Argued April 21, 1997 Decided June 19, 1997
The Racketeer Influenced and Corrupt Organizations Act (RICO) makes it a crime "to conduct" an "enterprise's affairs through a pattern of racketeering activity." 18 U.S.C. § 1962(c). A "pattern" requires at least two acts of racketeering activity, the last of which occurred within 10 years after the commission of a prior act. § 1961(5). A person injured by a violation of RICO's criminal provisions may recover treble damages and attorney's fees in a civil RICO action, § 1964(c), but civil actions are subject to the 4-year limitations period in § 4B of the Clayton Act — the statute of limitations governing private civil antitrust actions seeking treble damages, Agency Holding Corp. v. Malley-Duff Associates, Inc., 483 U.S. 143, 156. The petitioners Klehr filed a civil RICO action against respondents (hereinafter Harvestore) in August 1993, claiming that their injury began in 1974, when they purchased a Harvestore-brand silo for their dairy farm based on Harvestore's false representations that it would prevent moldy and fermented cattle feed, thereby producing healthier cows, more milk, and higher profits. In fact, the feed became moldy and fermented and both milk production and profits declined. They added that Harvestore committed other predicate acts, consisting of repeated misrepresentations to the Klehrs and to others, and sales to others, over many years. Harvestore moved to dismiss on the ground that the limitations period had run because the Klehrs' claim had accrued before August 1989, and no special legal doctrine applied to toll the running of the limitations period or to estop Harvestore from asserting a statute of limitations defense. The Klehrs responded that because Harvestore had taken affirmative steps to conceal its fraud, they did not become sufficiently suspicious to investigate the silo and to discover the mold until 1991. The District Court found the Klehrs' lawsuit untimely. The Eighth Circuit affirmed, holding that a civil RICO action accrues as soon as the plaintiff discovers, or reasonably should discover, both the existence and source of his injury and that the injury is part of a pattern; and that the Klehrs had suffered one single, continuous injury sometime in the 1970s which they should have discovered well before August 1989. The Circuit refused to toll the running of the statute on a "fraudulent concealment" theory because, among other things, the Klehrs had not been sufficiently diligent in discovering their claim. Like the Eighth Circuit, some Circuits apply an "injury and pattern discovery" civil RICO accrual rule; others apply an "injury discovery" rule, under which the statute begins to run when the plaintiff knows or reasonably should know of his injury; and the Third Circuit applies a "last predicate act" rule, under which the statute begins to run when the plaintiff knows or reasonably should know of the last injury or last predicate act in the pattern, whether or not the plaintiff himself has suffered any injury from that last act.
1. The "last predicate act" rule is not an appropriate interpretation of RICO. Pp. 186-193.
(a) Only the Third Circuit's accrual rule can help the Klehrs. For purposes of assessing its lawfulness, this Court assumes that the rule means that as long as Harvestore committed one predicate act within the limitations period, the Klehrs can recover, not just for any harm caused by that late-committed act, but for all the harm caused by all the acts that make up the total "pattern"; that the Klehrs can show at least one such late-committed act; and that they are knowledgeable about the pattern. Pp. 186-187.
(b) The rule is unlawful for two reasons. First, because a series of predicate acts can continue indefinitely, it creates a longer limitations period than Congress could have contemplated, in conflict with a basic objective — repose — underlying limitations periods. See, e.g., Wilson v. Garcia, 471 U.S. 262, 271. Civil RICO has no compensatory objective warranting so significant an extension of the limitations period, and civil RICO's further purpose — encouraging potential private plaintiffs diligently to investigate, see Malley-Duff, 483 U.S., at 151 — suggests the contrary. RICO's criminal limitations period, which runs from the most recent predicate act, does not provide an apt analogy for civil RICO actions. Id., at 155-156. Second, the rule is inconsistent with § 4B of the Clayton Act, under which "a cause of action accrues . . . when a defendant commits an act that injures a plaintiff's business." Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 338. The Clayton Act analogy is generally useful in civil RICO cases, since Congress consciously patterned civil RICO after that Act, and since, by the time civil RICO was enacted, the Clayton Act's accrual rule was well-established. The Clayton Act accrual rule may not apply without modification in every civil RICO case. However, in this case the petitioners knew of the facts underlying their cause of action, and thus the Clayton Act rule makes clear precisely where, and how, the Third Circuit's rule goes too far. The Klehrs invoke the "separate accrual" civil RICO rule adopted by some Circuits, which is similar to the "continuing violation" doctrine in antitrust, in that the commission of a separate, new predicate act within the 4-year limitations period permits a plaintiff to recover for the additional damages that act caused. Under the separate accrual rule, however, the plaintiff cannot use an independent, new act as a bootstrap to recover for injuries caused by other predicate acts that took place outside the limitations period. See, e.g., Grimmett v. Brown, 75 F.3d 506, 513. Thus acts taking place after August 1989 do not help the Klehrs, for they have not shown any additional damages, and the Third Circuit rule is incorrect insofar as it would allow the presence of a new act to help them recover for injuries caused by pre-1989 acts. This case also does not present the kind of special circumstance in which courts might permit plaintiffs to recover for injuries that were so speculative or unprovable at the time of Harvestore's unlawful act that starting the limitations period when the act first caused injury would have left the Klehrs without relief. Zenith, supra, at 339-340, distinguished. Pp. 187-191.
