the applicable statute of limitations. We agree.
On April 8, 1988, we decided In re Data Access Systems Securities Litigation, 843 F.2d 1537 (3d Cir. 1988) (in banc), holding that the "the proper period of limitations for a complaint charging violation of Section 10(b) and Rule 10b-5 is one year after the plaintiff discovers the facts constituting the violation, and in no event more than three years after such violation." Data Access, 843 F.2d at 1550. Data Access overruled our three prior cases on this point. See Sharp v. Coopers & Lybrand, 649 F.2d 175 (3d Cir. 1981), cert. denied, 455 U.S. 938, 71 L. Ed. 2d 648, 102 S. Ct. 1427 (1982); Biggans v. Bache Halsey Stuart Shields, Inc., 638 F.2d 605 (3d Cir. 1980); Roberts v. Magnetic Metals Co., 611 F.2d 450 (3d Cir. 1979). Under the prior overruled cases, plaintiffs' federal complaint, which followed by 5 1/2, years the accrual of the action was arguably filed in time, See A.J. Cunningham Packing v. Congress Financial Corp., 792 F.2d 330, 332-37 (3d Cir. 1986) (common law fraud claims accruing before February 1983 governed by six-year statute of limitations under Pennsylvania law). However, under Data Access, it was plainly filed out of time. We must decide whether Data Access is to be applied prospectively only. Data Access itself does not answer this question.
In general, an appellate court applies the law as it exists when the case is before it, i.e., retroactively, even though "that law may differ from the one in force . . . at the time the events sparking the litigation occurred." See Hill v. Equitable Trust Co., 851 F.2d 691, 695 (3d Cir. 1988), cert. denied, 488 U.S. 1008, 109 S. Ct. 791, 102 L. Ed. 2d 782 (1989). The Supreme Court has set out three criteria by which to decide whether a newly adopted rule should be applied prospectively only:
(1) The holding must establish a new principle of law, either by overruling clear past precedent on which litigants may have relied, or by deciding an issue of first impression whose resolution was not clearly foreshadowed;
(2) The merits and demerits in each case must be weighed by looking to the history of the rule in dispute, its purpose and effect, and whether retrospective operation will further or retard the rule's operation;
(3) Retrospective application must create the risk of producing substantially inequitable results.
Hill, 851 F.2d at 696 (paraphrasing Chevron, 404 U.S. at 106-07). The district court determined that these criteria were not met and therefore applied the shorter, Data Access, statute of limitations, holding that the federal securities claims were barred by the statute of limitations. In order to apply these criteria, we must summarize the prior precedent and the parties' c ontentions with respect thereto.
A. The Prior Precedent
Before April 8, 1988, the statute of limitations for Rule 10b-5 cases in this Court was determined by borrowing the statute of limitations from the most analogous state cause of action. Plaintiffs argue that the most analogous referent is not the Pennsylvania securities law, but common law fraud, which is governed by a six-year statute of limitations. See A. J. Cunningham Packing, 792 F.2d at 330 (common law fraud claims accruing before February 1983 governed by six-year statute of limitations under Pennsylvania law). They submit that the three overruled cases, see supra Slip Op. at 12, established that 10b-5 actions must be analogized to common law fraud and therefore that a six-year statute of limitations then applied.
Roberts addressed the issue of the applicable statute of limitations under New Jersey, not Pennsylvania, law, but it is instructive here in that it set out the protocols for addressing the problem of what statute of limitations to apply. In Roberts, the panel split over the appropriate approach to take in determining the applicable statute of limitations where a stockholder sued a corporation and others for alleged securities violations. Judge Gibbons, writing the opinion of the court, approached the issue by attempting to determine a "state policy of repose." 611 F.2d at 456. Judge Sloviter, concurring on the issue of what approach should be taken to the problem, stated that "the selection of the one most appropriate to use must be made by considerations which comport with federal policy. The state policy of repose is relevant only if, and to the extent to which, it is consistent with the underlying federal claim." Id. at 458. Judge Seitz, dissenting on the result, agreed in principle with Judge Sloviter's approach. Id. at 460. The majority held that because the New Jersey securities act did not provide for a cause of action by anyone other than a buyer of securities, New Jersey's common law remedies for breach of fiduciary duty and fraud applied. See id. at 453-54.
