NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press.
SUPREME COURT OF THE UNITED STATES
_________________
No. 20–1143
_________________
Denise A. Badgerow, PETITIONER
v. Greg Walters, et al.
on writ of certiorari to the united states court of appeals for the fifth circuit
[March 31, 2022]
Justice Kagan delivered the opinion of the Court.
The Federal Arbitration Act (FAA),
9 U. S. C. §1
et seq., authorizes a party to an arbitration agreement to seek several kinds of assistance from a federal court. Under Section 4, for example, a party may ask the court to compel an arbitration proceeding, as the agreement contemplates. And under Sections 9 and 10, a party may apply to the court to confirm, or alternatively to vacate, an arbitral award.
Yet the federal courts, as we have often held, may or may not have jurisdiction to decide such a request. The Act’s authorization of a petition does not itself create jurisdiction. Rather, the federal court must have what we have called an
“independent jurisdictional basis” to resolve the matter.
Hall Street Associates, L. L. C. v.
Mattel, Inc.,
552 U. S. 576, 582 (2008).
In
Vaden v.
Discover Bank,
556 U. S. 49 (2009), we assessed whether there was a jurisdictional basis to decide a Section 4 petition to compel arbitration by means of examining the parties’ underlying dispute. The text of Section 4, we reasoned, instructs a federal court to “look through” the petition to the “underlying substantive controversy” between the parties—even though that controversy is not before the court.
Id., at 62. If the underlying dispute falls within the court’s jurisdiction—for example, by presenting a federal question—then the court may rule on the petition to compel. That is so regardless whether the petition alone could establish the court’s jurisdiction.
The question presented here is whether that same “look-through” approach to jurisdiction applies to requests to confirm or vacate arbitral awards under the FAA’s Sections 9 and 10. We hold it does not. Those sections lack Section 4’s distinctive language directing a look-through, on which
Vaden rested. Without that statutory instruction, a court may look only to the application actually submitted to it in assessing its jurisdiction.
I
This case grows out of the arbitration of an employment dispute. Petitioner Denise Badgerow worked as a financial advisor for REJ Properties, a firm run by respondents Greg Walters, Thomas Meyer, and Ray Trosclair. (For ease of reference, we refer from now on only to Walters.) Badgerow’s contract required her to bring claims arising out of her employment to arbitration, rather than to court. So when she was (in her view, improperly) fired, she initiated an arbitration action against Walters, alleging unlawful termination under both federal and state law. The arbitrators sided with Walters, dismissing Badgerow’s claims.
What happened afterward—when Badgerow refused to give up—created the jurisdictional issue we address today. Believing that fraud had tainted the arbitration proceeding, Badgerow sued Walters in Louisiana state court to vacate the arbitral decision. Walters responded by removing the case to Federal District Court—and, once there, applying to confirm the arbitral award. Finally, Badgerow moved to remand the case to state court, arguing that the federal court lacked jurisdiction over the parties’ requests—under Sections 10 and 9, respectively—to vacate or confirm the award.
The District Court assessed its jurisdiction under the look through approach this Court adopted in
Vaden v.
Discover Bank. See 2019 WL 2611127, *1 (ED La., June 26, 2019). That approach, as just noted, allows a federal court to exercise jurisdiction over an FAA application when the parties’ underlying substantive dispute would have fallen within the court’s jurisdiction. See
supra, at 1–2. The District Court acknowledged that
Vaden involved a different kind of arbitration dispute: It concerned a petition to compel arbitration under the FAA’s Section 4, rather than an application to confirm or vacate an arbitral award under Section 9 or 10. And
Vaden’s “reasoning was grounded on specific text” in Section 4 that Sections 9 and 10
“do[ ] not contain.” 2019 WL 2611127, *2. But the court thought it should apply the look-through approach anyway, so that “consistent jurisdictional principles” would govern all kinds of FAA applications.
