BENEFICIAL NAT. BANK v. ANDERSON


Syllabus

BENEFICIAL NAT. BANK v. ANDERSON ( No. 02-306 )
287 F. 3d 1038, reversed.

BENEFICIAL NATIONAL BANK et al. <font i="1">v. ANDERSON et al.

certiorari to the united states court of appeals for the eleventh circuit


No. 02–306. Argued April 30, 2003—Decided June 2, 2003

Respondents, who secured loans from petitioner national bank filed a state-court suit against the bank and two other petitioners, seeking damages on the theory, among others, that the bank’s interest rates violated “the common law usury doctrine” and an Alabama usury statute. The complaint did not refer to any federal law. Petitioners removed the case to Federal District Court, asserting that the National Bank Act governs the interest rate that a national bank may charge, see 12 U. S. C. §85, that the rates charged to respondents complied with §85, that §86 provides the exclusive remedies available against a national bank charging excessive interest, and that respondents’ action was therefore one “arising under” federal law that could be removed under 28 U. S. C. §1441. The District Court denied respondents’ motion to remand the case to state court, but certified the question whether it had jurisdiction to the Eleventh Circuit. In reversing, the latter court held that under the “well-pleaded complaint” rule, removal is not permitted unless the complaint expressly alleges a federal claim, and that the narrow exception known as the complete pre-emption doctrine did not apply because there was no evidence of clear congressional intent to permit removal under §§85 and 86.

Held: Respondents’ cause of action arose only under federal law and could, therefore, be removed under §1441. Pp. 3–9.

(a) As a general rule, absent diversity jurisdiction, a case is not removable if the complaint does not affirmatively allege a federal claim. Potential defenses, including a federal statute’s pre-emptive effect, Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, do not provide a basis for removal. One exception to the general rule occurs when a federal statute completely pre-empts a cause of action. Where this Court has found such pre-emption, the federal statutes at issue—the Labor Management Relations Act, 1947, see Avco Corp. v. Machinists, 390 U. S. 557, and the Employee Retirement Income Security Act of 1974, see Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58—provided the exclusive cause of action for the claim asserted and also set forth procedures and remedies governing that cause of action. Pp. 3–6.

(b) Because respondents’ complaint expressly charged petitioners with usury, Metropolitan Life, Avco, and Franchise Tax Bd. provide the framework for answering the question whether the National Bank Act provides the exclusive cause of action for usury claims against national banks. Section 85 sets substantive limits on the interest rates that national banks may charge, while §86 prescribes the remedies available to borrowers who are charged higher rates and the procedures governing such claims. If the interest charged here did not violate §85 limits, the statute pre-empts any common-law or Alabama statutory rule that would treat those rates as usurious and would, thus, provide a federal defense. That defense would not justify removal. Only if Congress intended §86 to provide the exclusive cause of action for usury claims against national banks would the statute be comparable to the provisions construed in Avco and Metropolitan Life. This Court has long construed the National Bank Act as providing the exclusive federal cause of action for usury against national banks. See, e.g., Farmers’ and Mechanics’ Nat. Bank v. Dearing, 91 U. S. 29. The Court has also recognized the special nature of federally chartered banks. Uniform rules limiting their liability and prescribing exclusive remedies for their overcharges are an integral part of a banking system that needed protection from possible unfriendly state legislation. The same federal interest supports the established interpretation of §§85 and 86 that gives those provisions the requisite pre-emptive force to provide removal jurisdiction. Pp. 6–9.

287 F. 3d 1038, reversed.

Stevens, J., delivered the opinion of the Court, in which Rehnquist, C. J., and O’Connor, Kennedy, Souter, Ginsburg, and Breyer, JJ., joined. Scalia, J., filed a dissenting opinion, in which Thomas, J., joined.


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .


TOP

Dissent

BENEFICIAL NATIONAL BANK, et al ., PETITIONERS v. MARIE ANDERSON et al.

on writ of certiorari to the united states court of appeals for the eleventh circuit


[June 2, 2003]

Justice Scalia , with whom Justice Thomas joins, dissenting.

