(delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Alito, JJ., joined)
Section 27 of the Securities Exchange Act of 1934 (Exchange Act), 48Stat. 992, as amended, 15 U. S. C. §78a, et seq., grants federal district courts exclusive jurisdiction “of all suits in equity and actions at law brought to enforce any liability or duty created by [the Exchange Act] or the rules or regulations thereunder.” §78aa(a). We hold today that the jurisdictional test established by that provision is the same as the one used to decide if a case “arises under” a federal law. See 28 U. S. C. §1331.
Respondent Greg Manning held more than two million shares of stock in Escala Group, Inc., a company traded on the NASDAQ. Between 2006 and 2007, Escala’s share price plummeted and Manning lost most of his investment. Manning blames petitioners, Merrill Lynch and several other financial institutions (collectively, Merrill Lynch), for devaluing Escala during that period through “naked short sales” of its stock.
A typical short sale of a security is one made by a borrower, rather than an owner, of stock. In such a transaction, a person borrows stock from a broker, sells it to a buyer on the open market, and later purchases the same number of shares to return to the broker. The short seller’s hope is that the stock price will decline between the time he sells the borrowed shares and the time he buys replacements to pay back his loan. If that happens, the seller gets to pocket the difference (minus associated transaction costs).
In a “naked” short sale, by contrast, the seller has not borrowed (or otherwise obtained) the stock he puts on the market, and so never delivers the promised shares to the buyer. See “Naked” Short Selling Antifraud Rule, Securities Exchange Commission (SEC) Release No. 34–58774, 73 Fed. Reg. 61667 (2008). That practice (beyond its effect on individual purchasers) can serve “as a tool to drive down a company’s stock price” – which, of course, injures shareholders like Manning. Id., at 61670. The SEC regulates such short sales at the federal level: The Commission’s Regulation SHO, issued under the Exchange Act, prohibits short sellers from intentionally failing to deliver securities and thereby curbs market manipulation. See 17 CFR §§242.203–242.204 (2015).
In this lawsuit, Manning (joined by six other former Escala shareholders) alleges that Merrill Lynch facilitated and engaged in naked short sales of Escala stock, in violation of New Jersey law. His complaint asserts that Merrill Lynch participated in “short sales at times when [it] neither possessed, nor had any intention of obtaining[,] sufficient stock” to deliver to buyers. App. to Pet. for Cert. 57a, Amended Complaint ¶39. That conduct, Manning charges, contravened provisions of the New Jersey Racketeer Influenced and Corrupt Organizations Act (RICO), New Jersey Criminal Code, and New Jersey Uniform Securities Law; it also, he adds, ran afoul of the New Jersey common law of negligence, unjust enrichment, and interference with contractual relations. See id., at 82a–101a, ¶¶88–161. Manning chose not to bring any claims under federal securities laws or rules. His complaint, however, referred explicitly to Regulation SHO, both describing the purposes of that rule and cataloguing past accusations against Merrill Lynch for flouting its requirements. See id., at 51a–54a, ¶¶28–30; 75a–82a, ¶¶81–87. And the complaint couched its description of the short selling at issue here in terms suggesting that Merrill Lynch had again violated that regulation, in addition to infringing New Jersey law. See id., at 57a–59a, ¶¶39–43.
Manning brought his complaint in New Jersey state court, but Merrill Lynch removed the case to Federal District Court. See 28 U. S. C. §1441 (allowing removal of any civil action of which federal district courts have original jurisdiction). Merrill Lynch asserted federal jurisdiction on two grounds. First, it invoked the general federal question statute, §1331, which grants district courts jurisdiction of “all civil actions arising under” federal law. Second, it maintained that the suit belonged in federal court by virtue of §27 of the Exchange Act. That provision, in relevant part, grants district courts exclusive jurisdiction of “all suits in equity and actions at law brought to enforce any liability or duty created by [the Exchange Act] or the rules and regulations thereunder.” 15 U. S. C. §78aa(a). Manning moved to remand the case to state court, arguing that neither statute gave the federal court authority to adjudicate his collection of state-law claims. The District Court denied his motion. See No. 12–4466 (D NJ, Mar. 18, 2013), App. to Pet. for Cert. 24a–38a.