(c) Resolving the conflicts among the various discovery accrual rules used by other Circuits would not affect the outcome of this case, as the petitioners' civil RICO claim is barred under the most liberal accrual rule, as applied by the Eighth Circuit. There is no clear or obvious error in the Eighth Circuit's application of its "injury and pattern discovery" rule and it is beyond the scope of the writ to reconsider whether the Klehrs reasonably should have discovered the silo's flaws before 1989. Pp. 191-193.
2. A plaintiff who is not reasonably diligent in trying to discover his civil RICO cause of action may not rely upon "fraudulent concealment" to toll the limitations period or to estop a defendant from asserting a limitations defense. This requirement is uniformly supported by relevant authority in the related antitrust context, where the "fraudulent concealment" doctrine is invoked fairly often. And while those courts that do not require "reasonable diligence" in contexts other than antitrust cases have said that the doctrine is concerned only with defendants' behavior, that is not the case with respect to antitrust or civil RICO. In both of these contexts private civil actions seek not only to compensate victims but also to encourage those victims diligently to investigate and thereby to uncover unlawful activity. See Malley-Duff, supra, at 151. The Klehrs' fact-based question whether the Eighth Circuit properly applied the "due diligence" requirement to the evidentiary materials before it is beyond the scope of this Court's writ. Pp. 193-196.
87 F.3d 231, affirmed.
BREYER, J., delivered the opinion of the Court, in which REHNQUIST, C.J., and STEVENS, O'CONNOR, KENNEDY, SOUTER, and GINSBURG, JJ., joined. SCALIA, J., filed an opinion concurring in part and concurring in the judgment, in which THOMAS, J., joined, post, p. 196.
Charles A. Bird argued the cause for petitioners. With him on the briefs were Mary R. Vasaly, Michael C. McCarthy, and Malcolm McCune.
Bruce J. Ennis, Jr., argued the cause for respondents. With him on the brief were Frederick W. Morris, Blake Shepard, Jr., Jeffrey E. Grell, Nory Miller, and Kathleen M. Massey.
Briefs of amici curiae urging affirmance were filed for the National Association of Manufacturers by Alfred W. Cortese, Jr., Daniel I. Prywes, Michael F. Wasserman, Jan S. Amundson, and Quentin Riegel; Richard B. McNamara, Gregory A. Holmes, Stephanie A. Bray, Martin J. Oberman, Alice W. Ballard, Michael M. Baylson, Charles Barnhill, Jr., Judson Miner, and Edward R. Garvey filed a brief for Plaintiffs' Executive Committee, MDL No. 1069 et al. as amici curiae urging reversal. and for the Washington Legal Foundation et al. by Daniel J. Popeo and Richard A. Samp.
Philip Allen Lacovara, Evan M. Tager, and Phillip E. Stano filed a brief for the American Council of Life Insurance et al. as amici curiae.
The petition in this case asked us to consider two aspects of "statute of limitations" law. One concerns the date upon which a civil RICO action accrues and the limitations period starts to run. The other concerns "fraudulent concealment," a doctrine that extends the time for a plaintiff to file suit. In respect to the first, we focus upon, and disapprove, an accrual rule followed in the Third Circuit called the "last predicate act" rule. In respect to the second, we hold that a plaintiff may not rely upon "fraudulent concealment" unless he has been reasonably diligent in trying to discover his cause of action.
The Racketeer Influenced and Corrupt Organizations Act (RICO), 18 U.S.C. § 1961-1968, among other things, makes it a crime "to conduct" an "enterprise's affairs through a pattern of racketeering activity." § 1962(c). The phrase "racketeering activity" is a term of art defined in terms of activity that violates other laws, including more than 50 specifically mentioned federal statutes, which forbid, for example, murder-for-hire, extortion, and various kinds of fraud. § 1961(1). The word "pattern" is also a term of art defined to require "at least two acts of racketeering activity . . . the last of which occurred within ten years . . . after the commission of a prior act of racketeering activity." § 1961(5).
A special RICO provision — commonly known as civil RICO — permits "[a]ny person injured in his business or property by reason of a violation" of RICO's criminal provisions to recover treble damages and attorney's fees. § 1964(c). RICO does not say what limitations period governs the filing of civil RICO claims. But in Agency Holding Corp. v. Malley-Duff Associates, Inc., 483 U.S. 143, 156 (1987), this Court held that civil RICO actions are subject to the 4-year limitations period contained in § 4B of the Clayton Act (Antitrust), as added 69 Stat. 283, and amended, 15 U.S.C. § 15b — the statute of limitations that governs private civil antitrust actions seeking treble damages.
Marvin and Mary Klehr, the petitioners here, are dairy farmers. They filed this civil RICO action on August 27, 1993, claiming that A. O. Smith Corporation, and A. O. Smith Harvestore Products, Inc. (whom we shall simply call "Harvestore") had committed several acts of mail and wire fraud, 18 U.S.C. § 1341, 1343, thereby violating RICO and causing them injury. Their injury, they said, began in 1974, when Harvestore sold them a special "Harvestore" brand silo, which they used for storing cattle feed. The Klehrs alleged that they bought the silo in reliance on Harvestore's representations, made through advertisements and a local dealer, that the silo would limit the amount of oxygen in contact with the silage, thus preventing moldy and fermented feed, and thereby producing healthier cows, more milk, and higher profits. The representations, they claim, were false; the silo did not keep oxygen away from the feed, the feed became moldy and fermented, the cows ate the bad feed, and milk production and profits went down. They add that Harvestore committed other acts consisting primarily of additional representations made to them and to others and sales made to others — over a period of many years after 1974.