In Biggans, Judge Sloviter expressly recognized that her approach differed from that of Judge Gibbons in Roberts, but assumed that the same result would be attained in Biggans regardless of which approach was implemented. 638 F.2d at 608. In Biggans, this Court held that where the defendant was a broker and engaged in churning, the limitations period under the Pennsylvania law of securities was inapplicable because that statute only provides for causes of action against buyers and sellers. 638 F.2d at 610. Thus, the limitations period under the common law of fraud was analogous. In Sharp, we did not address the proper approach to take, but summarily followed Biggans and held that the Pennsylvania law of securities did not provide for causes of action except between buyers and sellers and therefore the cause of action in that case against an accounting firm was governed by the common law of fraud. 649 F.2d at 191-92.
B. Was There Clear Past Precedent?
Plaintiffs argue that there was clear precedent as to the statute of limitations to be applied. At the earliest time the plaintiffs could have discovered the fraud in this case, Roberts, Biggans, and Sharp had been decided. Plaintiffs argue that those cases stand for the proposition that "if the underlying state [securities] law does not afford a civil damage action to remedy the behavior challenged by the 10b-5 claim and the plaintiff would be relegated to a common law fraud action for state relief, the courts must apply the fraud limitations provision to the 10b-5 action." Hill v. Der, 521 F. Supp. 1370, 1383 (D.Del. 1981). While this broad statement of the law may be correct, it does not answer the particular question before us in this case -- whether there is clear past precedent. Upon scrutiny, we do not find that the applicable precedent would clearly render the common law fraud limitation period applicable to the facts of this case.
In Hill v. Equitable Trust, supra, we held that Data Access should be applied retroactively because there was not clear precedent on which plaintiffs could have relied in delaying the bringing of a Rule 10b-5 suit. Hill, 851 F.2d at 697. The court concluded that, as of the time of the filing of the complaint in that case, September 17, 1981, "the law in the circuit on this issue was in ferment" and that the extant decisions were essentially fact-based, making it difficult to predict what limitations period would apply to a particular case. Id. Similarly, the court in Data Access stated in 1988 that "our present case law calls for difficult interpretations of state limitation periods" requiring examination of each contention of a federal securities claim with "great particularity" and noted that district courts have found Third Circuit precedent to be of limited guidance. 843 F.2d at 1541. Generally speaking, then, the law with respect to 10b-5 was unsettled as late as 1988.
However, we must decide more particularly whether the facts of this case have been clearly decided in prior cases. In this case, we have plaintiff-sellers, who were allegedly defrauded through the sale of securities to defendant-buyers through a defendant-broker. Thus, we must determine whether there was clear past precedent with respect to the applicable statute of limitations period for cases brought against buyers and brokers.
In Biggans and Sharp, this Court, interpreting Pennsylvania state law, held, respectively, that the Pennsylvania Securities Act does not apply to brokers involved in churning or to accounting firms that had prepared opinion letters regarding the securities. Although we stated in dicta that the Pennsylvania Securities Act provides a cause of action only against buyers and sellers, we did not address in those cases the scope of the application of the Pennsylvania Securities Act to dealings between buyers and sellers. Neither did we consider whether brokers in the role of aiders and abettors would be liable under the Act. Thus, no case in the Third Circuit provided clear past precedent for the facts of this case.
Moreover, in 1982 the Superior Court of Pennsylvania called into question the assumption in Biggans (1980) and Sharp (1981) that the Pennsylvania Securities Act only provides causes of action against buyers and sellers, see Brennan v. Reed, Smith, Shaw & McClay, 304 Pa.Super. 399, 450 A.2d 740 (1982), making the relevant precedent even less clear. In Brennan, the court was faced, inter alia with the question whether a law firm could be jointly and severally liable under the Pennsylvania Securities Act. The court concluded that the law firm could be held liable under section 503 of the act, stating that an attorney "can be held liable to an investor as a violator of the Securities Act." Id. at 747
Although Brennan was not addressed in Hill v. Equitable Trust, it can be squared with the prior cases, because those cases suggested that section 501 of the Pennsylvania Securities Act of 1972, Pa.Stat.Ann. tit. 70, § 1-501, only provides a cause of action to buyers and sellers, but they did not consider whether a cause of action under the Act could be obtained under section 503. Pa.Stat.Ann. tit. 70, § 1503; see Bull v. American Bank and Trust Co., 641 F. Supp. 62, 66 n. 5 (E.D.Pa. 1986) ("Neither [Biggans nor Sharp) discusses the scope of civil liability stemming from the interrelationship of §§ 501 and 503."). Thus, even though the prior cases provided a general approach to the issue of what statute of limitations to apply (although even that is in some dispute, see supra Slip Op. at 15-16), they do not provide clear past precedent on which McCarter could have reasonably relied in delaying the filing of a claim under the federal securities laws. This is true with respect to the law as it existed at the time of the accrual of the claim in 1981 and it is also true with respect to the time of the filing of the federal suit on February 17, 1987.