The Court of Appeals for the Third Circuit reversed, ordering a remand of the case to state court. See 772 F. 3d 158 (2014). The Third Circuit first decided that the federal question statute, 28 U. S. C. §1331, did not confer jurisdiction of the suit, because all Manning’s claims were “brought under state law” and none “necessarily raised” a federal issue. 772 F. 3d, at 161, 163. Nor, the court held, did §27 of the Exchange Act make the district court the appropriate forum. Relying on this Court’s construction of a nearly identical jurisdictional provision, the Court of Appeals found that §27 covers only those cases involving the Exchange Act that would satisfy the “arising under” test of the federal question statute. See id., at 166–167 (citing Pan American Petroleum Corp. v Superior Court of Del. for New Castle Cty., 366 U. S. 656 (1961) ). Because the District Court lacked jurisdiction of Manning’s suit under §1331, so too it was not the exclusive forum under §27.
Merrill Lynch sought this Court’s review solely as to whether §27 commits Manning’s case to federal court. See Pet. for Cert. i. Because of a Circuit split about that provision’s meaning, we granted certiorari. 576 U. S. ___ (2015). We now affirm.
Like the Third Circuit, we read §27 as conferring exclusive federal jurisdiction of the same suits as “aris[e] under” the Exchange Act pursuant to the general federal question statute. See 28 U. S. C. §1331. The text of §27 more readily supports that meaning than it does either of the parties’ two alternatives. This Court’s precedents interpreting identical statutory language positively compel that conclusion. And the construction fits with our practice of reading jurisdictional laws, so long as consistent with their language, to respect the traditional role of state courts in our federal system and to establish clear and administrable rules.
Section 27, as noted earlier, provides federal district courts with exclusive jurisdiction “of all suits in equity and actions at law brought to enforce any liability or duty created by [the Exchange Act] or the rules and regulations thereunder.” 15 U. S. C. §78aa(a); see supra, at 3. Much the same wording appears in nine other federal jurisdictional provisions – mostly enacted, like §27, as part of New Dealera regulatory statutes.
Merrill Lynch argues that the “plain, unambiguous language” of §27 requires an expansive understanding of its scope. Brief for Petitioners 23. Whenever (says Merrill Lynch) a plaintiff’s complaint either explicitly or implicitly “assert[s]” that “the defendant breached an Exchange Act duty,” then the suit is “brought to enforce” that duty and a federal court has exclusive jurisdiction. Id., at 22; Reply Brief 10–11; see Tr. of Oral Arg. 7–8 (confirming that such allegations need not be express). That is so, Merrill Lynch contends, even if the plaintiff, as in this case, brings only state-law claims in his complaint – that is, seeks relief solely under state law. See Reply Brief 3–6. And it is so, Merrill Lynch continues, even if the plaintiff can prevail on those claims without proving that the alleged breach of an Exchange Act duty – here, the violation of Regulation SHO – actually occurred. See id., at 7–13; Tr. of Oral Arg. 3 (“[T]he words ‘brought to enforce’ [do not focus] on what the court would necessarily have to decide”).
But a natural reading of §27’s text does not extend so far. “Brought” in this context means “commenced,” Black’s Law Dictionary 254 (3d ed. 1933); “to” is a word “expressing purpose [or] consequence,” The Concise Oxford Dictionary 1288 (1931); and “enforce” means “give force [or] effect to,” 1 Webster’s New International Dictionary of the English Language 725 (1927). So §27 confers federal jurisdiction when an action is commenced in order to give effect to an Exchange Act requirement. That language, in emphasizing what the suit is designed to accomplish, stops short of embracing any complaint that happens to mention a duty established by the Exchange Act. Consider, for example, a simple state-law action for breach of contract, in which the plaintiff alleges, for atmospheric reasons, that the defendant’s conduct also violated the Exchange Act – or still less, that the defendant is a bad actor who infringed that statute on another occasion. On Merrill Lynch’s view, §27 would cover that suit; indeed, Merrill Lynch points to just such incidental assertions as the basis for federal jurisdiction here. See Brief for Petitioners 20–21; supra, at 3. But that hypothetical suit is “brought to enforce” state contract law, not the Exchange Act – because the plaintiff can get all the relief he seeks just by showing the breach of an agreement, without proving any violation of federal securities law. The suit, that is, can achieve all it is supposed to even if issues involving the Exchange Act never come up.