Harvestore, pointing out that the Klehrs had filed suit almost 20 years after they had bought the silo, moved to dismiss the lawsuit on the ground that the limitations period had long since run. The Klehrs could not file suit, Harvestore said, unless their claim had accrued within the four years prior to filing, i.e., after August 25, 1989, or unless some special legal doctrine nonetheless tolled the running of the limitations period or estopped Harvestore from asserting a statute of limitations defense. See Holmberg v. Armbrecht, 327 U.S. 392, 396-397 (1946); Bailey v. Glover, 21 Wall. 342, 349-350 (1875); Cada v. Baxter Healthcare Corp., 920 F.2d 446, 450-451 (CA7 1990), cert. denied, 501 U.S. 1261 (1991).
The Klehrs responded by producing evidentiary material designed to support a legal justification for the late filing. Essentially they claimed that Harvestore had covered up its fraud — preventing them from noticing the silo's malfunction — for example, by means of an unloading device that hid the mold by chopping up the feed instantly as it emerged; through continued dealer misrepresentations; with advertisements that tried to convince farmers that warm, brown, molasses-smelling feed was not fermented feed, but good feed; and even by hanging on the silo itself a plaque that said:
"DANGER DO NOT ENTER NOT ENOUGH OXYGEN TO SUPPORT LIFE"
Not until 1991, say the Klehrs, did they become sufficiently suspicious to investigate the silo, at which time, by opening the silo wall and chopping through the feed with an ice chisel, they discovered "`mold hanging all over the silage.'" Brief for Petitioners 16.
The District Court, after examining the Klehrs' evidence, found their lawsuit untimely. The Eighth Circuit affirmed the dismissal, and said that a civil RICO action accrues
"`as soon as the plaintiff discovers, or reasonably should have discovered, both the existence and source of his injury and that the injury is part of a pattern.'" 87 F.3d 231, 238 (1996) (quoting Association of Commonwealth Claimants v. Moylan, 71 F.3d 1398, 1402 (CA8 1995)).
After examining the Klehrs' evidence de novo, the Circuit held that they failed to satisfy the standard. It said they had suffered "one single, continuous injury . . . sometime in the 1970s"; and that they should have discovered "the existence and source of their injury," as well as any related "pattern," well before August 1989. 87 F.3d, at 239. The Circuit refused to find "fraudulent concealment" because, among other things, the Klehrs had not been sufficiently "diligen[t]." Id., at 238, 239, n. 11.
We granted certiorari in this case to consider the Klehrs' claim in light of a split of authority among the Courts of Appeals. Two other Circuits, like the Eighth Circuit here, have applied forms of an "injury and pattern discovery" civil RICO accrual rule. Bivens Gardens Office Building, Inc. v. Barnett Bank, 906 F.2d 1546, 1554-1555 (CA11 1990), cert. denied, 500 U.S. 910 (1991); Bath v. Bushkin, Gaims, Gaines Jonas, 913 F.2d 817, 820 (CA10 1990). Other Circuits have applied forms of an "injury discovery" rule, i.e., without the "pattern." See Grimmett v. Brown, 75 F.3d 506, 511 (CA9 1996), cert. dism'd as improvidently granted, 519 U.S. 233 (1997); McCool v. Strata Oil Co., 972 F.2d 1452, 1464-1465 (CA7 1992); Rodriguez v. Banco Central Corp., 917 F.2d 664, 665-666 (CA1 1990); Bankers Trust Co. v. Rhoades, 859 F.2d 1096, 1102 (CA2 1988), cert. denied, 490 U.S. 1007 (1989); Pocahontas Supreme Coal Co. v. Bethlehem Steel Corp., 828 F.2d 211, 220 (CA4 1987); see also Riddell v. Riddell Washington Corp., 866 F.2d 1480, 1489-1490 (CADC 1989) (assuming, but not deciding, that injury discovery rule applies). One court, the Third Circuit, has applied a "last predicate act" rule, which we shall discuss below. We also agreed to decide the Klehrs' argument that "reasonable diligence" is not a necessary component of the doctrine of "fraudulent concealment."
For reasons we shall describe, we affirm the judgment of the Court of Appeals.
We shall first discuss the Third Circuit's accrual rule — the "last predicate act" rule — for it is the only accrual rule that can help the Klehrs. Like the Eighth Circuit, the Third Circuit believes that the limitations period starts to run when a plaintiff knew or should have known that the RICO claim (including a "pattern of racketeering activity") existed, but the Third Circuit has added an important exception, which it states as follows:
"[If], as a part of the same pattern of racketeering activity, there is further injury to the plaintiff or further predicate acts occur, . . . the accrual period shall run from the time when the plaintiff knew or should have known of the last injury or the last predicate act which is part of the same pattern of racketeering activity. The last predicate act need not have resulted in injury to the plaintiff but must be part of the same pattern." Keystone Ins. Co. v. Houghton, 863 F.2d 1125, 1130 (1988).
For purposes of assessing the rule's lawfulness, we assume, as do the Klehrs, that this rule means that as long as Harvestore committed one predicate act within the limitations period ( i.e., the four years preceding suit), the Klehrs can recover, not just for any added harm caused them by that late-committed act, but for all the harm caused them by all the acts that make up the total "pattern." We also assume that they can show at least one such late-committed act. Finally, we note that the point of difference between the Third Circuit, and the other Circuits, has nothing to do with the plaintiff's state of mind or knowledge. It concerns only the accrual consequences of a late-committed act. Consequently, we can consider the merits of the rule on the simplifying assumption that the plaintiff is perfectly knowledgeable.