Today’s opinion takes the view that because §30 of the National Bank Act, 12 U. S. C. §§85, provides the exclusive cause of action for claims of usury against a national bank, all such claims—even if explicitly pleaded under state law—are to be construed as “aris[ing] under” federal law for purposes of our jurisdictional statutes. Ante , at 9. This view finds scant support in our precedents and no support whatever in the National Bank Act or any other Act of Congress. I respectfully dissent.

Unless Congress expressly provides otherwise, the federal courts may exercise removal jurisdiction over state-court actions “of which the district courts of the United States have original jurisdiction.” 28 U. S. C. §1441a. In this case, petitioners invoked as the predicate for removal the district courts’ original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” §1331.

This so-called “arising under” or “federal question” jurisdiction has long been governed by the well-pleaded-complaint rule, which provides that “federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint.” Caterpillar Inc. v. Williams, 482 U. S. 386, 392 (1987) . A federal question “is presented” when the complaint invokes federal law as the basis for relief. It does not suffice that the facts alleged in support of an asserted state-law claim would also support a federal claim. “The [well-pleaded-complaint] rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.” Ibid . See also The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course the party who brings a suit is master to decide what law he will rely upon”). Nor does it even suffice that the facts alleged in support of an asserted state-law claim do not support a state-law claim and would only support a federal claim. “Jurisdiction may not be sustained on a theory that the plaintiff has not advanced.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, n. 6 (1986).

Under the well-pleaded-complaint rule, “a federal court does not have original jurisdiction over a case in which the complaint presents a state-law cause of action, but also asserts that federal law deprives the defendant of a defense he may raise, … or that a federal defense the defendant may raise is not sufficient to defeat the claim.” Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 10 (1983) . Of critical importance here, the rejection of a federal defense as the basis for original federal-question jurisdiction applies with equal force when the defense is one of federal pre-emption. “By unimpeachable authority, a suit brought upon a state statute does not arise under an act of Congress or the Constitution of the United States because prohibited thereby.” Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 116 (1936) . “[A] case may not be removed to federal court on the basis of . . . the defense of pre-emption … .” Caterpillar, supra, at 393. To be sure, pre-emption requires a state court to dismiss a particular claim that is filed under state law, but it does not, as a general matter, provide grounds for removal .

This Court has twice recognized exceptions to the well-pleaded-complaint rule, upholding removal jurisdiction notwithstanding the absence of a federal question on the face of the plaintiff’s complaint. First, in Avco Corp. v. Machinists, 390 U. S. 557 (1968) , we allowed removal of a state-court action to enforce a no-strike clause in a collective-bargaining agreement. The complaint concededly did not advance a federal claim, but was subject to a defense of pre-emption under §301 of the Labor Management Relations Act, 1947 (LMRA), 29 U. S. C. §185. The well-pleaded-complaint rule notwithstanding, we treated the plaintiff’s state-law contract claim as one arising under §301, and held that the case could be removed to federal court. Avco, supra, at 560.

The only support mustered by the Avco Court for its conclusion was a statement wrenched out of context from our decision in Textile Workers v. Lincoln Mills of Ala., 353 U. S. 448, 457 (1957) , that “[a]ny state law applied [in a §301 case] will be absorbed as federal law and will not be an independent source of private rights.” To begin with, this statement is entirely unnecessary to the landmark holding in Lincoln Mills —that §301 not only gives federal courts jurisdiction to decide labor relations cases but also supplies them with authority to create the governing substantive law. Id., at 456. More importantly, understood in the context of that holding, the quoted passage in no way supports the proposition for which it is relied upon in Avco —that state-law claims relating to labor relations necessarily arise under §301. If one reads Lincoln Mills with any care, it is clear beyond doubt that the relevant passage merely confirms that when, in deciding cases arising under §301, courts employ legal rules that overlap with, or are even explicitly borrowed from, state law, such rules are nevertheless rules of federal law. It is in this sense that “[a]ny state law applied [in a §301 case] will be absorbed as federal law”—in the sense that federally adopted state rules become federal rules, not in the sense that a state-law claim becomes a federal claim.