Critiquing Merrill Lynch’s position on similar grounds, Manning proposes a far more restrictive interpretation of §27’s language – one going beyond what he needs to prevail. See Brief for Respondents 27–33. According to Manning, a suit is “brought to enforce” the Exchange Act’s duties or liabilities only if it is brought directly under that statute – that is, only if the claims it asserts (and not just the duties it means to vindicate) are created by the Exchange Act. On that view, everything depends (as Justice Holmes famously said in another jurisdictional context) on which law “creates the cause of action.” American Well Works Co. v Layne & Bowler Co., 241 U. S. 257, 260 (1916). If a complaint asserts a right of action deriving from the Exchange Act (or an associated regulation), the suit must proceed in federal court. But if, as here, the complaint brings only state-created claims, then the case belongs in a state forum. And that is so, Manning claims, even if – contrary to what the Third Circuit held below – the success of the state claim necessarily hinges on proving that the defendant breached an Exchange Act duty. See Brief for Respondents 31.
Manning’s view of the text’s requirements, although better than Merrill Lynch’s, veers too far in the opposite direction. There is no doubt, as Manning says, that a suit asserting an Exchange Act cause of action fits within §27’s scope: Bringing such a suit is the prototypical way of enforcing an Exchange Act duty. But it is not the only way. On rare occasions, as just suggested, a suit raising a state-law claim rises or falls on the plaintiff’s ability to prove the violation of a federal duty. See, e.g., Grable & Sons Metal Products, Inc. v Darue Engineering & Mfg., 545 U. S. 308–315 (2005); Smith v Kansas City Title & Trust Co., 255 U. S. 180, 201 (1921). If in that manner, a state-law action necessarily depends on a showing that the defendant breached the Exchange Act, then that suit could also fall within §27’s compass. Suppose, for example, that a state statute simply makes illegal “any violation of the Exchange Act involving naked short selling.” A plaintiff seeking relief under that state law must undertake to prove, as the cornerstone of his suit, that the defendant infringed a requirement of the federal statute. (Indeed, in this hypothetical, that is the plaintiff’s only project.) Accordingly, his suit, even though asserting a state-created claim, is also “brought to enforce” a duty created by the Exchange Act.
An existing jurisdictional test well captures both classes of suits “brought to enforce” such a duty. As noted earlier, 28 U. S. C. §1331 provides federal jurisdiction of all civil actions “arising under” federal law. See supra, at 3. This Court has found that statutory term satisfied in either of two circumstances. Most directly, and most often, federal jurisdiction attaches when federal law creates the cause of action asserted. That set of cases is what Manning highlights in offering his view of §27. But even when “a claim finds its origins” in state law, there is “a special and small category of cases in which arising under jurisdiction still lies.” Gunn v Minton, 568 U. S. ___, ___ (2013) (slip op., at 6) (internal quotation marks omitted). As this Court has explained, a federal court has jurisdiction of a state-law claim if it “necessarily raise[s] a stated federal issue, actually disputed and substantial, which a federal forum may entertain without disturbing any congressionally approved balance” of federal and state power. Grable, 545 U. S., at 314; see Gunn, 568 U. S., at ___ (slip op., at 6) (framing the same standard as a four-part test). That description typically fits cases, like those described just above, in which a state-law cause of action is “brought to enforce” a duty created by the Exchange Act because the claim’s very success depends on giving effect to a federal requirement. Accordingly, we agree with the court below that §27’s jurisdictional test matches the one we have formulated for §1331, as applied to cases involving the Exchange Act. If (but only if) such a case meets the “arising under” standard, §27 commands that it go to federal court.