Reliance on clear past precedent would militate in favor of the longer six-year statute of limitations. Because there was no clear past precedent upon which appellants could have reasonably relied in deciding when to file suit, this Chevron prong favors application of the shorter Data Access limitations period.
C. Would Retroactive Application Retard or Further the Rule's Operation?
We next address the second Chevron criterion, which takes into account the history, purpose and effect of the rule in dispute in determining whether retroactive application will further or retard the rule's operation. The district court concluded that this criterion was neutral, following the Hill v. Equitable Trust determination that the second criterion was neutral in that case. See 693 F.2d at 352.
We stated in Data Access that the factual, claim-based approach taken in the prior cases would not promote the strong federal interest in "uniformity, certainty, and the minimization of unnecessary litigation." 843 F.2d at 1543 (quoting Wilson v. Garcia, 471 U.S. at 275). Thus, the purpose of the rule as announced in Data Access was to further uniformity and certainty. Permitting the least number of exceptions to the general rule of retroactivity would, of course, create more uniformity and certainty than would obtain under application of the prior Third Circuit precedent. Therefore, the second criterion might counsel retroactive application of the rule. However, this reasoning swallows the rule; under its regime, whenever a new statute of limitations is set, it would never properly be applied retroactively. We conclude that the second criterion does not "militate clearly either in favor of or against retroactive application" and is therefore neutral. Al-Khazraji v. Saint Francis College, 784 F.2d 505, 513 (3d Cir. 1986), aff'd, 481 U.S. 604, 107 S. Ct. 2022, 95 L. Ed. 2d 582 (1986).
D. Would Retroactive Application Create the Risk of Substantially Inequitable Results?
We agree with the district court that there would be no inequity in applying the Data Access statute of limitations retroactively, because the federal complaint also would have been out of time under analogous state law. 693 F. Supp. at 352. If we were not to apply Data Access, we would have to identify the most analogous state law.
Although the issue has not been squarely decided in Pennsylvania, it would appear that the most analogous state law cause of action to plaintiffs' section 10(b) claim against the defendant broker (Mitcham) and brokerage firm (Butcher and Singer), for deceptively inducing plaintiffs to sell the securities in question would be the Pennsylvania Securities Act, which prohibits brokers from inducing the sale or purchase of any security by means of deception, Pa.Stat.Ann. tit. 70, § 1-403, and which apparently provides for an action by an investor against a broker in section 503, Pa.Stat.Ann. tit. 70, § 1-503. See Bull v. American Bank, 641 F. Supp. at 65-67 & n. 5; Brennan, 304 Pa.Super. at 409, 450 A.2d at 745. The most analogous claim to the plaintiffs' claim against defendant Gabreski, one of the buyers of the securities, appears to be a claim under section 501(b) of the Pennsylvania Securities Act, which provides a cause of action against those who buy securities "by means of any untrue statement . . . or any omission to state a material fact." Pa.Stat.Ann. tit. 70, § 1-501(b).
The limitations period under the Pennsylvania Securities Act is four years from the date of violation or one year after notice, whichever is earlier. Pa.Stat.Ann. tit. 70, § 1-504(a). The filing in federal court 5 1/2, years after the accrual of the action falls beyond that limitation period. Thus, plaintiffs' claims would be barred even if the holding in Data Access were not applied retroactively. Hence, retroactive application of the rule would not be inequitable.