We conclude that the Third Circuit's rule is not a proper interpretation of the law. We have two basic reasons. First, as several other Circuits have pointed out, the last predicate act rule creates a limitations period that is longer than Congress could have contemplated. Because a series of predicate acts (including acts occurring at up to 10-year intervals) can continue indefinitely, such an interpretation, in principle, lengthens the limitations period dramatically. It thereby conflicts with a basic objective — repose — that underlies limitations periods. See Wilson v. Garcia, 471 U.S. 261, 271 (1985) (citing Adams v. Woods, 2 Cranch 336, 342 (1805)); Crown, Cork Seal Co. v. Parker, 462 U.S. 345, 352 (1983). Indeed, the rule would permit plaintiffs who know of the defendant's pattern of activity simply to wait, "sleeping on their rights," ibid., as the pattern continues and treble damages accumulate, perhaps bringing suit only long after the "memories of witnesses have faded or evidence is lost." Wilson, supra, at 271. We cannot find in civil RICO a compensatory objective that would warrant so significant an extension of the limitations period, and civil RICO's further purpose — encouraging potential private plaintiffs diligently to investigate, see Malley-Duff, 483 U.S., at 151 — suggests the contrary.
We recognize that RICO's criminal statute of limitations runs from the last, i.e., the most recent, predicate act. But there are significant differences between civil and criminal RICO actions, and this Court has held that criminal RICO does not provide an apt analogy. Id., at 155-156 (declining to apply criminal RICO's 5-year statute of limitations to civil RICO actions and noting "competing equities unique to civil RICO actions or, indeed, any other federal civil remedy").
Second, the Third Circuit rule is inconsistent with the ordinary Clayton Act rule, applicable in private antitrust treble damage actions, under which "a cause of action accrues and the statute begins to run when a defendant commits an act that injures a plaintiff's business." Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 338 (1971); Connors v. Hallmark Son Coal Co., 935 F.2d 336, 342, n. 10 (CADC 1991); 1 C. Corman, Limitation of Actions § 188.8.131.52, p. 449 (1991) (hereinafter Corman); 2 P. Areeda H. Hovenkamp, Antitrust Law ¶ 338b, p. 145 (rev. ed. 1995) (hereinafter Areeda). We do not say that a pure injury accrual rule always applies without modification in the civil RICO setting in the same way that it applies in traditional antitrust cases. For example, civil RICO requires not just a single act, but rather a "pattern" of acts. Furthermore, there is some debate as to whether the running of the limitations period depends on the plaintiff's awareness of certain elements of the cause of action. As we said earlier, however, for purposes of evaluating the Third Circuit's rule we can assume knowledgeable parties. Hence the special problems associated with a discovery rule, see Part II-B, infra, are not at issue. And we believe, in these circumstances, the Clayton Act analogy is helpful.
In Malley-Duff, this Court indicated why the analogy is useful. It concluded
"that there is a need for a uniform statute of limitations for civil RICO, that the Clayton Act clearly provides a far closer analogy than any available state statute, and that the federal policies that lie behind RICO and the practicalities of RICO litigation make the selection of the 4-year statute of limitations for Clayton Act actions . . . the most appropriate limitations period for RICO actions." 483 U.S., at 156 (citing 15 U.S.C. § 15b).
The Court left open the accrual question. But it did not rule out the use of a Clayton Act analogy. As the Court has explained, Congress consciously patterned civil RICO after the Clayton Act. 483 U.S., at 150-151 (comparing 15 U.S.C. § 15(a) with 18 U.S.C. § 1964( c)); see also Sedima, S. P. R. L. v. Imrex Co., 473 U.S. 479, 489 (1985). And by the time civil RICO was enacted, the Clayton Act's accrual rule was well established. See Crummer Co. v. DuPont, 223 F.2d 238, 247-248 (CA5), cert. denied, 350 U.S. 848 (1955); Foster Kleiser Co. v. Special Site Sign Co., 85 F.2d 742, 750-751 (CA9 1936), cert. denied, 299 U.S. 613 (1937); Bluefields S. S. Co. v. United Fruit Co., 243 F. 1, 20 (CA3 1917).
The Clayton Act helps here because it makes clear precisely where, and how, the Third Circuit's rule goes too far. Antitrust law provides that, in the case of a "continuing violation," say a price-fixing conspiracy that brings about a series of unlawfully high priced sales over a period of years, "each overt act that is part of the violation and that injures the plaintiff," e.g., each sale to the plaintiff, "starts the statutory period running again, regardless of the plaintiff's knowledge of the alleged illegality at much earlier times." 2 Areeda, ¶ 338b, at 145 (footnote omitted); see also Zenith, supra, at 338; Hanover Shoe, Inc. v. United Shoe Machinery Corp., 392 U.S. 481, 502, n. 15 (1968); DXS, Inc. v. Siemens Medical Systems, Inc., 100 F.3d 462, 467 (CA6 1996). But the commission of a separate new overt act generally does not permit the plaintiff to recover for the injury caused by old overt acts outside the limitations period. Zenith, supra, at 338; Pennsylvania Dental Assn. v. Medical Serv. Assn., 815 F.2d 270, 278 (CA3), cert. denied, 484 U.S. 851 (1987); Hennegan v. Pacifico Creative Serv., Inc., 787 F.2d 1299, 1300 (CA9), cert. denied, 479 U.S. 886 (1986); National Souvenir Center v. Historic Figures, Inc., 728 F.2d 503, 509 (CADC), cert. denied sub nom. C. M. Uberman Enterprises, Inc. v. Historical Figures, Inc., 469 U.S. 825 (1984); Imperial Point Colonnades Condominium, Inc. v. Mangurian, 549 F.2d 1029, 1034-1035 (CA5 1977); Crummer Co., supra, 247-248. Compare 2 Areeda ¶ 338b, at 149.