Merrill Lynch objects that our rule construes “completely different language” – i.e., the phrases “arising under" and “brought to enforce” in §1331 and §27, respectively – “to mean exactly the same thing.” Reply Brief 7. We cannot deny that point. But we think it far less odd than Merrill Lynch does. After all, the test for §1331 jurisdiction is not grounded in that provision’s particular phrasing. This Court has long read the words “arising under” in Article III to extend quite broadly, “to all cases in which a federal question is ‘an ingredient’ of the action.” Merrell Dow Pharmaceuticals Inc. v Thompson, 478 U. S. 804(1986) (quoting Osborn v Bank of United States, 9 Wheat. 738, 823 (1824)). In the statutory context, however, we opted to give those same words a narrower scope “in the light of [§1331’s] history[,] the demands of reason and coherence, and the dictates of sound judicial policy.” Romero v International Terminal Operating Co., 358 U. S. 354, 379 (1959). Because the resulting test does not turn on §1331’s text, there is nothing remarkable in its fitting as, or even more, neatly a differently worded statutory provision.
Nor can Merrill Lynch claim that Congress’s use of the new “brought to enforce” language in §27 shows an intent to depart from a settled (even if linguistically ungrounded) test for statutory “arising under” jurisdiction. That is because no such well-defined test then existed. As we recently noted, our caselaw construing §1331 was for many decades – including when the Exchange Act passed – highly “unruly.” Gunn, 568 U. S., at __ (slip op., at 6) (referring to the “canvas” of our old opinions as “look[ing] like one that Jackson Pollock got to first”). Against that muddled backdrop, it is impossible to infer that Congress, in enacting §27, wished to depart from what we now understand as the “arising under” standard.
This Court has reached the same conclusion before. In two unrelated decisions, we addressed the “brought to enforce” language at issue here. See Pan American, 366 U. S. 656; Matsushita Elec. Industrial Co. v Epstein, 516 U. S. 367 (1996). Each time, we viewed that phrase as coextensive with our construction of “arising under.”
Pan American involved §22 of the Natural Gas Act (NGA), 15 U. S. C. §717u – an exclusive jurisdiction provision containing language materially indistinguishable from §27’s. The case began in state court when a natural gas purchaser sued a producer for breach of a contract setting sale prices. Prior to the alleged breach, the producer had filed those contractual rates with the Federal Power Commission, as the NGA required. Relying on that submission (which the complaint did not mention), the producer claimed that the buyer’s suit was “brought to enforce” a liability deriving from the NGA – i.e., a filed rate – and so must proceed in federal court. See 366 U. S., at 662. This Court rejected the argument.
Our decision explained that §22’s use of the term “brought to enforce,” rather than “arising under,” made no difference to the jurisdictional analysis. The inquiry, we wrote, was “not affected by want” of the language contained in the federal question statute. Id., at 665, n. 2. The “limitation[s]” associated with “arising under” jurisdiction, we continued, were “clearly implied” in §22’s alternative phrasing. Ibid. In short, the linguistic distinction between the two jurisdictional provisions did not extend to their meaning.
Pan American thus went on to analyze the jurisdictional issue in the manner set out in our “arising under” precedents. Federal question jurisdiction lies, the Court wrote, only if “it appears from the face of the complaint that determination of the suit depends upon a question of federal law.” Id., at 663. That inquiry focuses on “the particular claims a suitor makes” in his complaint – meaning, whether the plaintiff seeks relief under state or federal law. Id., at 662. In addition, the Court suggested, a federal court could adjudicate a suit stating only a state-law claim if it included as “an element, and an essential one,” the violation of a federal right. Id., at 663 (quoting Gully v First Nat. Bank in Meridian, 299 U. S. 109, 112 (1936)). With those principles of “arising under” jurisdiction laid out, the Court held that §22 did not enable a federal court to resolve the buyer’s case, because he could prevail merely by proving breach of the contract. See 366 U. S., at 663–665. Pan American establishes, then, that an action “brought to enforce” a duty or liability created by a federal statute is nothing more (and nothing less) than an action “arising under” that law.