We conclude that the first and third Chevron criteria counsel in favor of retroactive application of Data Access and that the second criterion is neutral, and therefore the shorter one-year or three-year statute of limitations under Data Access governs this case. Thus, the claims, filed 5 1/2 years after accrual of the action, were filed out of time.
We hold that plaintiffs' civil RICO and state claims are barred by res judicata. We also hold that plaintiffs' federal securities law claims are barred by the statute of limitations. For these reasons, the order of the district court dismissing the plaintiffs' complaint will be affirmed.
SLOVITER, Circuit Judge dissenting in part, concurring in part, and concurring in judgment.
I concur in Parts I, II-A, II-B, II-D, and III of the majority opinion. Although I believe the issue is close and the majority's position is supported by respectable authority, I believe that federal courts have exclusive jurisdiction over civil claims under the federal Racketeer Influenced and Corrupt Organizations Act (RICO), 18 U.S.C. § 1964 (1982 & Supp. V 1987). Therefore, I dissent from Part II-C of the majority opinion and would hold that the dismissal with prejudice of plaintiffs' prior suit in state court does not bar assertion of the federal RICO claims. I would, however, find plaintiffs' RICO claims to be barred by the applicable statute of limitations period and therefore concur in the judgment of the majority affirming the district court's dismissal of the complaint.
Gulf Offshore Co. v. Mobil Oil Corp., 453 U.S. 473, 478, 69 L. Ed. 2d 784, 101 S. Ct. 2870 (1981), instructs courts to begin the inquiry as to whether concurrent state and federal jurisdiction exists over a federal law claim "with the presumption that state courts enjoy concurrent jurisdiction."*fn1 The presumption of concurrent jurisdiction may be rebutted in any of three ways: "by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests." Id.
Gulf Offshore explains, "Congress . . . may confine jurisdiction to the federal courts either explicitly or implicitly." Id. at 478 (emphasis added). An examination of the legislative history of RICO leads me to conclude that Congress implicitly confined jurisdiction over civil RICO to the federal courts, a conclusion supported by the incompatibility between state court jurisdiction and several important federal interests at stake.
Looking first to the statutory language, it appears that the jurisdictional provision of civil RICO does not explicitly provide for exclusive jurisdiction in the federal courts. The relevant section reads:
Any person injured in his business or property by reason of a violation of section 1962 of this chapter may sue therefor in any appropriate United States district court and shall recover threefold the damages he sustains and the cost of the suit, including a reasonable attorney's fee.
18 U.S.C. § 1964(c) (1982). While the section does not refer to a plaintiff filing a federal RICO suit in state court, analogous language providing that a plaintiff may sue "in any district court of the United States" has been interpreted to allow concurrent jurisdiction in the context of other federal laws. See Charles Dowd Box Co. v. Courtney, 368 U.S. 502, 7 L. Ed. 2d 483, 82 S. Ct. 519 (1962) (state and federal courts have concurrent jurisdiction under Labor Management Relations Act); see also Gulf Offshore, 453 U.S. at 478-79 (concurrent jurisdiction over Outer Continental Shelf Lands Act).
However, in the case of RICO section 1964(c), the language was taken from the Clayton Act,*fn2 which had been consistently interpreted as providing for exclusive federal jurisdiction over antitrust claims. See, e.g., Marrese v. American Academy of Orthopaedic Surgeons, 470 U.S. 373, 379-80, 84 L. Ed. 2d 274, 105 S. Ct. 1327 (1985); General Inv. Co. v. Lake Shore & M.S. Ry., 260 U.S. 261, 287, 67 L. Ed. 244, 43 S. Ct. 106 (1922).
Therefore, we must examine whether the explicit borrowing of language from a statute granting exclusive jurisdiction to the federal courts provides the requisite "unmistakable implication" in the legislative history of RICO to rebut the presumption of concurrent jurisdiction.
