LAMPF v. GILBERTSON

JUSTICE BLACKMUN delivered the opinion of the Court, except as to Part II-A.

In this litigation we must determine which statute of limitations is applicable to a private suit brought pursuant to § 10(b) of the Securities Exchange Act of 1934, 48 Stat. 891, 15 U. S. C. § 78j(b), and to Securities and Exchange Commission Rule 10b-5, 17 CFR §240.10b-5 (1990), promulgated thereunder.

I

The controversy arises from the sale of seven Connecticut limited partnerships formed for the purpose of purchasing and leasing computer hardware and software. Petitioner Lampf, Pleva, Lipkind, Prupis & Petigrow is a West Orange, N. J., law firm that aided in organizing the partnerships and that provided additional legal services, including the preparation of opinion letters addressing the tax consequences of investing in the partnerships. The several plaintiff-respondents purchased units in one or more of the partnerships during the years 1979 through 1981 with the expectation of realizing federal income tax benefits therefrom.

The partnerships failed, due in part to the technological obsolescence of their wares. In late 1982 and early 1983, plaintiff-respondents received notice that the United States Internal Revenue Service was investigating the partnerships. The IRS subsequently disallowed the claimed tax benefits because of overvaluation of partnership assets and lack of profit motive.

On November 3, 1986, and June 4, 1987, plaintiff-respondents filed their respective complaints in the United States District Court for the District of Oregon, naming as defendants petitioner and others involved in the preparation of offering memoranda for the partnerships. The complaints alleged that plaintiff-respondents were induced to invest in the partnerships by misrepresentations in the offering memoranda, in violation of, among other things, § 10(b) of the 1934 Act and Rule 10b-5. The claimed misrepresentations were said to include assurances that the investments would entitle the purchasers to substantial tax benefits; that the leasing of the hardware and software packages would generate a profit; that the software was readily marketable; and that certain equipment appraisals were accurate and reasonable. Plaintiff-respondents asserted that they became aware of the alleged misrepresentations only in 1985 following the disallowance by the IRS of the tax benefits claimed.

After consolidating the actions for discovery and pretrial proceedings, the District Court granted summary judgment for the defendants on the ground that the complaints were not timely filed. App. to Pet. for Cert. 22A. Following precedent of its controlling court, see, e. g., Robuck v. Dean Witter & Co., 649 F.2d 641 (CA9 1980), the District Court ruled that the securities claims were governed by the state statute of limitations for the most analogous forum-state cause of action. The court determined this to be Oregon's 2-year limitations period for fraud claims, Ore. Rev. Stat. § 12.110(1) (1989). The court found that reports to plaintiff-respondents detailing the declining financial status of each partnership and allegations of misconduct made known to the general partners put plaintiff-respondents on "inquiry notice" of the possibility of fraud as early as October 1982. App. to Pet. for Cert. 43A. The court also ruled that the distribution of certain fiscal reports and the installation of a general partner previously associated with the defendants did not constitute fraudulent concealment sufficient to toll the statute of limitations. Applying the Oregon statute to the facts underlying plaintiff-respondents' claims, the District Court determined that each complaint was time barred.

The Court of Appeals for the Ninth Circuit reversed and remanded the cases. See, e. g., Reitz v. Leasing Consultants Associates, 895 F.2d 1418 (1990) (judgment order). In its unpublished opinion, the Court of Appeals found that unresolved factual issues as to when plaintiff-respondents discovered or should have discovered the alleged fraud precluded summary judgment. Then, as did the District Court, it selected the 2-year Oregon limitations period. In so doing, it implicitly rejected petitioner's argument that a federal limitations period should apply to Rule 10b-5 claims. App. to Pet. for Cert. 8A. In view of the divergence of opinion among the Circuits regarding the proper limitations period for Rule 10b-5 claims,1 we granted certiorari to address this important issue. 498 U.S. 894 (1990).

II

Plaintiff-respondents maintain that the Court of Appeals correctly identified common-law fraud as the source from which § 10(b) limitations should be derived. They submit that the underlying policies and practicalities of § 10(b) litigation do not justify a departure from the traditional practice of "borrowing" analogous state-law statutes of limitations. Petitioner, on the other hand, argues that a federal period is appropriate, contending that we must look to the "1-and-3-year" structure applicable to the express causes of action in § 13 of the Securities Act of 1933, 48 Stat. 84, as amended, 15 U. S. C. § 77m, and to certain of the express actions in the 1934 Act, see 15 U. S. C. §§ 78i(e), 78r(c), and 78cc(b).2 The Solicitor General, appearing on behalf of the Securities and Exchange Commission, agrees that use of a federal period is indicated, but urges the application of the 5-year statute of repose specified in § 20A of the 1934 Act, 15 U. S. C. § 78t-1(b)(4), as added by § 5 of the Insider Trading and Securities Fraud Enforcement Act of 1988, 102 Stat. 4681. The 5-year period, it is said, accords with "Congress's most recent views on the accommodation of competing interests, provides the closest federal analogy, and promises to yield the best practical and policy results in Rule 10b-5 litigation." Brief for Securities and Exchange Commission as Amicus Curiae 8. For the reasons discussed below, we agree that a uniform federal period is indicated, but we hold that the express causes of action contained in the 1933 and 1934 Acts provide the source.