Merrill Lynch reads Pan American more narrowly, as holding only that §22 does not confer federal jurisdiction when a complaint (unlike Manning’s) fails to reference federal law at all. See Brief for Petitioners 32–33, 38. But that argument ignores Pan American’s express statement of equivalence between §27’s language and the federal question statute’s: “Brought to enforce” has the same “limitation[s]” (meaning, the same scope) as “arising under.” 366 U. S., at 665, n. 2. And just as important, Merrill Lynch disregards Pan American’s analytical structure: The decision proceeds by reviewing this Court’s “arising under” precedents, articulating the principles animating that caselaw, and then applying those tenets to the dispute at hand. Id., at 662–665. The Court thus showed (as well as told) that “brought to enforce” jurisdiction mirrors that of “arising under.”
As a fallback, Merrill Lynch claims that Pan American is irrelevant here because it relied on legislative history distinct to the NGA in finding §22’s “brought to enforce” language coterminous with “arising under.” See Brief for Petitioners 38–39. The premise of that argument is true enough: In support of its holding, the Court quoted a Committee Report describing §22 as conferring federal jurisdiction “over cases arising under the act.” 366 U. S., at 665, n. 2. But we cannot accept the conclusion Merrill Lynch draws from that statement: that courts should give two identically worded statutory provisions, passed less than five years apart, markedly different meanings. Indeed, the result of Merrill Lynch’s approach is still odder, for what of the eight other jurisdictional provisions containing “brought to enforce” language? See n. 3, supra. Presumably, Merrill Lynch would have courts inspect each of their legislative histories to decide whether to read those statutes as reproducing the “arising under” standard, adopting Merrill Lynch’s alternative view, or demanding yet another jurisdictional test. We are hard pressed to imagine a less sensible way of construing the repeated iterations of the phrase “brought to enforce” in the jurisdictional provisions of the Federal Code.
In any event, this Court in Matsushita addressed §27 itself, and once again equated the “brought to enforce” and “arising under” standards. That decision arose from a state-law action against corporate directors for breach of fiduciary duty. The issue was whether the state court handling the suit could approve a settlement releasing, in addition to the state claims actually brought, potential Exchange Act claims that §27 would have committed to federal court. In deciding that the state court could do so, we described §27 – not once, not twice, but three times – as conferring exclusive jurisdiction of suits “arising under” the Exchange Act. See 516 U. S., at 380 (Section 27 “confers exclusive jurisdiction upon the federal courts for suits arising under the [Exchange] Act”); id., at 381 (Section 27 “prohibits state courts from adjudicating claims arising under the Exchange Act”); id., at 385 (Section 27 “prohibit[s] state courts from exercising jurisdiction over suits arising under the Exchange Act”) (emphases added). Over and over, then, the Court took as a given that §27’s jurisdictional test mimicked the one in the general federal question statute.
And still more: The Matsushita Court thought clear that the suit as filed – which closely resembled Manning’s in its mix of state and federal law – fell outside §27’s grant of exclusive jurisdiction. As just noted, the claims brought in the Matsushita complaint sought relief for breach of a state-law duty. But in support of those claims, the plaintiffs charged, much as Manning did here, that the defendants’ conduct also violated federal securities laws. See 516 U. S., at 370; supra, at 2–3. We found the presence of that accusation insufficient to trigger §27. “[T]he cause pleaded,” we wrote, remained “a state common-law action,” 516 U. S., at 382, n. 7: Notwithstanding the potential federal issue, the suit “was not ‘brought to enforce’ any rights or obligations under the [Exchange] Act,” id., at 381. The Court thus rejected the very position Merrill Lynch takes here – i.e., that §27 precludes a state court from adjudicating any case, even if brought under state law, in which the plaintiff asserts an Exchange Act breach.
Construing §27, consistent with both text and precedent, to cover suits that arise under the Exchange Act serves the goals we have consistently underscored in interpreting jurisdictional statutes. Our reading, unlike Merrill Lynch’s, gives due deference to the important role of state courts in our federal system. And the standard we adopt is more straightforward and administrable than the alternative Merrill Lynch offers.