Similarly, some Circuits have adopted a "separate accrual" rule in civil RICO cases, under which the commission of a separable, new predicate act within a 4-year limitations period permits a plaintiff to recover for the additional damages caused by that act. But, as in the antitrust cases, the plaintiff cannot use an independent, new predicate act as a bootstrap to recover for injuries caused by other earlier predicate acts that took place outside the limitations period. See, e.g., Grimmett, 75 F.3d, at 512-514; McCool v. Strata Oil Co., 972 F.2d, at 1465-1466, and n. 10; Bivens Gardens Office Building, Inc. v. Barnett Bank, 906 F.2d, at 1552, n. 9; State Farm Mut. Auto. Ins. Co. v. Ammann, 828 F.2d 4, 5 (CA9 1987) (Kennedy, J., concurring). But see Bingham v. Zolt, 66 F.3d 553, 560 (CA2 1995) (citing Bankers Trust, 859 F.2d, at 1103). Thus the Klehrs may point to new predicate acts that took place after August 1989, such as sales to other farmers or the printing of new Harvestore advertisements. But that fact does not help them, for, as the Court of Appeals pointed out, they have not shown how any new act could have caused them harm over and above the harm that the earlier acts caused. 87 F.3d, at 239. Nor can the presence of the new act help them recover for the injuries caused by pre-1989 acts, for it is in this respect that we find the Third Circuit's rule incorrect.
Plaintiffs also point to Zenith, a case in which this Court considered antitrust damages that were so "speculative" or "unprovable," 401 U.S., at 339, at the time of a defendant's unlawful act (and plaintiff's initial injury) that to follow the normal accrual rule (starting the limitations period at the point the act first causes injury) would have left the plaintiff without relief. This Court held that, in such a case, a claim for the injuries that had been speculative would accrue when those injuries occurred, even though the act that caused them had taken place more than four years earlier. Id., at 339-340. This case does not help the plaintiffs here, however, for their injuries — the harm to their farm — have always been specific and calculable.
We recognize that our holding in Part II-A does not resolve other conflicts among the Circuits. For example, the Circuits have applied "discovery" accrual rules, which extend accrual periods for plaintiffs who could not reasonably obtain certain key items of information. The use of a discovery rule may reflect the fact that a high percentage of civil RICO cases, unlike typical antitrust cases, involve fraud claims. See Sedima, 473 U.S., at 499, n. 16 (most civil RICO claims involve underlying fraud offense); 1 A. Mathews, A. Weissman, J. Sturc, Civil RICO Litigation, p. 1-6 (2d ed. 1992) (citing Report of the Ad Hoc Civil RICO Task Force of the ABA Section of Corporation, Banking and Business Law 243 (1985)) (as of 1985, approximately 90% of civil RICO cases resulting in a published decision involved mail, wire, or securities fraud as a predicate offense); cf. Connors, 935 F.2d, at 342 (federal courts generally apply discovery accrual rule when statute does not call for a different rule); 1 Corman § 184.108.40.206, at 449 (same). Moreover, different Circuits have applied discovery accrual rules that differ, one from the other, in important ways. Compare, e.g., Bankers Trust, supra, at 1103 (civil RICO cause of action accrues when the plaintiff discovers or should have discovered his injury) with 87 F.3d, at 238 (civil RICO cause of action accrues when, in addition, plaintiff discovers or should have discovered the "source" of injury and a "pattern").
We further realize that, contrary to our assumption in Part II-A, supra (where we discussed a legal issue in respect to which knowledge was irrelevant), the Klehrs did claim that they lacked knowledge of the faulty silo — the "source" of their injury. But that particular "lack of knowledge" claim does not require us to consider the various "discovery rule" differences among the Circuits, because the Klehrs failed the "knowledge" test that favors them the most — the Eighth Circuit's "injury plus source plus pattern" rule. That rule would have found the Klehrs' action timely had it not been the case that the Klehrs reasonably "should have discovered" all of those elements prior to 1989. 87 F.3d, at 239. If the Klehrs cannot fit their case through the Eighth Circuit's larger hole, they cannot squeeze it through a smaller one.
In addition, the major difference among the Circuits — whether a discovery rule includes knowledge about a "pattern" — is clearly not at issue here. Harvestore marketed and sold its "oxygen-limiting" silos for many years before the Klehrs purchased theirs, and the Klehrs have not claimed lack of knowledge of a "pattern." Nor has anyone argued any other legal differences among the Circuits' various tests that would affect the outcome in this case.
In these circumstances, we believe we should not consider differences among the various discovery accrual rules used by the Circuits. The legal questions involved may be subtle and difficult. Compare id., at 238 (claim accrues with discovery of existence and source of injury, plus pattern) with Bivens Gardens, 906 F.2d, at 1554 (claim accrues with discovery of injury and pattern); see also Cada, 920 F.2d, at 451 (describing differences among various discovery rules and doctrines of "equitable tolling" and "equitable estoppel"). And the facts of this case do not force focused argument as to how the traditional Clayton Act "injury" accrual rule, principles of equitable tolling, and doctrines of equitable estoppel should interact in circumstances where the application of one, or another, of these different limitations doctrines would make a significant legal difference. To say this is not, as the concurrence claims, to advocate a "mix-and-match" statute of limitations theory. Rather, it is to recognize that the Clayton Act's express statute of limitations does not necessarily provide all the answers. We shall, at the very least, wait for a case that clearly presents these, or related issues, providing an opportunity for full argument, before we attempt to resolve them.