The fact that Congress modeled the enforcement procedures of civil RICO closely after those of the Clayton Act is unmistakably clear in the legislative history of RICO. See Agency Holding Corp. v. Malley-Duff & Assocs., 483 U.S. 143, 107 S. Ct. 2759, 2764, 97 L. Ed. 2d 121 (1987) ("The 'clearest current' in the legislative history of RICO 'is the reliance on the Clayton Act model.'" (quoting Sedima, S.P.R.L. v. Imrex Co., 473 U.S. 479, 489, 87 L. Ed. 2d 346, 105 S. Ct. 3275 (1985)). RICO represented Congress' effort to adapt the "time tested machinery of the antitrust laws" to the problem of organized crime. See Agency Holding Corp., 107 S. Ct. at 2764 (quoting from Report of American Bar Association's Antitrust Section, reprinted at 115 Cong.Rec. 6994, 6995 (1969)). The influential ABA Report presented to Congress during hearings on the development of an anti-racketeering statute proposed that the Clayton Act's provisions authorizing treble damage suits by private parties for antitrust violations be incorporated in an independent statute aimed at combatting organized crime. See 115 Cong.Rec. at 6995. This suggestion immediately won support in Congress and the final version of the statute enacted into law contained precisely such a provision in the form of 18 U.S.C. § 1964(c).
Prof. G. Robert Blakey, principal architect of RICO in his then-position as chief counsel of the Senate Subcommittee on Criminal Laws and Procedures, is reported to have stated that Congress never focused its attention on the question of whether jurisdiction over civil RICO should be exclusively federal or concurrent with state courts, but that "'courts can infer from the statute that if Congress had thought about it, they would have made [jurisdiction] exclusive . . . [since] the antitrust law is an exclusive-jurisdiction statute.'" Flaherty, Two States Lay Claim to RICO, Nat'l L.J., May 7, 1984, at 3, col. 1, quoted in Greenview Trading Co. v. Hershman & Leicher, 108 A.D.2d 468, , 489 N.Y.S.2d 502, 505 (1985).
In Agency Holding Corp., the Supreme Court found the "clear legislative intent to pattern RICO's civil enforcement provision on the Clayton Act" to "strongly counsel[ ] in favor" of applying the statute of limitations period specified in the Clayton Act to civil RICO claims despite the lack of an express intent by Congress to so incorporate the Clayton Act's statute of limitations provision. 107 S. Ct. at 2765. The same reasoning counsels applying the interpretation of the Clayton Act giving federal courts exclusive jurisdiction to the interpretation of the similar language in civil RICO. Because there is no indication that Congress intended any change in this respect, the language should be given the same effect as it had in the prior statute. See, e.g., Albemarle Paper Co. v. Moody, 422 U.S. 405, 419-20 & n. 19, 45 L. Ed. 2d 280, 95 S. Ct. 2362 (1975) (where Congress modeled backpay provision of Title VII on that of National Labor Relations Act (NLRA), Court would assume that Congress was aware of judicial interpretation given NLRA provision and would give same interpretation to Title VII).
The Court's reliance on the Clayton Act in interpreting RICO has not been limited to the statute of limitations context. In its earlier decision in Sedima, S.P.R.L. v. Imrex Co., 473 U.S. 479, 87 L. Ed. 2d 346, 105 S. Ct. 3275 (1985), the Court looked to the Clayton Act to interpret the term "violation" in section 1964(c). It rejected the argument that plaintiffs in a civil RICO suit must demonstrate that defendants had been convicted of the predicate offenses alleged, noting that the Clayton Act, upon which RICO was based, did not require prior conviction in a government prosecution as a prerequisite for bringing a civil suit. Id. at 489. The Court stated that changes in the relevant language of the Clayton Act made by the drafters of the civil RICO provision did not manifest a difference in congressional intent where the Congressman who proposed the wording "nowhere indicated a desire to depart from the antitrust model." Id. at n. 8.
We recognize that the Sedima Court refused to impose a requirement that civil RICO plaintiffs prove a distinct "racketeering injury," even though plaintiffs under the Clayton Act are required to show an "antitrust injury." Id. at 495-99. The Court explained that to interpret RICO otherwise would "create[ ] exactly the problems Congress sought to avoid by placing RICO outside the antitrust laws." Id. at 499. Sedima thus suggests that interpretation of RICO by way of reference to the Clayton Act is not appropriate where the two statutes differ with respect to their underlying goals, but is appropriate and efficacious where there is no indication that Congress had any intent other than to incorporate the civil enforcement provisions of the Clayton Act without modification. The grant of original jurisdiction in the federal district courts appears to involve precisely such wholesale borrowing from the Clayton Act, and I would therefore interpret the identical words of the two acts as possessing the same meaning.