A

It is the usual rule that when Congress has failed to provide a statute of limitations for a federal cause of action, a court "borrows" or "absorbs" the local time limitation most analogous to the case at hand. Wilson v. Garcia, 471 U.S. 261, 266-267 (1985); Automobile Workers v. Hoosier Cardinal Corp., 383 U.S. 696, 704 (1966); Campbell v. Haverhill, 155 U.S. 610, 617 (1895). This practice, derived from the Rules of Decision Act, 28 U. S. C. § 1652, has enjoyed sufficient longevity that we may assume that, in enacting remedial legislation, Congress ordinarily "intends by its silence that we borrow state law." Agency Holding Corp. v. Malley-Duff & Associates, Inc., 483 U.S. 143, 147 (1987).

The rule, however, is not without exception. We have recognized that a state legislature rarely enacts a limitations period with federal interests in mind, Occidental Life Ins. Co. of Cal. v. EEOC, 432 U.S. 355, 367 (1977), and when the operation of a state limitations period would frustrate the policies embraced by the federal enactment, this Court has looked to federal law for a suitable period. See, e. g., DelCostello v. Teamsters, 462 U.S. 151 (1983); Agency Holding Corp., supra; McAllister v. Magnolia Petroleum Co., 357 U.S. 221, 224 (1958). These departures from the state-borrowing doctrine have been motivated by this Court's conclusion that it would be "inappropriate to conclude that Congress would choose to adopt state rules at odds with the purpose or operation of federal substantive law." DelCostello, 462 U. S., at 161.

Rooted as it is in the expectations of Congress, the "state-borrowing doctrine" may not be lightly abandoned. We have described federal borrowing as "a closely circumscribed exception," to be made "only `when a rule from elsewhere in federal law clearly provides a closer analogy than available state statutes, and when the federal policies at stake and the practicalities of litigation make that rule a significantly more appropriate vehicle for interstitial lawmaking.'" Reed v. United Transportation Union, 488 U.S. 319, 324 (1989), quoting DelCostello, 462 U. S., at 172.

Predictably, this determination is a delicate one. Recognizing, however, that a period must be selected,3 our cases do provide some guidance as to whether state or federal borrowing is appropriate and as to the period best suited to the cause of action under consideration. From these cases we are able to distill a hierarchical inquiry for ascertaining the appropriate limitations period for a federal cause of action where Congress has not set the time within which such an action must be brought.

First, the court must determine whether a uniform statute of limitations is to be selected. Where a federal cause of action tends in practice to "encompass numerous and diverse topics and subtopics," Wilson v. Garcia, 471 U. S., at 273, such that a single state limitations period may not be consistently applied within a jurisdiction, we have concluded that the federal interests in predictability and judicial economy counsel the adoption of one source, or class of sources, for borrowing purposes. Id., at 273-275. This conclusion ultimately may result in the selection of a single federal provision, see Agency Holding Corp., supra, or of a single variety of state actions. See Wilson v. Garcia (characterizing all actions under 42 U. S. C. § 1983 as analogous to a state-law personal injury action).

Second, assuming a uniform limitations period is appropriate, the court must decide whether this period should be derived from a state or a federal source. In making this judgment, the court should accord particular weight to the geographic character of the claim:

Finally, even where geographic considerations counsel federal borrowing, the aforementioned presumption of state borrowing requires that a court determine that an analogous federal source truly affords a "closer fit" with the cause of action at issue than does any available state-law source. Although considerations pertinent to this determination will necessarily vary depending upon the federal cause of action and the available state and federal analogues, such factors as commonality of purpose and similarity of elements will be relevant.

B

In the present litigation, our task is complicated by the nontraditional origins of the § 10(b) cause of action. The text of § 10(b) does not provide for private claims.4 Such claims are of judicial creation, having been implied under the statute for nearly half a century. See Kardon v. National Gypsum Co., 69 F.Supp. 512 (ED Pa. 1946), cited in Ernst & Ernst v. Hochfelder, 425 U.S. 185, 196, n. 16 (1976). Although this Court repeatedly has recognized the validity of such claims, see Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 730 (1975); Affiliated Ute Citizens of Utah v. United States, 406 U.S. 128, 150-154 (1972); Superintendent of Ins. of N. Y. v. Bankers Life & Casualty Co., 404 U.S. 6, 13, n. 9 (1971), we have made no pretense that it was Congress' design to provide the remedy afforded. See Ernst & Ernst, 425 U. S., at 196 ("[T]here is no indication that Congress, or the Commission when adopting Rule 10b-5, contemplated such a remedy") (footnotes omitted). It is therefore no surprise that the provision contains no statute of limitations.

In a case such as this, we are faced with the awkward task of discerning the limitations period that Congress intended courts to apply to a cause of action it really never knew existed. Fortunately, however, the drafters of § 10(b) have provided guidance.