Other than its entirely misguided reliance on Lincoln Mills , the opinion in Avco failed to clarify the analytic basis for its unprecedented act of jurisdictional alchemy. The Court neglected to explain why state-law claims that are pre-empted by §301 of the LMRA are exempt from the strictures of the well-pleaded-complaint rule, nor did it explain how such a state-law claim can plausibly be said to “arise under” federal law. Our subsequent opinion in Franchise Tax Board , struggled to prop up Avco ’s puzzling holding:

“The necessary ground of decision [in Avco ] was that the pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301. Avco stands for the proposition that if a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law.” 463 U. S., at 23–24 (footnote omitted).

This passage has repeatedly been relied upon by the Court as an explanation for its decision in Avco . See, e.g. , ante , at 4–5, Caterpillar, supra, at 394; Metropolitan Life Ins. Co. v. Taylor, 481 U. S. 58, 64 (1987) . Of course it is not an explanation at all. It provides nothing more than an account of what Avco accomplishes, rather than a justification (unless ipse dixit is to count as justification) for the radical departure from the well-pleaded-complaint rule, which demands rejection of the defense of federal pre-emption as a basis for federal jurisdiction. Gully, supra, at 116. Neither the excerpt quoted above, nor any other fragment of the decision in Franchise Tax , explains how or why the nonviability (due to pre-emption) of the state-law contract claim in Avco magically transformed that claim into one “arising under” federal law.

Metropolitan Life Ins. Co. v. Taylor, supra, was our second departure from the prohibition against resting federal “arising under” jurisdiction upon the existence of a federal defense. In that case, Taylor sued his former employer and its insurer, alleging breach of contract and seeking, inter alia , reinstatement of certain disability benefits and insurance coverages. Id ., at 61. Though Taylor invoked no federal law in his complaint, we treated his case as one arising under §502 of the Employee Retirement Income Security Act of 1974 (ERISA) and upheld the District Court’s exercise of removal jurisdiction. Id. , at 66–67.

In reaching this conclusion, the Taylor Court broke no new analytic ground; its opinion follows the exception established in Avco and described in Franchise Tax Board , but says nothing to commend that exception to logic or reason. Instead, Taylor simply relies on the “clos[e] parallels,” 481 U. S., at 65, between the language of the pre-emptive provision in ERISA and the language of the LMRA provision deemed in Avco to be so dramatically pre-emptive as to summon forth a federal claim where none had been asserted. “No more specific reference to the Avco rule can be expected,” we said, than what was found in §502(a); and we accordingly concluded that “Congress has clearly manifested an intent to make causes of action within the scope of the civil enforcement provisions of §502(a) removable to federal court.” 481 U. S., at 66. As in Avco and Franchise Tax Board , no explanation was provided for Avco ’s abrogation of the rule that “[f]ederal pre-emption is ordinarily a federal defense to the plaintiff ’s suit[, and as such] it does not appear on the face of a well-pleaded complaint, [nor does it] authorize removal to federal court.” 1 481 U. S., at 63.

It is noteworthy that the straightforward (though similarly unsupported) rule announced in today’s opinion— under which (1) removal is permitted “[w]hen [a] federal statute completely pre-empts a state-law cause of action,” ante , at 6, and (2) a federal statute is completely pre-emptive when it “provide[s] the exclusive cause of action for the claim asserted,” ibid .—is nowhere to be found in either Avco or Taylor . To the contrary, the analysis in today’s opinion implicitly contradicts (by rendering inexplicable) Taylor ’s discussion of pre-emption and removal. ( Avco , as I observed earlier, has no discussion to be contradicted.) Had it thought that today’s decision was the law, the Taylor Court need not have taken pains to emphasize the “clos[e] parallels” between §502(a)(1)(B) of ERISA and §301 of the LMRA and need not have pored over the legislative history of §502(a) to show that Congress expected ERISA to be treated like the LMRA. See Taylor, supra, at 65–66 (citing H. R. Conf. Rep. No. 93–1280, p. 327, (1974); 120 Cong. Rec. 29933 (1974) (remarks of Sen. Williams); id ., at 29942 (remarks of Sen. Javits)). Instead, it could have rested after noting the “unique pre-emptive force of ERISA,” Taylor, supra, at 65. Indeed, it could even have spared itself the trouble of adding the adjective “unique.” While there is something unique about statutes whose pre-emptive force is closely patterned after that of the LMRA (which we had held to support removal), there is nothing whatever unique about a federal cause of action that displaces state causes of action. Displacement alone, if today’s opinion is to be believed, would have sufficed to establish the existence of removal jurisdiction.