Ibid. And under that approach, the court had jurisdiction because Badgerow’s underlying employment action raised federal-law claims. The court thus went on to
resolve the dispute over whether fraud had infected the arbitration proceeding. Finding it had not, the court granted Walters’s application to confirm, and denied Badgerow’s application to vacate, the arbitral award.
The United States Court of Appeals for the Fifth Circuit affirmed the District Court’s finding of jurisdiction, relying on a just-issued Circuit precedent. See 975 F. 3d 469, 472–474 (2020) (citing
Quezada v.
Bechtel OG&C Constr. Servs.,
Inc., 946 F. 3d 837, 843 (2020)). In that decision, the Fifth Circuit had echoed the reasoning of the District Court here. Yes, the language of Section 4 directing use of the look-through approach “is in fact absent in” the FAA’s other sections. 946 F. 3d, at 842. But, the court continued, a “principle of uniformity” applying to the FAA “dictates using the same approach for determining jurisdiction under each section of the statute.”
Ibid.; but see
id., at 845–846 (Ho, J., dissenting) (rejecting that asserted principle in favor of “[f ]idelity to text”).
As applied to this case, that analysis meant that the district court had jurisdiction over Walters’s Section 9 and Badgerow’s Section 10 applications.
Courts have divided over whether the look-through approach used in
Vaden can establish jurisdiction in a case like this one—when the application before the court seeks not to compel arbitration under Section 4 but to confirm, vacate, or modify an arbitral award under other sections of the FAA.[
1] We granted certiorari to resolve the conflict, 593 U. S. ___ (2021), and now reverse the judgment below.
II
The district courts of the United States are courts of limited jurisdiction, defined (within constitutional bounds) by federal statute. See,
e.g.,
Kokkonen v.
Guardian Life Ins. Co. of America,
511 U. S. 375, 377 (1994). Congress has granted those courts jurisdiction over two main kinds of cases. District courts have power to decide diversity cases—suits between citizens of different States as to any matter valued at more than $75,000. See
28 U. S. C. §1332(a). And they have power to decide federal-question cases—suits “arising under” federal law. §1331. Typically, an action arises under federal law if that law “creates the cause of action asserted.”
Gunn v.
Minton,
568 U. S. 251, 257 (2013). So when federal law authorizes the action, the party bringing it—once again, typically—gets to go to federal court.
But that is not necessarily true of FAA-created arbitration actions. As noted above, the FAA authorizes parties to arbitration agreements to file specified actions in federal court—most prominently, petitions to compel arbitration (under Section 4) and applications to confirm, vacate, or modify arbitral awards (under Sections 9 through 11). See
supra, at 1. But those provisions, this Court has held, do not themselves support federal jurisdiction. See
Hall Street, 552 U. S., at 581–582;
Vaden, 556 U. S., at 59. (Were it otherwise, every arbitration in the country, however distant from federal concerns, could wind up in federal district court.) A federal court may entertain an action brought under the FAA only if the action has an “independent jurisdictional basis.”
Hall Street, 552 U. S., at 582. That means an applicant seeking, for example, to vacate an arbitral award under Section 10 must identify a grant of jurisdiction, apart from Section 10 itself, conferring “access to a federal forum.”
Vaden, 556 U. S., at 59. If she cannot, the action belongs in state court. The FAA requires those courts, too, to honor arbitration agreements; and we have long recognized their “prominent role” in arbitral enforcement.
Ibid.; see
id., at 71;
Southland Corp. v.
Keating,
465 U. S. 1, 12–16 (1984).[
2]
The issue here is about where a federal court should look to determine whether an action brought under Section 9 or 10 has an independent jurisdictional basis. An obvious place is the face of the application itself. If it shows that the contending parties are citizens of different States (with over $75,000 in dispute), then §1332(a) gives the court diversity jurisdiction. Or if it alleges that federal law (beyond Section 9 or 10 itself ) entitles the applicant to relief, then §1331 gives the court federal-question jurisdiction. But those possibilities do Walters no good. He and Badgerow are from the same State. And their applications raise no federal issue. Recall that the two are now contesting not the legality of Badgerow’s firing but the enforceability of an arbitral award. That award is no more than a contractual resolution of the parties’ dispute—a way of settling legal claims. See
Vaden, 556 U. S., at 63.