We conclude that where, as here, the claim asserted is one implied under a statute that also contains an express cause of action with its own time limitation, a court should look first to the statute of origin to ascertain the proper limitations period. We can imagine no clearer indication of how Congress would have balanced the policy considerations implicit in any limitations provision than the balance struck by the same Congress in limiting similar and related protections. See DelCostello, 462 U. S., at 171; United Parcel Service, Inc. v. Mitchell, 451 U.S. 56, 69-70 (1981) (opinion concurring in judgment). When the statute of origin contains comparable express remedial provisions, the inquiry usually should be at an end. Only where no analogous counterpart is available should a court then proceed to apply state-borrowing principles.

In the present litigation, there can be no doubt that the contemporaneously enacted express remedial provisions represent "a federal statute of limitations actually designed to accommodate a balance of interests very similar to that at stake here — a statute that is, in fact, an analogy to the present lawsuit more apt than any of the suggested state-law parallels." DelCostello, 462 U. S., at 169. The 1934 Act contained a number of express causes of action, each with an explicit limitations period. With only one more restrictive exception,5 each of these includes some variation of a 1-year period after discovery combined with a 3-year period of repose.6 In adopting the 1934 Act, the 73d Congress also amended the limitations provision of the 1933 Act, adopting the 1-and-3-year structure for each cause of action contained therein.7

Section 9 of the 1934 Act, 15 U. S. C. § 78i, pertaining to the willful manipulation, of security prices, and § 18, 15 U. S. C. § 78r, relating to misleading filings, target the precise dangers that are the focus of § 10(b). Each is an integral element of a complex web of regulations. Each was intended to facilitate a central goal: "to protect investors against manipulation of stock prices through regulation of transactions upon securities exchanges and in over-the-counter markets, and to impose regular reporting requirements on companies whose stock is listed on national securities exchanges." Ernst & Ernst, 425 U. S., at 195, citing S. Rep. No. 792, 73d Cong., 2d Sess., 1-5 (1934).

C

We therefore conclude that we must reject the Commission's contention that the 5-year period contained in § 20A, added to the 1934 Act in 1988, is more appropriate for § 10(b) actions than is the 1-and-3-year structure in the Act's original remedial provisions. The Insider Trading and Securities Fraud Enforcement Act of 1988, which became law more than 50 years after the original securities statutes, focuses upon a specific problem, namely, the "purchasing or selling [of] a security while in possession of material, nonpublic information," 15 U. S. C. § 78t-1(a), that is, "insider trading." Recognizing the unique difficulties in identifying evidence of such activities, the 100th Congress adopted § 20A as one of "a variety of measures designed to provide greater deterrence, detection and punishment of violations of insider trading." H. R. Rep. No. 100-910, p. 7 (1988). There is no indication that the drafters of § 20A sought to extend that enhanced protection to other provisions of the 1934 Act. Indeed, the text of § 20A indicates the contrary. Section 20A(d) states: "Nothing in this section shall be construed to limit or condition the right of any person to bring an action to enforce a requirement of this chapter or the availability of any cause of action implied from a provision of this chapter." 15 U. S. C. § 78t-1(d).

The Commission further argues that because some conduct that is violative of § 10(b) is also actionable under § 20A, adoption of a 1-and-3-year structure would subject actions based on § 10(b) to two different statutes of limitations. But § 20A also prohibits insider trading activities that violate sections of the 1934 Act with express limitations periods. The language of § 20A makes clear that the 100th Congress sought to alter the remedies available in insider trading cases, and only in insider trading cases. There is no inconsistency.

Finally, the Commission contends that the adoption of a 3-year period of repose would frustrate the policies underlying § 10(b). The inclusion, however, of the 1-and-3-year structure in the broad range of express securities actions contained in the 1933 and 1934 Acts suggests a congressional determination that a 3-year period is sufficient. See Ceres Partners v. GEL Associates, 918 F.2d 349, 363 (CA2 1990).

Thus, we agree with every Court of Appeals that has been called upon to apply a federal statute of limitations to a § 10(b) claim that the express causes of action contained in the 1933 and 1934 Acts provide a more appropriate statute of limitations than does § 20A. See Ceres Partners, supra; Short v. Belleville Shoe Mfg. Co., 908 F.2d 1385 (CA7 1990), cert. pending, No. 90-526; In re Data Access Systems Securities Litigation, 843 F.2d 1537 (CA3), cert. denied sub nom. Vitiello v. I. Kahlowsky & Co., 488 U.S. 849 (1988).

Necessarily, we also reject plaintiff-respondents' assertion that state-law fraud provides the closest analogy to § 10(b). The analytical framework we adopt above makes consideration of state-law alternatives unnecessary where Congress has provided an express limitations period for correlative remedies within the same enactment.8

III

Finally, we address plaintiff-respondents' contention that, whatever limitations period is applicable to § 10(b) claims, that period must be subject to the doctrine of equitable tolling. Plaintiff-respondents note, correctly, that "[t]ime requirements in lawsuits . . . are customarily subject to `equitable tolling.'" Irwin v. Department of Veterans Affairs, 498 U.S. 89, 95 (1990), citing Hallstrom v. Tillamook County, 493 U.S. 20, 27 (1989). Thus, this Court has said that in the usual case, "where the party injured by the fraud remains in ignorance of it without any fault or want of diligence or care on his part, the bar of the statute does not begin to run until the fraud is discovered, though there be no special circumstances or efforts on the part of the party committing the fraud to conceal it from the knowledge of the other party." Bailey v. Glover, 21 Wall. 342, 348 (1875); see also Holmberg v. Armbrecht, 327 U.S. 392, 396-397 (1946). Notwithstanding this venerable principle, it is evident that the equitable tolling doctrine is fundamentally inconsistent with the 1-and-3-year structure.