Out of respect for state courts, this Court has time and again declined to construe federal jurisdictional statutes more expansively than their language, most fairly read, requires. We have reiterated the need to give “[d]ue regard [to] the rightful independence of state governments” – and more particularly, to the power of the States “to provide for the determination of controversies in their courts.” Romero, 358 U. S., at 380 (quoting Healy v Ratta, 292 U. S. 263, 270 (1934); Shamrock Oil & Gas Corp. v Sheets, 313 U. S. 100, 109 (1941). Our decisions, as we once put the point, reflect a “deeply felt and traditional reluctance .... to expand the jurisdiction of federal courts through a broad reading of jurisdictional statutes.” Romero, 358 U. S., at 379. That interpretive stance serves, among other things, to keep state-law actions like Manning’s in state court, and thus to help maintain the constitutional balance between state and federal judiciaries.
Nor does this Court’s concern for state court prerogatives disappear, as Merrill Lynch suggests it should, in the face of a statute granting exclusive federal jurisdiction. See Brief for Petitioners 23–27. To the contrary, when a statute mandates, rather than permits, federal jurisdiction – thus depriving state courts of all ability to adjudicate certain claims – our reluctance to endorse “broad reading[s],” Romero, 358 U. S., at 379, if anything, grows stronger. And that is especially so when, as here, the construction offered would place in federal court actions bringing only claims created by state law – even if those claims might raise federal issues. To be sure, a grant of exclusive federal jurisdiction, as Merrill Lynch reminds us, indicates that Congress wanted “greater uniformity of construction and more effective and expert application” of federal law than usual. Brief for Petitioners 24 (quoting Matsushita, 516 U. S., at 383). But “greater” and “more” do not mean “total,” and the critical question remains how far such a grant extends. In resolving that issue, we will not lightly read the statute to alter the usual constitutional balance, as it would by sending actions with all state-law claims to federal court just because a complaint references a federal duty.
Our precedents construing other exclusive grants of federal jurisdiction illustrate those principles. In Pan American, for example, we denied that a state court’s resolution of state-law claims potentially implicating the NGA’s meaning would “jeopardize the uniform system of regulation” that the statute established. 366 U. S., at 665. We reasoned that this Court’s ability to review state court decisions of federal questions would sufficiently protect federal interests. And similarly, in Tafflin v Levitt, 493 U. S. 455–467 (1990), we permitted state courts to adjudicate civil RICO actions that might raise issues about the scope of federal crimes alleged as predicate acts, even though federal courts have exclusive jurisdiction “of all offenses against the laws of the United States,” 18 U. S. C. §3231. There, we expressed confidence that state courts would look to federal court interpretations of the relevant criminal statutes. Accordingly, we saw “no significant danger of inconsistent application of federal criminal law” and no “incompatibility with federal interests.” Tafflin, 493 U. S., at 464–465, 467 (internal quotation marks omitted).
So too here, when state courts, in deciding state-law claims, address possible issues of the Exchange Act’s meaning. Not even Merrill Lynch thinks those decisions wholly avoidable: It admits that §27 does nothing to prevent state courts from resolving Exchange Act questions that result from defences or counterclaims. See Brief for Petitioners 32–33; Pan American, 366 U. S., at 664–665. We see little difference, in terms of the uniformity-based policies Merrill Lynch invokes, if those issues instead appear in a complaint like Manning’s. And indeed, Congress likely contemplated that some complaints intermingling state and federal questions would be brought in state court: After all, Congress specifically affirmed the capacity of such courts to hear state-law securities actions, which predictably raise issues coinciding, overlapping, or intersecting with those under the Act itself. See 15 U. S. C. §78bb(a)(2); Matsushita, 516 U. S., at 383. So, for example, it is hardly surprising in a suit like this one, alleging short sales in violation of state securities law, that a plaintiff might say the defendant previously breached a federal prohibition of similar conduct. See supra, at 2–3 (describing Manning’s complaint). And it is less troubling for a state court to consider such an issue than to lose all ability to adjudicate a suit raising only state-law causes of action.