The best that can be said, from a precedential perspective, for the rule of law announced by the Court today is that variations on it have twice appeared in our cases in the purest dicta. Rivet v. Regions Bank of La., 522 U. S. 470, 476 (1998) (“[O]nce an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state-law claim is considered, from its inception, a federal claim, and therefore arises under federal law” (internal quotation marks omitted)); Caterpillar, 482 U. S., at 393 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law” (some internal quotation marks omitted)). Dicta of course have no precedential value, see U. S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U. S. 18, 24 (1994) , even when they do not contradict, as they do here, prior holdings of the Court.

The difficulty with today’s holding, moreover, is not limited to the flimsiness of its precedential roots. As has been noted already, the holding cannot be squared with bedrock principles of removal jurisdiction. One or another of two of those principles must be ignored: Either (1) the principle that merely setting forth in state court facts that would support a federal cause of action—indeed, even facts that would support a federal cause of action and would not support the claimed state cause of action—does not produce a federal question supporting removal, Caterpillar , 482 U. S., at 391, or (2) the principle that a federal defense to a state cause of action does not support federal-question jurisdiction, see id., at 393. Relatedly, today’s holding also represents a sharp break from our long tradition of respect for the autonomy and authority of state courts. For example, in Healy v. Ratta, 292 U. S. 263, 270 (1934) , we explained that “[d]ue regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” And in Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 108 (1941) , we insisted on a “strict construction” of the federal removal statutes. 2 Today’s decision ignores these venerable principles and effectuates a significant shift in decisional authority from state to federal courts.

In an effort to justify this shift, the Court explains that “[b]ecause §§85 and 86 [of the National Bank Act] provide the exclusive cause of action for such claims, there is … no such thing as a state-law claim of usury against a national bank.” Ante , at 9. But the mere fact that a state-law claim is invalid no more deprives it of its character as a state-law claim which does not raise a federal question, than does the fact that a federal claim is invalid deprive it of its character as a federal claim which does raise a federal question. The proper response to the presentation of a nonexistent claim to a state court is dismissal , not the “federalize-and-remove” dance authorized by today’s opinion. For even if the Court is correct that the National Bank Act obliterates entirely any state-created right to relief for usury against a national bank, that does not explain how or why the claim of such a right is transmogrified into the claim of a federal right. Congress’s mere act of creating a federal right and eliminating all state-created rights in no way suggests an expansion of federal jurisdiction so as to wrest from state courts the authority to decide questions of pre-emption under the National Bank Act.

Petitioners seek to justify their end-run around the well-pleaded-complaint rule by insisting that, in determining whether federal jurisdiction exists, we are required to “ ‘look beyond the pleadings.’ ” Brief for Petitioners 18 (quoting Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69 (1941) ). They point out:

“[A] long line of cases disallow[s] manipulations by plaintiffs designed to create or avoid diversity jurisdiction, such as misaligning the interests of the parties, naming parties (whether plaintiffs or defendants) who have no real interest in or relationship to the controversy, misstating the citizenship of a party (whether plaintiffs or defendants), or misstating the amount in controversy.” Brief for Petitioners 17–18.

Petitioners insist that, like the “manipulative” complaints in these diversity cases, “[r]espondents’ complaint is disingenuously pleaded, not ‘well pleaded’ in any respect, for it purports to raise a state law claim that does not exist.” Id. , at 16. Accordingly, the argument continues, just as federal courts may assert jurisdiction where a plaintiff seeks to hide the true citizenship of the parties, so too they may assert jurisdiction where a plaintiff cloaks a necessarily federal claim in state-law garb.