The 1-year period, by its terms, begins after discovery of the facts constituting the violation, making tolling unnecessary. The 3-year limit is a period of repose inconsistent with tolling. One commentator explains: "[T]he inclusion of the three-year period can have no significance in this context other than to impose an outside limit." Bloomenthal, The Statute of Limitations and Rule 10b-5 Claims: A Study in Judicial Lassitude, 60 U. Colo. L. Rev. 235, 288 (1989). See also ABA Committee on Federal Regulation of Securities, Report of the Task Force on Statute of Limitations for Implied Actions 645, 655 (1986) (advancing "the inescapable conclusion that Congress did not intend equitable tolling to apply in actions under the securities laws"). Because the purpose of the 3-year limitation is clearly to serve as a cutoff, we hold that tolling principles do not apply to that period.

IV

Litigation instituted pursuant to § 10(b) and Rule 10b-5 therefore must be commenced within one year after the discovery of the facts constituting the violation and within three years after such violation.9 As there is no dispute that the earliest of plaintiff-respondents' complaints was filed more than three years after petitioner's alleged misrepresentations, plaintiff-respondents' claims were untimely.10

The judgment of the Court of Appeals is reversed.

It is so ordered.

JUSTICE SCALIA, concurring in part and concurring in the judgment.

Although I accept the stare decisis effect of decisions we have made with respect to the statutes of limitations applicable to particular federal causes of action, I continue to disagree with the methodology the Court has very recently adopted for purposes of making those decisions. In my view, absent a congressionally created limitations period state periods govern, or, if they are inconsistent with the purposes of the federal Act, no limitations period exists. See Agency Holding Corp. v. Malley-Duff & Associates, Inc., 483 U.S. 143, 157-170 (1987) (SCALIA, J., concurring in judgment), see also Reed v. United Transportation Union, 488 U.S. 319, 334 (1989) (SCALIA, J., concurring in judgment).

The present case presents a distinctive difficulty because it involves one of those so-called "implied" causes of action that, for several decades, this Court was prone to discover in — or, more accurately, create in reliance upon — federal legislation. See Thompson v. Thompson, 484 U.S. 174, 190 (1988) (SCALIA, J., concurring in judgment). Raising up causes of action where a statute has not created them may be a proper function for common-law courts, but not for federal tribunals. See id., at 191-192; Cannon v. University of Chicago, 441 U.S. 677, 730-749 (1979) (Powell, J., dissenting). We have done so, however, and thus the question arises what statute of limitations applies to such a suit. Congress has not had the opportunity (since it did not itself create the cause of action) to consider whether it is content with the state limitations or would prefer to craft its own rule. That lack of opportunity is particularly apparent in the present case, since Congress did create special limitations periods for the Securities Exchange Act of 1934 causes of actions that it actually enacted. See 15 U. S. C. §§ 78p(b), 78i(e), 78r(c); see also § 77m.

When confronted with this situation, the only thing to be said for applying my ordinary (and the Court's pre-1983 traditional) rule is that the unintended and possibly irrational results will certainly deter judicial invention of causes of action. That is not an unworthy goal, but to pursue it in that fashion would be highly unjust to those who must litigate past inventions. An alternative approach would be to say that since we "implied" the cause of action we ought to "imply" an appropriate statute of limitations as well. That is just enough, but too lawless to be imagined. It seems to me the most responsible approach, where the enactment that has been the occasion for our creation of a cause of action contains a limitations period for an analogous cause of action, is to use that. We are imagining here. And I agree with the Court that "[w]e can imagine no clearer indication of how Congress would have balanced the policy considerations implicit in any limitations provision than the balance struck by the same Congress in limiting similar and related protections." Ante, at 359.

I join the judgment of the Court, and all except Part II-A of the Court's opinion.

JUSTICE STEVENS, with whom JUSTICE SOUTER joins, dissenting.

In my opinion, the Court has undertaken a lawmaking task that should properly be performed by Congress. Starting from the premise that the federal cause of action for violating § 10(b) of the Securities Exchange Act of 1934, 48 Stat. 891, 15 U. S. C. § 78j(b), was created out of whole cloth by the Judiciary, it concludes that the Judiciary must also have the authority to fashion the time limitations applicable to such an action. A page from the history of § 10(b) litigation will explain why both the premise and the conclusion are flawed.

The private cause of action for violating § 10(b) was first recognized in Kardon v. National Gypsum Co., 69 F.Supp. 512 (ED Pa. 1946). In recognizing this implied right of action, Judge Kirkpatrick merely applied what was then a well-settled rule of federal law. As was true during most of our history, the federal courts then presumed that a statute enacted to benefit a special class provided a remedy for those members injured by violations of the statute. See Texas & Pacific R. Co. v. Rigsby, 241 U.S. 33, 39-40 (1916).1 Judge Kirkpatrick did not make "new law" when he applied this presumption to a federal statute enacted for the benefit of investors in securities that are traded in interstate commerce.