Reading §27 in line with our §1331 caselaw also promotes “administrative simplicity[, which] is a major virtue in a jurisdictional statute.” Hertz Corp. v Friend, 559 U. S. 77, 94 (2010). Both judges and litigants are familiar with the “arising under” standard and how it works. For the most part, that test provides ready answers to jurisdictional questions. And an existing body of precedent gives guidance whenever borderline cases crop up. See supra, at 8–9. By contrast, no one has experience with Merrill Lynch’s alternative standard, which would spring out of nothing to govern suits involving not only the Exchange Act but up to nine other discrete spheres of federal law. See n. 3, supra (listing statutes with “brought to enforce” language); supra, at 12–13 (noting Merrill Lynch’s backup claim that legislative histories might compel different tests for different statutes). Adopting such an untested approach, and forcing courts to toggle back and forth between it and the “arising under” standard, would undermine consistency and predictability in litigation. That result disserves courts and parties alike.
Making matters worse, Merrill Lynch’s rule is simple for plaintiffs to avoid – or else, excruciating for courts to police. Under that rule, a plaintiff electing to bring state-law claims in state court will purge his complaint of any references to federal securities law, so as to escape removal. Such omissions, after all, will do nothing to change the way the plaintiff can present his case at trial; they will merely make the complaint less informative. Recognizing the potential for that kind of avoidance, Merrill Lynch argues that a judge should go behind the face of a complaint to determine whether it is the product of “artful pleading.” See Tr. of Oral Arg. 7 (If the plaintiffs “had just literally whited out, deleted the references to Reg[ulation] SHO,” the court should still understand the complaint to allege a breach of that rule; “the fact [that the plaintiffs] didn’t cite it wouldn’t change the fact”). We have no idea how a court would make that judgment, and get cold comfort from Merrill Lynch’s assurance that the question would arise not in this case but in “the next third, fourth, fifth case down the road.” Id., at 8. Jurisdictional tests are built for more than a single dispute: That Merrill Lynch’s threatens to become either a useless drafting rule or a tortuous inquiry into artful pleading is one more good reason to reject it.
Section 27 provides exclusive federal jurisdiction of the same class of cases as “arise under” the Exchange Act for purposes of §1331. The text of §27, most naturally read, supports that rule. This Court has adopted the same view in two prior cases. And that reading of the statute promotes the twin goals, important in interpreting jurisdictional grants, of respecting state courts and providing administrable standards.
Our holding requires remanding Manning’s suit to state court. The Third Circuit found that the District Court did not have jurisdiction of Manning’s suit under §1331 because all his claims sought relief under state law and none necessarily raised a federal issue. See supra, at 3. Merrill Lynch did not challenge that ruling, and we therefore take it as a given. And that means, under our decision today, that the District Court also lacked jurisdiction under §27. Accordingly, we affirm the judgment below.
It is so ordered.
(with whom Justice Sotomayor joins, concurring in the judgment)
The Court concludes that respondents’ suit belongs in state court because it does not satisfy the multifactor, a textual standard that we have used to assess whether a suit is one “arising under” federal law, 28 U. S. C. §1331. Ante, at 18. I agree that this suit belongs in state court, but I would rest that conclusion on the statute before us, §27 of the Securities Exchange Act of 1934, 15 U. S. C. §78aa. That statute does not use the phrase “arising under” or provide a sound basis for adopting the arising-under standard. It instead provides federal jurisdiction where a suit is “brought to enforce” Exchange Act requirements. §78aa(a). That language establishes a straightforward test: If a complaint alleges a claim that necessarily depends on a breach of a requirement created by the Act, §27 confers exclusive federal jurisdiction over that suit. Because the complaint here does not allege such claims – and because no other statute confers federal jurisdiction – this suit should return to state court. Accordingly, I concur in the judgment.
Section 27 provides that “[t]he district courts .... shall have exclusive jurisdiction .... of all suits in equity and actions at law brought to enforce any liability or duty created by this chapter or the rules and regulations thereunder.” §78aa(a). As the Court explains, under a “natural reading,” §27 “confers federal jurisdiction when an action is commenced in order to give effect to an Exchange Act requirement.” Ante, at 6; see also Webster’s New International Dictionary of the English Language 725 (1927) (“enforce” means “give force to” or “give effect to”). And by providing “exclusive jurisdiction” to federal district courts over certain suits, §27 strips state courts of jurisdiction over such suits.