To begin with, the cases involving diversity jurisdiction are probably distinguishable on the ground that there is a crucial difference between, on the one hand, “looking beyond the pleadings” to determine whether a factual assertion is true, and, on the other hand, doing so in order to determine whether the plaintiff has proceeded on the basis of the “correct” legal theory. But even assuming that the analogy to the diversity cases is apt, petitioners can derive no support from it in this case. Their argument proceeds from the faulty premise that if one looks behind the pleadings in this case, one discovers that the plaintiffs have, in fact, presented a federal claim. But that begs the question—that is, it assumes the answer to the very question presented. It assumes that whenever a claim of usury is brought against a national bank, that claim is a federal one. As I have discussed above, neither logic nor precedent supports that conclusion; they support, at best, the proposition that the only viable claim against a national bank for usury is a federal one. Federal jurisdiction is ordinarily determined—invariably determined, except for Avco and Taylor —on the basis of what claim is pleaded, rather than on the basis of what claim can prevail.

There may well be good reasons to favor the expansion of removal jurisdiction that petitioners urge and that the Court adopts today. As the United States explains in its amicus brief:

“Absent removal, the state court would have only two legitimate options—to recharacterize the claim in federal-law terms or to dismiss the claim altogether. Any plaintiff who truly seeks recovery on that claim would prefer the first option, which would make the propriety of removal crystal clear. A third possibility, however, is that the state court would err and allow the claim to proceed under state law notwithstanding Congress’s decision to make the federal cause of action exclusive. The complete pre-emption rule avoids that potential error.” Brief for United States as Amicus Curiae 17–18.

True enough, but inadequate to render today’s decision either rational or properly within the authority of this Court. Inadequate for rationality, because there is no more reason to fear state-court error with respect to federal pre-emption accompanied by creation of a federal cause of action than there is with respect to federal pre-emption unaccompanied by creation of a federal cause of action—or, for that matter, than there is with respect to any federal defense to a state-law claim. The rational response to the United States’ concern is to eliminate the well-pleaded-complaint rule entirely. And inadequate for judicial authority, because it is up to Congress, not the federal courts, to decide when the risk of state-court error with respect to a matter of federal law becomes so unbearable as to justify divesting the state courts of authority to decide the federal matter. Unless and until we receive instruction from Congress that claims pre-empted under the National Bank Act—in contrast to almost all other claims that are subject to federal pre-emption—“arise under” federal law, we simply lack authority to “avoi[d] … potential errors,” id., at 18, by permitting removal.

* * *

Today’s opinion has succeeded in giving to our Avco decision a theoretical foundation that neither Avco itself nor Taylor provided. Regrettably, that theoretical foundation is itself without theoretical foundation. That is to say, the more general proposition that (1) the existence of a pre-emptive federal cause of action causes the invalid assertion of a state cause of action to raise a federal question, has no more logic or precedent to support it than the very narrow proposition that (2) the LMRA (Avco) and statutes modeled after the LMRA (Taylor) cause invalid assertions of state causes of action pre-empted by those particular statutes to raise federal questions. Since I believe that, as between an inexplicable narrow holding and an inexplicable broad one, the former is the lesser evil, I would adhere to the approach taken by Taylor and on the basis of stare decisis simply affirm, without any real explanation, that the LMRA and statutes modeled after it have a “unique pre-emptive force” that (quite illogically) suspends the normal rules of removal jurisdiction. Since no one asserts that the National Bank Act is modeled after the LMRA, the state-law claim pleaded here cannot be removed, and it is left to the state courts to dismiss it. From the Court’s judgment to the contrary, I respectfully dissent.


Notes

1 This is not to say that Taylor was wrongly decided. Having been informed through the Avco Corp. v. Machinists, 390 U. S. 557 (1968) , decision that the language of §301 triggered “arising under” jurisdiction even with respect to certain state-law claims, Congress’ subsequent decision to insert language into ERISA that “closely parallels” the text of §301 can be viewed to be, as we said, a “specific reference to the Avco rule.” 481 U. S., at 65–66. Taylor, in other words, rests upon a sort of statutory incorporation of Avco. Avco itself, on the other hand, continues to rest upon nothing.

2 Our traditional regard for the role played by state courts in interpreting and enforcing federal law has other doctrinal manifestations. We indulge, for example, a “presumption of concurrent [state and federal] jurisdiction,” which can be rebutted only “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981) .