Finally, the Klehrs have asked us to review the Eighth Circuit's application of its rule in this case. Doing so would involve examining an evidentiary record of several thousand pages to determine the validity of the independent conclusion of each of two lower courts that the Klehrs should reasonably have discovered the silo's flaws before 1989 (and that a reasonable factfinder could not conclude to the contrary). That conclusion is highly fact-based, depending not only upon how much mold the Klehrs noticed in their silage and when, but also upon such matters as the effect of the Klehrs' failure to consult the herd performance records they were continuously sent, and whether their having done so would have led them to tell veterinarians a more revealing story, to question Harvestore's representatives more fully, or to investigate the silo sooner. See 87 F.3d, at 234. We have no reason to believe that there is any very obvious or exceptional error below. And our writ of certiorari commits us to decide only whether the purely legal question of whether or not a claim accrues "where the Respondent continues to commit predicate acts" in the 4-year period immediately preceding suit. Pet. for Cert. i. We have answered that question in Part II-A. And we shall not go beyond the writ's question to reexamine the fact-based rule-application issue that the Klehrs now raise, and which the Eighth Circuit decided in Harvestore's favor.
Our writ of certiorari contained one further question, namely whether
"affirmative continuing acts of fraud . . . coupled with active cover up of the fraud, act to equitably toll the statute of limitations . . . whether or not Petitioners have exercised reasonable diligence to discover their claim." Ibid. (emphasis added).
This question refers to the doctrine of "fraudulent concealment," which some courts have said "equitably tolls" the running of a limitations period, see, e.g., Grimmett, 75 F.3d, at 514, while other courts have said it is a form of "equitable estoppel," see, e.g., Wolin v. Smith Barney Inc., 83 F.3d 847, 852 (CA7 1996). Regardless, the question presented here focuses upon a relevant difference among the Circuits in respect to the requirement of "reasonable diligence" on the part of the plaintiff. Some Circuits have held that when a plaintiff does not, in fact, know of a defendant's unlawful activity, and when the defendant takes "affirmative steps" to conceal that unlawful activity, those circumstances are sufficient to toll the limitations period (or to "estop" the defendant from asserting a limitations defense) irrespective of what the plaintiff should have known. See, e.g., id., at 852-853. Other courts have held that a plaintiff who has not exercised reasonable diligence may not benefit from the doctrine. See, e.g., Wood v. Carpenter, 101 U.S. 135, 143 (1879); Bailey, 21 Wall., at 349-350; J. Geils Band Employee Benefit Plan v. Smith Barney Shearson, Inc., 76 F.3d 1245, 1252-1255 (CA1 1996) (diligence required for fraudulent concealment under federal law); Urland v. Merrell-Dow Pharmaceuticals, Inc., 822 F.2d 1268, 1273-1274 (CA3 1987) (same with respect to Pennsylvania law); see also 2 Corman § 9.7.1, at 56-57, 60-61, 64-66.
We limit our consideration of the question to the context of civil RICO. In that context, we conclude that "reasonable diligence" does matter, and a plaintiff who is not reasonably diligent may not assert "fraudulent concealment." We reach this conclusion for two reasons. First, in the related antitrust context, where the "fraudulent concealment" doctrine is invoked fairly often, relevant authority uniformly supports the requirement. Professor Areeda says, for example, that the "[t]he concealment requirement is satisfied only if the plaintiff shows that he neither knew nor, in the exercise of due diligence, could reasonably have known of the offense." 2 Areeda ¶ 338, at 152; see also I. Scher, Antitrust Adviser § 10.27, p. 10-62 (4th ed. 1995). We have found many antitrust cases that say the same, and none that says the contrary. See, e.g., Conmar Corp. v. Mitsui Co., 858 F.2d 499, 502 (CA9 1988), cert. denied sub nom. VSL Corp. v. Conmar Corp., 488 U.S. 1010 (1989); Texas v. Allan Constr. Co., 851 F.2d 1526, 1533 (CA5 1988); Pinney Dock Transport Co. v. Penn Central Corp., 838 F.2d 1445, 1465 (CA6), cert. denied sub nom. Pinney Dock Transport Co. v. Norfolk Western R. Co., 488 U.S. 880 (1988); New York v. Hendrickson Bros., Inc., 840 F.2d 1065, 1083 (CA2), cert. denied, 488 U.S. 848 (1988); Berkson v. Del Monte Corp., 743 F.2d 53, 56 (CA1 1984), cert. denied, 470 U.S. 1056 (1985); Charlotte Telecasters, Inc. v. Jefferson-Pilot Corp., 546 F.2d 570, 574 (CA4 1976).