And quarrels about legal settlements—even settlements of federal claims—typically involve only state law, like disagreements about other contracts. See
Kokkonen, 511 U. S., at 378–382. So the District Court here, as Walters recognizes, had to go beyond the face of the Section 9 and 10 applications to find a basis for jurisdiction. See Brief for Respondents 26–27. It had to proceed downward to Badgerow’s employment action, where a federal-law claim satisfying §1331 indeed exists. In other words, the court had to look through the Section 9 and 10 applications to the underlying substantive dispute, although that dispute was not before it. Could the court do so?
In
Vaden, this Court approved the look-through approach for a Section 4 petition, relying on that section’s express language. Under Section 4, a party to an arbitration agreement may petition for an order to compel arbitration in a “United States district court which, save for [the arbitration] agreement, would have jurisdiction” over “the controversy between the parties.”[
3] That text, we stated, “drives our conclusion that a federal court should determine its jurisdiction by ‘looking through’ a §4 petition to the underlying substantive controversy”—to see, for example, if that dispute “ ‘arises under’ federal law.” 556 U. S., at 62.
To show why that is so, we proceeded methodically through Section 4’s wording. “The phrase ‘save for [the arbitration] agreement,’ ” we began, “indicates that the district court should assume the absence of the arbitration agreement and determine whether [the court] ‘would have jurisdiction . . .’ without it.”
Ibid. (first alteration in original). But “[j]urisdiction over what?”
Ibid. “The text of Section 4,” we continued, “refers us to ‘the controversy between the parties.’ ”
Ibid. And that “controversy,” we explained, could not mean the dispute before the court about “the existence or applicability of an arbitration agreement”; after all, the preceding save-for clause had just “direct[ed] courts” to assume that agreement away.
Id., at 63. The “controversy between the parties” instead had to mean their “underlying substantive controversy.”
Id., at 62 (internal quotation marks omitted). “Attending to the language” of Section 4 thus required “approv[ing] the ‘look through’ approach” as a means of assessing jurisdiction over petitions to compel arbitration.
Ibid. The opposite view was not merely faulty; it was “textual[ly] implausib[le].”
Id., at 65.
But Sections 9 and 10, in addressing applications to confirm or vacate an arbitral award, contain none of the statutory language on which
Vaden relied. Most notably, those provisions do not have Section 4’s “save for” clause. They do not instruct a court to imagine a world without an arbitration agreement, and to ask whether it would then have jurisdiction over the parties’ dispute. Indeed, Sections 9 and 10 do not mention the court’s subject-matter jurisdiction at all.[
4] So under ordinary principles of statutory construction, the look-through method for assessing jurisdiction should not apply. “[W]hen Congress includes particular language in one section of a statute but omits it in another section of the same Act,” we generally take the choice to be deliberate.
Collins v.
Yellen, 594 U. S. ___, ___ (2021) (slip op., at 23) (internal quotation marks omitted). We have no warrant to redline the FAA, importing Section 4’s consequential language into provisions containing nothing like it. Congress could have replicated Section 4’s look-through instruction in Sections 9 and 10. Or for that matter, it could have drafted a global look-through provision, applying the approach throughout the FAA. But Congress did neither. And its decision governs.
Nothing in that conclusion changes because a jurisdictional question is before us. The federal “district courts may not exercise jurisdiction absent a statutory basis.”
Exxon Mobil Corp. v.