During the ensuing four decades of administering § 10(b) litigation, the federal courts also applied settled law when they looked to state law to find the rules governing the timeliness of claims. See DelCostello v. Teamsters, 462 U.S. 151, 172-173 (1983) (STEVENS, J., dissenting).2 It was not until 1988 that a federal court decided that it would be better policy to have a uniform federal statute of limitations apply to claims of this kind. See In re Data Access Systems Securities Litigation, 843 F.2d 1537 (CA3). I agree that such a uniform limitations rule is preferable to the often chaotic traditional approach of looking to the analogous state limitation. I believe, however, that Congress, rather than the Federal Judiciary, has the responsibility for making the policy determinations that are required in rejecting a rule selected under the doctrine of state borrowing, long applied in § 10(b) cases, and choosing a new limitations period and its associated tolling rules.3 When a legislature enacts a new rule of law governing the timeliness of legal action, it can — and usually does — specify the effective date of the rule and determine the extent to which it shall apply to pending claims. See, e. g., 104 Stat. 5114, quoted ante, at 364, n. 10. When the Court ventures into this lawmaking arena, however, it inevitably raises questions concerning the retroactivity of its new rule that are difficult and arguably inconsistent with the neutral, nonpolicymaking role of the judge. See Chevron Oil Co. v. Huson, 404 U.S. 97 (1971); In re Data Access, 843 F. 2d, at 1551 (Seitz, J., dissenting).

The Court's rejection of the traditional rule of applying a state limitations period when the federal statute is silent is not justified by this Court's prior cases. Despite the majority's recognition of the traditional rule, ante, at 355, it effectively repudiates it by holding that "[o]nly where no analogous counterpart [within the statute] is available should a court then proceed to apply state-borrowing principles." Ante, at 359. The Court's principal justification for this departure is that it took similar action in DelCostello, supra. I registered my dissent in that case for reasons similar to those I express today. In that case there was nothing in the statute to lead me to believe that Congress intended to depart from our settled practice of looking to analogous state limitations. Id., at 171-173. Likewise in this case, I can find nothing in the Securities Exchange Act of 1934 that leads me to believe that Congress intended us to depart from our traditional rule and overrule four decades of established law.

The other case on which the Court primarily relies, Agency Holding Corp. v. Malley-Duff & Associates, Inc., 483 U.S. 143 (1987), is distinguishable from this case. Agency Holding did not involve a change in a rule of law that had been settled for 40 years. Furthermore, in that case, the Court found an explicit intent to pattern the RICO private remedy after the Clayton Act's private antitrust remedy. The remedy in the Clayton Act was subject to a 4-year statute of limitations, and the Court reasonably inferred that Congress wanted the same limitations period to apply to both statutes. The Court has not found a similar intent to pattern § 10 of the Securities Exchange Act of 1934 after those sections subject to a 1-and-3-year limitation. See ante, at 359-361.

The policy choices that the Court makes today may well be wise — even though they are at odds with the recommendation of the Executive Branch — but that is not a sufficient justification for making a change in what was well-settled law during the years between 1946 and 1988 governing the timeliness of action impliedly authorized by a federal statute. This Court has recognized that a rule of statutory construction that has been consistently applied for several decades acquires a clarity that "is simply beyond peradventure." Herman & MacLean v. Huddleston, 459 U.S. 375, 380 (1983). I believe that the Court should continue to observe that principle in this case. The Court's occasional departure from that principle does not justify today's refusal to comply with the Rules of Decision Act. See, e. g., Shearson/ American Express Inc. v. McMahon, 482 U.S. 220, 268 (1987) (STEVENS, J., dissenting). Accordingly, I respectfully dissent.

JUSTICE O'CONNOR, with whom JUSTICE KENNEDY joins, dissenting.

I agree that predictability and judicial economy counsel the adoption of a uniform federal statute of limitations for actions brought under § 10(b) and Rule 10b-5. For the reasons stated by JUSTICE KENNEDY, however, I believe we should adopt the 1-year-from-discovery-rule, but not the 3-year period of repose. I write separately only to express my disagreement with the Court's decision in Part IV to apply the new limitations period in this case. In holding that respondents' suit is time barred under a limitations period that did not exist before today, the Court departs drastically from our established practice and inflicts an injustice on the respondents. The Court declines to explain its unprecedented decision, or even to acknowledge its unusual character.

Respondents, plaintiffs below, filed this action in Federal District Court in 1986. Everyone agrees that, at that time, their claims were governed by the state statute of limitations for the most analogous state cause of action. This was mandated by a solid wall of binding Ninth Circuit authority dating back more than 30 years.1 See ante, at 353. The case proceeded in the District Court and the Court of Appeals for almost four years. During that time, the law never changed; the governing limitations period remained the analogous state statute of limitations.2 Notwithstanding respondents' entirely proper reliance on this limitations period, the Court now holds that their suit must be dismissed as untimely because respondents did not comply with a federal limitations period announced for the first time today — 4½ years after the suit was filed. Quite simply, the Court shuts the courthouse door on respondents because they were unable to predict the future.