Second, those courts that do not require "reasonable diligence" have said that the "fraudulent concealment" doctrine seeks to punish defendants for affirmative, discrete acts of concealment; the behavior of plaintiffs is consequently irrelevant. See Wolin, supra, at 852; Robertson v. Seidman Seidman, 609 F.2d 583, 593 (CA2 1979); cf. Urland, supra, at 1280-1281 (Becker, J., dissenting). Whether or not that is so in the legal contexts at issue in those cases (which were not antitrust cases), it is not so in respect either to antitrust or to civil RICO. Rather, in both of those latter contexts private civil actions seek not only to compensate victims but also to encourage those victims themselves diligently to investigate and thereby to uncover unlawful activity. See Malley-Duff, 483 U.S., at 151. That being so, we cannot say that the "fraudulent concealment" is concerned only with the behavior of defendants. For that reason, and in light of the consensus of authority, we conclude that "fraudulent concealment" in the context of civil RICO embodies a "due diligence" requirement.
In their brief on the merits, petitioners have asked us to examine whether the Eighth Circuit properly applied the "due diligence" requirement to the evidentiary materials before it. That fact-based question, however, is beyond the scope of our writ; and for reasons similar to those discussed earlier, see supra, at 193, we shall not consider it.
The judgment of the Court of Appeals is
Twice this Term we have received full briefing and heard oral argument on the question of when a civil RICO cause of action accrues; when we rise for our summer recess, the question will remain unanswered. We did not reach it in Grimmett v. Brown, 519 U.S. 233 (1997), because we dismissed the writ of certiorari as improvidently granted. And we do not reach it today for no particular reason except timidity — declining to say what the correct accrual rule is, but merely rejecting the only one of the four candidates under which these petitioners could recover. We thus leave reduced but unresolved the well-known split in authority that prompted us to take this case. There will remain in effect, in some Circuits, one of the three remaining accrual rules — the one that their Courts of Appeals or District Courts have adopted; in the remaining Circuits litigants will have to guess which of the three to follow; and in all of the Circuits no one will know for sure which rule is right — until, at some future date, we receive briefing and argument a third or fourth time, and finally summon up the courage to "unravel," as one commentator has put it, "the mess that characterizes civil RICO accrual decisions," Abrams, Crime Legislation and the Public Interest: Lessons from Civil RICO, 50 SMU L. Rev. 33, 70 (1996).
The Court's opinion could be read to suggest that there are only three different possible accrual rules — last predicate act, injury discovery, and injury and pattern discovery. See ante, at 185-186, 191-193. In fact, as is alluded to in its rejection of the Third Circuit's last predicate act rule, see ante, at 188-189, there is a fourth accrual rule — the Clayton Act "injury" rule.
Worse still, the reason the Court gives for regarding the accrual issue as too complex ("subtle and difficult," ante, at 192) to be decided on only the second try is a reason that implicates the merits, and that in my view gets the merits wrong. One cannot, the Court says, leap impetuously to the conclusion that the antitrust "injury" accrual rule applies, rather than a "discovery" accrual rule, because civil RICO cases are unlike antitrust cases, in that "a high percentage" of them "involve fraud claims." Ante, at 191. This erases, it seems to me, the one clear path back out of the current forest of confusion, which is the proposition that RICO is similar to the Clayton Act. This is the proposition that caused us to adopt the Clayton Act statute of limitations in the first place, specifically rejecting the argument the Court now finds plausible, that the preponderance of fraud claims under RICO makes the Clayton Act an inappropriate model. We said the similarity was close enough: "Although the large majority of civil RICO complaints use [fraud] as the required predicate offenses, a not insignificant number of complaints allege criminal activity of a type generally associated with professional crimes such as arson, bribery, theft and political corruption." Agency Holding Corp. v. Malley-Duff Associates, Inc., 483 U.S. 143, 149 (1987) (rejecting for this reason the use of state-law fraud statutes of limitations). Elsewhere in today's opinion, curiously enough, the Court is quite willing to say that what is good for antitrust is good for RICO — even with respect to a matter much more intimately connected with fraud than the accrual rule, namely, whether invocation of the "fraudulent concealment" rule requires "reasonable diligence" on the plaintiff's part. On this point the Court finds arguments taken from "the related antitrust context" entirely persuasive. Ante, at 194. (Apart from that illogical reliance, it seems to me also illogical even to resolve the question of whether a statute should be tolled by fraudulent concealment without having resolved the antecedent question of when the statute begins to run.) Similarly, the Court relies heavily on the antitrust injury accrual rule in its analysis rejecting the Third Circuit's last predicate act rule. Ante, at 188-191.
I would resolve the Circuit split we granted certiorari to consider, and would hold that, of the four main accrual rules (injury, injury discovery, injury and pattern discovery, and last predicate act), the appropriate accrual rule is the Clayton Act "injury" rule — the "cause of action accrues and the statute begins to run when a defendant commits an act that injures a plaintiff's business." Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 338 (1971) (referring, of course, to "an act" that violates the governing statute.) In Malley- Duff, we held that the appropriate statute of limitations for civil RICO actions is the 4-year limitations period found in the Clayton Act. We reasoned that "RICO was patterned after the Clayton Act," 483 U.S., at 150, and that the purpose, structure, and aims of the two schemes were quite similar, id., at 151-152. Although we expressly acknowledged in Malley-Duff that we "ha[d] no occasion to decide the appropriate time of accrual for a RICO claim," id., at 157, it takes no profound analysis to figure out what that decision must be. "Presumably the accrual standards developed by the lower federal courts in . . . civil antitrust litigation should be equally applicable to civil enforcement RICO actions." 1 C. Corman, Limitation of Actions § 220.127.116.11, pp. 447-448 (1991).