Allapattah Services, Inc.,
545 U. S. 546, 552 (2005). And the jurisdiction Congress confers may not “be expanded by judicial decree.”
Kokkonen, 511 U. S., at 377. Those bedrock principles prevent us from pulling look-through jurisdiction out of thin air—from somehow finding, without textual support, that federal courts may use the method to resolve various state-law-based, non-diverse Section 9 and 10 applications. The look-through rule is a highly unusual one: It locates jurisdiction not in the action actually before the court, but in another controversy neither there nor ever meant to be. We recognized that rule in
Vaden because careful analysis of Section 4’s text showed that Congress wanted it applied to petitions brought under that provision. See 556 U. S., at 62–65. But Congress has not so directed in Sections 9 and 10. Congress has not authorized a federal court to adjudicate a Section 9 or 10 application just because the contractual dispute it presents grew out of arbitrating different claims, turning on different law, that (save for the parties’ agreement) could have been brought in federal court. And because a statutory basis for look-through jurisdiction is lacking here, we cannot reach the same result as in
Vaden: That would indeed be jurisdictional “expan[sion] by judicial decree.”
Kokkonen, 511 U. S., at 377.
Walters contests that view of the statute. Unlike the Courts of Appeals to have adopted his position, including the Fifth Circuit below, see
supra, at 3–4, he offers a reading of the FAA’s text to justify exercising jurisdiction here.[
5] Walters’s argument comes in two parts. First, Walters says, the language in Section 4 that
Vaden construed does not in fact authorize the use of the look-through jurisdictional method. In his view, that sentence is only a capacious “
venue provision,” serving to “expand[ ] venue to the limits of [federal] jurisdiction” (and thus to give an applicant a broad choice
among federal courts possessing jurisdiction). Brief for Respondents 12, 23. Second, Walters claims that Section 6 provides the basis for an FAA-wide look-through jurisdictional rule. Under Section 6, any FAA application “shall be made and heard in the manner provided by law for the making and hearing of motions.” That provision, Walters claims, requires use of the look-through approach because “[f]ederal courts have jurisdiction over motions when they have jurisdiction over the underlying action.” Brief for Respondents 19 (internal quotation marks omitted). So to recap Walters’s theory: Section 4 does not establish any jurisdictional rule for applications to compel in particular, while Section 6 establishes the look-through jurisdictional rule for all kinds of FAA applications.
But Walters’s understanding of Section 4 does not comport with what it says. The language of that provision never mentions “venue”; it refers only to “jurisdiction.” That is a signal, sharp and clear, that the section provides a jurisdictional rule. And even suppose (against all odds) that Congress had meant to state the venue rule Walters proposes without ever using the word “venue.” In that event, Congress could have simply permitted filing the petition in any district court with jurisdiction (or even more simply—because a court can never act without jurisdiction—in any district court). Given that (in Walters’s view) the jurisdictional rule comes from another provision, Congress would not have needed to (again) spell out its content. But spelling out the rule’s content—by describing the look-through method—is exactly what Section 4 does. That description can serve one purpose only: to establish jurisdiction where it would otherwise not exist.
And that is how
Vaden understood Section 4. Our decision, like the relevant text, never once referred to venue. Instead, we spoke, throughout the opinion, of the way Section 4 provides for jurisdiction. We formulated the question presented as whether the district court could “exercise jurisdiction over [the party’s] §4 petition.” 556 U. S., at 53; see
id., at 57 (stating that “[w]e granted certiorari” to decide whether district courts could use the look-through method “to determine whether federal-question jurisdiction exists over [a] §4 petition”). And we framed our holding as about jurisdiction: “[A] federal court should determine its jurisdiction by ‘looking through’ a §4 petition.”
Id., at 62; see
id., at 72 (Roberts, C. J., dissenting) (differing about the rule’s application, but agreeing that a court presented with a Section 4 petition should use the look-through method “in determining whether it has jurisdiction”). In short, Section 4’s “save for” text “dr[ove] our conclusion” not about venue, but about “jurisdiction.”