One might get the impression from the Court's matter-of-fact handling of the retroactivity issue that this is our standard practice. Part IV of the Court's opinion comprises, after all, only two sentences: the first sentence sets out the 1-and-3-year rule; the second states that respondents' complaint is untimely for failure to comply with the rule. Surely, one might think, if the Court were doing anything out of the ordinary, it would comment on the fact.

Apparently not. This Court has, on several occasions, announced new statutes of limitations. Until today, however, the Court had never applied a new limitations period retroactively to the very case in which it announced the new rule so as to bar an action that was timely under binding Circuit precedent. Our practice has been instead to evaluate the case at hand by the old limitations period, reserving the new rule for application in future cases.

A prime example is Chevron Oil Co. v. Huson, 404 U.S. 97 (1971). The issue in that case was whether state or federal law governed the timeliness of an action brought under a particular federal statute. At the time the lawsuit was initiated, the rule was that federal law governed. This Court changed the rule, holding that the timeliness of an action should be governed by state law. The Court declined to apply the state statute of limitations in that case, however, because the action had been filed long before the new rule was announced. The Court recognized, sensibly, that its decision overruled a long line of Court of Appeals' decisions on which the respondent had properly relied, id., at 107; that retroactive application would be inconsistent with the purpose of using state statutes of limitations, id., at 107-108; and that it would be highly inequitable to pretend that the respondent had "`slept on his rights'" when, in reality, he had complied fully with the law as it existed and could not have foreseen that the law would change. Id., at 108.

We followed precisely the same course several years later in Saint Francis College v. Al-Khazraji, 481 U.S. 604 (1987). We declined to apply a decision specifying the applicable statute of limitations retroactively because doing so would bar a suit that, under controlling Circuit precedent, had been filed in a timely manner. We relied expressly on the analysis of Chevron Oil, holding that a decision identifying a new limitations period should be applied only prospectively where it overrules clearly established Circuit precedent, where retroactive application would be inconsistent with the purpose of the underlying statute, and where doing so would be "manifestly inequitable." Saint Francis College, supra, at 608-609.

Chevron Oil and Saint Francis College are based on fundamental notions of justified reliance and due process. They reflect a straightforward application of an earlier line of cases holding that it violates due process to apply a limitations period retroactively and thereby deprive a party arbitrarily of a right to be heard in court. See Wilson v. Iseminger, 185 U.S. 55, 62 (1902); Brinkerhoff-Faris Trust & Savings Co. v. Hill, 281 U.S. 673, 681-682 (1930). Not surprisingly, then, the Court's decision in Chevron Oil and Saint Francis College not to apply new limitations periods retroactively generated no disagreement among Members of the Court: The opinion in Chevron Oil was joined by all but one Justice, who did not reach the retroactivity question; Saint Francis College was unanimous.

Only last Term, eight Justices reaffirmed the commonsense rule that decisions specifying the applicable statute of limitations apply only prospectively. See American Trucking Assns., Inc. v. Smith, 496 U.S. 167 (1990). The question presented in American Trucking was whether an earlier decision of the Court — striking down as unconstitutional a particular state highway tax scheme — would apply retroactively. In the course of explaining why the ruling would not apply retroactively, the plurality opinion relied heavily on our statute of limitations cases:

Four other Justices, while disagreeing that Chevron Oil's retroactivity analysis should apply in other contexts, reaffirmed its application to statutes of limitations. The dissenting Justices stated explicitly that it would be "most inequitable to [hold] that [a] plaintiff ha[s] `"slept on his rights"' during a period in which neither he nor the defendant could have known the time limitation that applied to the case." American Trucking, supra, at 220 (STEVENS, J., dissenting), quoting Chevron Oil, supra, at 108.

After American Trucking, the continued vitality of Chevron Oil with respect to statutes of limitations is — or should be — irrefutable; nothing in James B. Beam Distilling Co. v. Georgia, post, p. 529, alters this fact. The present case is indistinguishable from Chevron Oil and retroactive application should therefore be denied. All three Chevron Oil factors are met. First, in adopting a federal statute of limitations, the Court overrules clearly established Circuit precedent; the Court admits as much. Ante, at 353. Second, the Court explains that "the federal interes[t] in predictability" demands a uniform standard. Ante, at 357. I agree, but surely predictability cannot favor applying retroactively a limitations period that the respondents could not possibly have foreseen. Third, the inequitable results are obvious. After spending 4½ years in court and tens of thousands of dollars in attorney's fees, respondents' suit is dismissed for failure to comply with a limitations period that did not exist until today.

Earlier this Term, the Court observed that "the doctrine of stare decisis serves profoundly important purposes in our legal system." California v. Acevedo, 500 U.S. 565, 579 (1991). If that is so, it is difficult to understand the Court's decision today to apply retroactively a brand new statute of limitations. Part IV of the Court's opinion, without discussing the relevant cases or even acknowledging the issue, declines to follow the precedent established in Chevron Oil, Saint Francis College, and American Trucking, not to mention Wilson and Brinkerhoff-Faris.