"Both RICO and the Clayton Act are designed to remedy economic injury by providing for the recovery of treble damages, costs, and attorney's fees. Both statutes bring to bear the pressure of `private attorneys general' on a serious national problem for which public prosecutorial resources are deemed inadequate; the mechanism chosen to reach the objective in both the Clayton Act and RICO is the carrot of treble damages. Moreover, both statutes aim to compensate the same type of injury; each requires that a plaintiff show injury `in his business or property by reason of' a violation." 483 U.S., at 151.
We have said that "[a]ny period of limitation . . . is understood fully only in the context of the various circumstances that suspend it from running against a particular cause of action." Johnson v. Railway Express Agency, Inc., 421 U.S. 454, 463 (1975). It is just as true, I think, that any period of limitation is utterly meaningless without specification of the event that starts it running. As a practical matter, a 4-year statute of limitations means nothing at all unless one knows when the four years start running. If they start, for example, on the tenth anniversary of the injury, the 4-year statute is more akin to a 14-year statute than to the Clayton Act. We would thus have been foolish, in Malley-Duff, to speak of "adopting" the Clayton Act statute, and of "patterning" the RICO limitation period after the Clayton Act, if all we meant was using the Clayton Act number of years.
We have recognized this principle in our more established practice (first departed from in DelCostello v. Teamsters, 462 U.S. 151 (1983)) of borrowing state rather than federal statutes of limitations. We have consistently followed "[s]tate law . . . in a variety of cases that raised questions concerning the overtones and details of application of the state limitation period to the federal cause of action. Auto Workers v. Hoosier Corp., 383 U.S. [696,] 706 [(1966)] (characterization of the cause of action); Cope v. Anderson, 331 U.S. [461,] 465-467 [(1947)] (place where cause of action arose); Barney v. Oelrichs, 138 U.S. 529 (1891) (absence from State as a tolling circumstance)." Johnson, supra, at 464. See also, e.g., Chardon v. Fumero Soto, 462 U.S. 650, 657, 662 (1983). "In virtually all statutes of limitations the chronological length of the limitation period is interrelated with provisions regarding tolling, revival, and questions of application. Courts thus should not unravel state limitations rules unless their full application would defeat the goals of the federal statute at issue." Hardin v. Straub, 490 U.S. 536, 539 (1989) (internal quotation marks and citation omitted). There is no conceivable reason why the same principle should not apply to the borrowing of an analogous federal, rather than state, limitations period.
Both the allurement and the vice of the "mix-and-match" approach to statutes-of-limitation borrowing (the possibility of which the Court today entertains) is that it provides broad scope for judicial lawmaking. We should have resisted that allurement today, as we resisted it in the past: "[W]e find no support in our cases for the practice of borrowing only a portion of an express statute of limitations. Indeed, such a practice comes close to the type of judicial policymaking that our borrowing doctrine was intended to avoid." Lampf, Pleva, Lipkind, Prupis Petigrow v. Gilbertson, 501 U.S. 350, 362, n. 8 (1991) (emphasis added). It is, in other words, no wonder that the Court finds the question it has posed for itself today "subtle and difficult"; judicial policyworking is endlessly demanding, and constructing a statute of limitations is much more complicated than adopting one. Finding the most analogous cause of action whose limitations provision can be adopted is relatively simple (for the cause of action before us, we did it in Malley-Duff); but limiting the adoption to merely the term of years set forth in the limitations provision, and then selecting, to go with that term of years, the precise accrual rule, tolling rule, estoppel rule, etc. that will clothe the limitations-naked statute with an ensemble of policy perfection — well that is, I concede, a task that should not be attacked all at once, but rather undertaken piecemeal, over several decades, as the Court has chosen to do today. I prefer to stand by the ruder, humbler, but more efficient and predictable practice we have followed in the past: When we adopt a statute of limitations from an analogous federal cause of action we adopt it in whole, with all its accoutrements. Perhaps (though I am dubious) there is room for an exception similar to the one made in our state-borrowing practice, see Hardin, supra, that would permit rejection of an element that "would defeat the goals of the federal statute at issue," 490 U.S., at 539. But unless this exception is to gobble up the rule, nothing so extreme is represented by the Clayton Act accrual rule.
The Court disclaims any intent to adopt a "mix-and-match" approach, ante, at 193, but that seems to me inconsistent with its repeated references to the possibility of a discovery accrual rule which is (and has been thought to be) the antithesis of the Clayton Act injury accrual rule. If the Court merely means to say that it is not sure how the Clayton Act accrual rule would apply in this case, then it should simply say so — thereby going a long way towards resolving the Circuit split and rendering this concurrence unnecessary.
Applying the Clayton Act accrual rule, I agree with the Court that petitioners' cause of action accrued more than four years before the filing of this action on August 27, 1993. See ante, at 192. Since the Court of Appeals determined, under a more relaxed accrual rule, that petitioners should have discovered all of the RICO elements (which would include their injury) prior to 1989, it follows, a fortiori, that under the Clayton Act injury accrual rule, petitioners' cause of action is untimely.
I also agree with the Court that petitioners are not entitled to invoke the fraudulent concealment doctrine. As the Court persuasively demonstrates, in the antitrust context "`[t]he concealment requirement is satisfied only if the plaintiff shows that he neither knew nor, in the exercise of due diligence, could reasonably have known of the offense.'" Ante, at 195 (quoting 2 P. Areeda H. Hovenkamp, Antitrust Law ¶ 338b, p. 152 (rev. ed. 1995)). I therefore join Part III of the Court's opinion.
For the foregoing reasons, I concur in the judgment of the Court.