Id., at 62. And so that text, as shown above, contradicts Walters’s position—for it appears in Section 4 alone, rather than also in Sections 9 and 10. See
supra, at 8–9.
Walters’s theory fares no better in construing Section 6’s mention of motions to prescribe a look-through rule for the whole FAA. Here, Walters commits the opposite of his fault in reading Section 4: He now reads a provision containing no express reference to jurisdiction in fact to set out a jurisdictional rule. There may be rare contexts in which courts can, without such a reference, “infer that Congress has expanded our jurisdiction”—but this is not one.
Welch v.
Texas Dept. of Highways and Public Transp.,
483 U. S. 468, 474 (1987) (plurality opinion). The look-through method, as noted before, is a jurisdictional outlier. See
supra, at 9. For Congress to prescribe it by telling courts, a la Section 6, to treat FAA applications like motions in other kinds of litigation would be not just oblique but simply bizarre. Courts, after all, do not possess jurisdiction to decide ordinary motions by virtue of the look-through method. A motion (unlike a typical FAA application) is part of a case actually in court. Jurisdiction to decide the case includes jurisdiction to decide the motion; there is no need to “look through” the motion in search of a jurisdictional basis outside the court. And if the look-through rule does not apply to motions, then Section 6’s reference to motions cannot direct the look-through rule. We have formerly described that provision’s function as something different: Section 6, we said, ensures that FAA applications “get streamlined treatment”—a kind of “expedited review,” as compared to what a party would receive if she brought a normal contract suit.
Hall Street, 552 U. S., at 582, and n. 3. However hard we squint, we cannot also discern in Section 6 an FAA-wide look-through rule; the only such rule in the FAA, applying only to petitions to compel, resides in Section 4.
Walters’s more thought-provoking arguments sound not in text but in policy. Here, Walters—now joined by the dissent—preaches the virtues of adopting look-through as a “single, easy-to-apply jurisdictional test” that will produce “sensible” results. Brief for Respondents 28 (internal quotation marks omitted); see
post, at 4–9 (opinion of Breyer, J.) (lauding the “advantages” of look-through’s “practical consequences”). First, Walters says, a uniform jurisdictional rule, applying to all FAA applications alike, will necessarily promote “administrative simplicity” because a court will not have to figure out which rule to apply. Brief for Respondents 27. Second, he claims, the look-through rule is “easier to apply” than a test that would ground jurisdiction on the face of the FAA application itself.
Id., at 28 (internal quotation marks omitted). In particular, he says, the latter approach confronts courts with “hard questions” about how to determine diversity jurisdiction (including its amount-in-controversy component) across a range of settings—for the Section 9 and 10 applications at issue here, as well as for Section 5 and 7 petitions (obviously not at issue) to appoint arbitrators or compel the presence of witnesses.
Id., at 41. (The dissent’s vaunted practical “advantages” also mostly concern avoiding those diversity issues.
Post, at 4; see
post, at 4–7.)[
6] Finally, Walters contends that only the look-through rule will provide federal courts with comprehensive control over the arbitration process, including the period after the award. The opposite position, he says, will “close the federal courthouse doors to many” post-arbitration motions, even when they grow out of disputes raising “
exclusively federal claims.” Brief for Respondents 37, 46.
Walters himself quotes back to us the topline answer to those theories, reflecting its obviousness: “Even the most formidable policy arguments cannot overcome a clear statutory directive.”
Id., at 44 (quoting
BP p.l.c. v.
Mayor and City Council of Baltimore, 593 U. S. ___, ___ (2021) (slip op., at 12); alteration omitted). Walters’s (and the dissent’s) what-makes-best-sense assertions rest on the view that “the FAA contains no” such clear “directive” limiting look-through jurisdiction to Section 4. Brief for Respondents 44–45; see
post, at 10. Having rejected that view, we cannot find much relevance in his ideas, even if plausible, about the optimal jurisdictional rule for the FAA. “It is not for this Court to employ untethered notions of what might be good public policy to expand our jurisdiction.”