The Court's cursory treatment of the retroactivity question cannot be an oversight. The parties briefed the issue in this Court. See Brief for Respondents 45-48; Reply Brief for Petitioner 18-20. In addition, the United States, filing an amicus curiae brief on behalf of the Securities and Exchange Commission, addressed the issue explicitly, urging the Court to remand so that the lower court may address the retroactivity question in the first instance. Nevertheless, the Court, for reasons unknown and unexplained, chooses to ignore the issue, thereby visiting unprecedented unfairness on respondents.

Even if I agreed with the limitations period adopted by the Court, I would dissent from Part IV of the Court's opinion. Our prior cases dictate that the federal statute of limitations announced today should not be applied retroactively. I would remand so that the lower courts may determine in the first instance the timeliness of respondents' lawsuit.

JUSTICE KENNEDY, with whom JUSTICE O'CONNOR joins, dissenting.

I am in full agreement with the Court's determination that, under our precedents, a uniform federal statute of limitations is appropriate for private actions brought under § 10(b) of the Securities Exchange Act of 1934 and that we should adopt as a limitations period the 1-year-from-discovery rule Congress employed in various provisions of the 1934 Act. I must note my disagreement, however, with the Court's simultaneous adoption of the 3-year period of repose Congress also employed in a number of the 1934 Act's provisions. This absolute time bar on private § 10(b) suits conflicts with traditional limitations periods for fraud-based actions, frustrates the usefulness of § 10(b) in protecting defrauded investors, and imposes severe practical limitations on a federal implied cause of action that has become an essential component of the protection the law gives to investors who have been injured by unlawful practices.

As the Court recognizes, in the absence of an express limitations period in a federal statute, courts as a general matter should apply the most analogous state limitations period or, in rare cases, no limitations period at all. This rule does not apply, however, "when a rule from elsewhere in federal law clearly provides a closer analogy than available state statutes, and when the federal policies at stake and the practicalities of litigation make that rule a significantly more appropriate vehicle for interstitial lawmaking." DelCostello v. Teamsters, 462 U.S. 151, 172 (1983); see Reed v. United Transportation Union, 488 U.S. 319, 324 (1989). Applying this principle, the Court looks first to the express private rights of action in the 1934 Act itself to find what it believes are the appropriate limitations periods to apply here. One cannot fault the Court's mode of analysis; given that § 10(b) actions are implied under the 1934 Act, it makes sense for us to look to the limitations periods Congress established under the Act. See DelCostello, supra, at 171; United Parcel Service, Inc. v. Mitchell, 451 U.S. 56, 68, n. 4 (1981). That does not relieve us, however, of our obligation to reject a limitations rule that would "frustrate or significantly interfere with federal policies." Reed, 488 U. S., at 327. When determining the appropriate statute of limitations to apply, we must give careful consideration to the policies underlying a federal statute and to the practical difficulties aggrieved parties may have in establishing a violation. Ibid.; Wilson v. Garcia, 471 U.S. 261, 268 (1985):

This is not a case where the Court identifies a specific statute and follows each of its terms. As the Court is careful to note, the 1934 Act does not provide a single limitations period for all private actions brought under its express provisions. Rather, the Act makes three separate and distinct references to statutes of limitations. The Court rejects outright one of these references, a 2-year statute of repose for actions brought under § 16 of the 1934 Act, 15 U. S. C. § 78p(b), and purports to follow the other two. §§ 78i(e), 78r(c). The latter two references employ 1-year, 3-year schemes similar to one the Court establishes here, but each has its own unique wording. The Court does not identify any reasons for finding one to be controlling, so it is unnecessary to engage in close grammatical construction to separate the 1-year discovery period from the 3-year statute of repose.

It is of even greater importance to note that both of the statutes in question relate to express causes of action which in their purpose and underlying rationale differ from causes of action implied under § 10(b). The limitations statutes to which the Court refers apply to strict liability violations or, in the case of § 78i(e), to a rarely used remedy under § 9 of the 1934 Act. See L. Loss, Fundamentals of Securities Regulation 920 (2d ed. 1988). Neither relates to a cause of action of the scope and coverage of an implied action under § 10(b). Nor does either rest on the common-law fraud model underlying most § 10(b) actions.

Section 10(b) provides investors with significant protections from fraudulent practices in the securities markets. Intended as a comprehensive antifraud provision operating even when more specific laws have no application, § 10(b) makes it unlawful to employ in connection with the purchase or sale of any security "any manipulative or deceptive device or contrivance" in violation of the Securities and Exchange Commission's rules. 15 U. S. C. § 78j. Although Congress gave the Commission the primary role in enforcing this section, private § 10(b) suits constitute "an essential tool for enforcement of the 1934 Act's requirements," Basic Inc. v. Levinson, 485 U.S. 224, 231 (1988), and are "`a necessary supplement to Commission action.'" Bateman Eichler, Hill Richards, Inc. v. Berner, 472 U.S. 299, 310 (1985) (quoting J. I. Case Co. v. Borak, 377 U.S. 426, 432 (1964)). We have made it clear that rules facilitating § 10(b) litigation "suppor[t] the congressional policy embodied in the 1934 Act" of combating all forms of securities fraud. Basic, supra, at 245.