With respect to the third of the factors enumerated in Gulf Offshore which may rebut the presumption against exclusive federal jurisdiction over a federal law, i.e., incompatibility of state court jurisdiction and federal interests, the Gulf Offshore Court listed several potential sources of such incompatibility, including "the desirability of uniform interpretation, the expertise of federal judges in federal law, and the assumed greater hospitality of federal courts to peculiarly federal claims." 453 U.S. at 483-84 (footnote omitted). Examination of each of these factors enforces my conclusion that jurisdiction over civil RICO should be exclusively federal.
The first factor strongly counseling in favor of exclusive federal jurisdiction is the desirability of uniform construction of this federal statute whose wide-ranging applicability and open-ended terminology have repeatedly taxed the interpretative powers of the courts. See Sedima, 473 U.S. at 500 (invalidating Second Circuit's attempt to read certain narrowing constructions into the provisions of civil RICO but expressing sympathy with that court's "doubts" as to the direction in which interpretation of the statute has evolved). Imposition of jurisdiction over civil RICO claims on the state courts will only compound the problem of burgeoning and inconsistent interpretations of the statute.
The "expertise of federal judges in federal law, and the assumed greater hospitality of federal courts to peculiarly federal claims," Gulf Offshore, 453 U.S. at 484, also counsels in favor of vesting jurisdiction over civil RICO exclusively in the federal courts. Although a handful of the predicate acts defined as racketeering activity in 18 U.S.C. § 1961 are state law offenses, see 18 U.S.C. § 1961(1)(A), a much larger portion of the predicate acts are federal crimes, see id. at (B), (C) & (D), often involving "highly specialized areas" of federal law. Chivas Prods. Ltd. v. Owen, 864 F.2d 1280, 1285 (6th Cir. 1988). The state courts' experience with the large majority of the statutes constituting RICO predicate offenses will have been limited to those ocassions on which issues under these statutes arose collaterally. See Hathorn v. Lovorn, 457 U.S. 255, 268, 72 L. Ed. 2d 824, 102 S. Ct. 2421 (1982).
It is also significant that at least twenty-five states have passed their own anti-racketeering statutes, many of which contain provisions for private civil suits patterned after federal RICO. See, e.g., Del.Code Ann. tit. 11 § 1505(c) (1987); N.J.Stat.Ann. § 2C:41-4(c) (1982); see generally RICO Bus.Disputes Guide (CCH) paras. 4000-4490 (1985) (reprinting state statutes). These state statutes reflect the states' decisions on the nature and extent of the resources of their judicial systems which they wish to devote to anti-racketeering measures in much the same way that states have legislated in the fields of securities and antitrust regulation where they saw fit. In this situation of flourishing state experimentation in the development of anti-racketeering initiatives, it seems to me that the "diversities and conflicts" which are a beneficial aspect of our federal system and its general rule of concurrent jurisdiction, Charles Dowd, 368 U.S. at 514, are best promoted by leaving the state courts free to focus on the interpretation and application of their own anti-racketeering laws and confining jurisdiction over the federal statute to the federal courts in keeping with the clear implications in RICO's legislative history.
Because I agree with the majority that Pennsylvania, like almost all states, would not bar a party from asserting in a subsequent suit those claims which could not have been entertained by a court in hearing a prior suit because it lacked subject matter jurisdiction over such claims, I would hold that plaintiffs' subsequent assertion of their federal RICO claims in their federal court suit was not barred by res judicata principles.
I concur in the judgment, however, because under the applicable statute of limitations period, see Agency Holding Corp., 107 S. Ct. at 2767 (Clayton Act's four-year statute of limitations applies to civil RICO claims), plaintiffs' claims under RICO, predicated on allegedly fraudulent transactions occurring in 1981, see maj. typescript op. at 13-14, are time-barred. Cf. maj. typescript op. at 23 (plaintiffs' state securities law claims time barred under applicable four-year statute of limitations). I would therefore affirm the district court's dismissal of plaintiffs' complaint, but for different reasons than those stated in Part II-C of the majority's opinion, from which I respectfully dissent.