Whitmore v.
Arkansas,
495 U. S. 149, 161 (1990). However the pros and cons shake out, Congress has made its call. We will not impose uniformity on the statute’s non-uniform jurisdictional rules.
And anyway, we think Walters oversells the superiority of his proposal. First, uniformity in and of itself provides no real advantage in this sphere. A court can tell in an instant whether an application arises under Section 4 or, as here, under Section 9 or 10; so it can also tell in an instant whether to apply the look-through method or the usual jurisdictional rules. Second, the use of those ordinary rules—most notably, relating to diversity jurisdiction—is hardly beyond judicial capacity. Federal courts have faced, and federal courts have resolved, diversity questions for over two centuries, in diverse and ever-changing legal contexts. Throughout, they have developed workable rules; and we see no reason to think they will do differently here. Indeed, past practice belies Walters’s and the dissent’s gloomy predictions. Although they spin out hypotheticals designed to make the project look ultra-confusing, they fail to identify any actual problems that have arisen from courts’ longstanding application of diversity standards to FAA applications (without using look-through). And Walters’s solution does not even avoid the (purported) difficulty of which he complains. For he does not claim (nor could he) that look-through is the exclusive means of establishing federal jurisdiction. Even if the underlying action does not fall within a district court’s jurisdiction, the application still might do so—say, because the parties have changed, and are now diverse. See
supra, at 6. So courts, on Walters’s own view, will still have to resolve questions about—and develop rules for—determining diversity in the FAA context. The difference is only one of degree—and too small, under any plausible theory of statutory interpretation, to adopt Walters’s proposal to rewrite the law.
Finally, we can see why Congress chose to place fewer arbitration disputes in federal court than Walters wishes. The statutory plan, as suggested above, makes Section 9 and 10 applications conform to the normal—and sensible—judicial division of labor: The applications go to state, rather than federal, courts when they raise claims between non-diverse parties involving state law. See
supra, at 5–6. As Walters notes, those claims may have originated in the arbitration of a federal-law dispute. But the underlying dispute is not now at issue. Rather, the application concerns the contractual rights provided in the arbitration agreement, generally governed by state law. And adjudication of such state-law contractual rights—as this Court has held in addressing a non-arbitration settlement of federal claims—typically belongs in state courts. See
Kokkonen, 511 U. S., at 381–382;
supra, at 6. To be sure, Congress created an exception to those ordinary jurisdictional principles for Section 4 petitions to compel. But it is one thing to make an exception, quite another to extend that exception everywhere. See
post, at 8 (disregarding this point). As this Court has often said, the “preeminent” purpose of the FAA was to overcome some judges’ reluctance to enforce arbitration agreements when a party tried to sue in court instead.
E.g., Dean Witter Reynolds Inc. v.
Byrd,
470 U. S. 213, 221 (1985). We have never detected a similar congressional worry about judges’ willingness to enforce arbitration awards already made. So Congress might well have thought an expansion of federal jurisdiction appropriate for petitions to compel alone. Applications about arbitral decisions could and should follow the normal rules.
The result, as Walters laments, is to give state courts a significant role in implementing the FAA. But we have long recognized that feature of the statute. “[E]nforcement of the Act,” we have understood, “is left in large part to the state courts.”
Moses H. Cone Memorial Hospital v.
Mercury Constr. Corp.,
460 U. S. 1, 25, n. 32 (1983); see
Vaden, 556 U. S., at 59;
Hall Street, 552 U. S., at 582. As relevant here, Congress chose to respect the capacity of state courts to properly enforce arbitral awards. In our turn, we must respect that evident congressional choice.
* * *
For the reasons stated, we reverse the judgment of the Court of Appeals for the Fifth Circuit and remand the case for further proceedings consistent with this opinion.
It is so ordered.