The practical and legal obstacles to bringing a private § 10(b) action are significant. Once federal jurisdiction is established, a § 10(b) plaintiff must prove elements that are similar to those in actions for common-law fraud. See Herman & MacLean v. Huddleston, 459 U.S. 375 (1983). Each requires proof of a false or misleading statement or material omission, Santa Fe Industries, Inc. v. Green, 430 U.S. 462 (1977), reliance thereon, Basic, 485 U. S., at 243; cf. id., at 245 (reliance presumed in § 10(b) cases proving "fraud-on-the-market"), damages caused by the wrongdoing, Randall v. Loftsgaarden, 478 U.S. 647, 663 (1986), and scienter on the part of the defendant, Ernst & Ernst v. Hochfelder, 425 U.S. 185 (1976). Given the complexity of modern securities markets, these facts may be difficult to prove.

The real burden on most investors, however, is the initial matter of discovering whether a violation of the securities laws occurred at all. This is particularly the case for victims of the classic fraudlike case that often arises under § 10(b). "[C]oncealment is inherent in most securities fraud cases." American Bar Association, Report of the Task Force on Statute of Limitations for Implied Actions, 41 Bus. Lawyer 645, 654 (1985). The most extensive and corrupt schemes may not be discovered within the time allowed for bringing an express cause of action under the 1934 Act. Ponzi schemes, for example, can maintain the illusion of a profit-making enterprise for years, and sophisticated investors may not be able to discover the fraud until long after its perpetration. Id., at 656. Indeed, in Ernst & Ernst, the alleged fraudulent scheme had gone undetected for over 25 years before it was revealed in a stockbroker's suicide note. 425 U. S., at 189.

The practicalities of litigation, indeed the simple facts of business life, are such that the rule adopted today will "thwart the legislative purpose of creating an effective remedy" for victims of securities fraud. Agency Holding Corp. v. Malley-Duff & Associates, Inc., 483 U.S. 143, 154 (1987). By adopting a 3-year period of repose, the Court makes a § 10(b) action all but a dead letter for injured investors who by no conceivable standard of fairness or practicality can be expected to file suit within three years after the violation occurred. In so doing, the Court also turns its back on the almost uniform rule rejecting short periods of repose for fraud-based actions. In the vast majority of States, the only limitations periods on fraud actions run from the time of a victim's discovery of the fraud. Shapiro & Blauner, Securities Litigation in the Aftermath of In Re Data Access Securities Litigation, 24 New England L. Rev. 537, 549-550 (1989). Only a small minority of States constrain fraud actions with absolute periods of repose, and those that do typically permit actions to be brought within at least five years. See, e. g., Fla. Stat. § 95.11(4)(e) (1991) (5-year period of repose); Ky. Rev. Stat. Ann. § 413.120(11) (Michie 1990) (10-year period of repose); Mo. Rev. Stat. § 516.120(5) (1986) (10-year period of repose). Congress itself has recognized the importance of granting victims of fraud a reasonable time to discover the facts underlying the fraud and to prepare a case against its perpetrators. See, e. g., Interstate Land Sales Full Disclosure Act, 15 U. S. C. § 1711(a)(2) (action may be brought within three years from discovery of violation); Insider Trading and Securities Fraud Enforcement Act of 1988, 15 U. S. C. § 78t-1(b)(4) (action may be brought within five years of the violation). The Court, however, does not.

A reasonable statute of repose, even as applied against fraud-based actions, is not without its merits. It may sometimes be easier to determine when a fraud occurred than when it should have been discovered. But more important, limitations periods in general promote important considerations of fairness. "Just determinations of fact cannot be made when, because of the passage of time, the memories of witnesses have faded or evidence is lost." Wilson, 471 U. S., at 271. Notwithstanding these considerations, my view is that a 3-year absolute time bar is inconsistent with the practical realities of § 10(b) litigation and the congressional policies underlying that remedy. The 1-year-from-discovery rule is sufficient to ensure a fair balance between protecting the legitimate interests of aggrieved investors, yet preventing stale claims. In the extreme case, moreover, when the period between the alleged fraud and its discovery is of extraordinary length, courts may apply equitable principles such as laches should it be unfair to permit the claim. See DelCostello, 462 U. S., at 162; Holmberg v. Armbrecht, 327 U.S. 392 (1946). A 3-year absolute bar on § 10(b) actions simply tips the scale too far in favor of wrongdoers.

The Court's decision today forecloses any means of recovery for a defrauded investor whose only mistake was not discovering a concealed fraud within an unforgiving period of repose. As fraud in the securities markets remains a serious national concern, Congress may decide that the rule announced by the Court today should be corrected. But even if prompt congressional action is taken, it will not avail defrauded investors caught by the Court's new and unforgiving rule, here applied on a retroactive basis to a pending action.

With respect, I dissent and would remand with instructions that a § 10(b) action may be brought at any time within one year after an investor discovered or should have discovered a violation. In any event, I would permit the litigants in this case to rely upon settled Ninth Circuit precedent as setting the applicable limitations period in this case, and join JUSTICE O'CONNOR's dissenting opinion in full.