Put differently, under §27 a suit belongs in federal court when the complaint requires a court to enforce an Exchange Act duty or liability. In contrast, a suit belongs in state court when the complaint “assert[s] purely state-law causes of action” that do not require “binding legal determinations of rights and liabilities under the Exchange Act” or “a judgment on the merits of” an Exchange Act breach. Matsushita Elec. Industrial Co. v Epstein, 516 U. S. 367, 382, 384 (1996). Such a suit is “not ‘brought to enforce’ any rights or obligations under the Act,” and thus does not fall within §27’s scope. Id., at 381. So §27 does not provide federal jurisdiction over suits brought to enforce liabilities or duties under state law or over every case that happens to involve allegations that the Act was violated. The provision leaves state courts with some authority over suits involving the Act or its regulations.
The statutory context bolsters this understanding. That context confirms that Congress reserved some authority to state courts to adjudicate securities-law matters. Although the Act provides numerous federal “rights and remedies,” it also generally preserves “all other rights and remedies that may exist at law or in equity,” such as claims that could be litigated in state courts of general jurisdiction. 15 U. S. C. §78bb(a)(2). That provision shows that “Congress plainly contemplated the possibility of dual litigation in state and federal courts relating to securities transactions.” Matsushita, supra, at 383. A natural reading of §27’s text preserves the dual role for federal and state courts that Congress contemplated, and it confirms that mere allegations of Exchange Act breaches do not alone deprive state courts of jurisdiction.
A natural reading promotes the uniform interpretation of the federal securities laws that Congress sought to ensure when it gave federal courts “exclusive jurisdiction” over federal securities-law suits. §78aa(a). The textual approach fosters uniformity because it leaves to federal courts – which are presumptively more familiar with the intricate federal securities laws – the task of “adjudicat[ing] .... Exchange Act claims.” Matsushita, 516 U. S., at 383. When state courts decide cases where the complaint pleads only state-law claims and do not resolve the merits of Exchange Act rights or liabilities, they are not “trespass[ing] upon the exclusive territory of the federal courts.” Id., at 382.
The statutory text and structure thus support a straightforward test: Section 27 confers federal jurisdiction over a case if the complaint alleges claims that necessarily depend on establishing a breach of an Exchange Act requirement.
The Third Circuit was correct to remand this suit to state court. Respondents’ complaint does not seek “to enforce any liability or duty created by” the Exchange Act or its regulations. §78aa(a).
Although respondents’ complaint alleges at different places that petitioners violated the Exchange Act or its regulations, the complaint does not bring claims requiring enforcement of the Exchange Act or its regulations. The complaint instead brings 10 state-law causes of action that seek to enforce duties and liabilities created by state law. Count 2 alleges that petitioners violated state law by investing money derived from racketeering. See App. to Pet. for Cert. 91a–93a, Amended Complaint ¶¶114–122 (citing N. J. Stat. Ann. §2C:41–2a (West 2005)). Counts 3 through 9 allege standard state-law contract and tort claims: unjust enrichment, unlawful interference with economic advantage, tortious interference with contractual relations, unlawful interference with contractual relations, third-party-beneficiary claims, breach of the covenant of good faith and fair dealing, and negligence. See App. to Pet. for Cert. 93a–101a, Amended Complaint ¶¶123–158. Count 10 pleads a freestanding claim for punitive and exemplary damages. See id., at 101a, Amended Complaint ¶¶159–161. None of these claims requires a court to “enforce” – to give effect to – a requirement created by the Act, thus, §27 does not confer federal jurisdiction over them.
Count 1 presents a closer call, but it too does not trigger federal jurisdiction. That count pleads that petitioners violated a state law that makes it unlawful for a person to participate in a racketeering enterprise. Id., at 82a–90a, Amended Complaint ¶¶88–113 (citing N. J. Stat. Ann. §2C:41–2c). The alleged racketeering includes violating the New Jersey Uniform Securities Law (through fraud, deception, and misappropriation), committing “theft by taking” under state law, and committing “theft by deception” under state law. App. to Pet. for Cert. 82a–90a, Amended Complaint ¶¶88–113. Respondents allege that “[t]he SEC has expressly noted that naked short selling involves the omission of a material fact” as part of their state-law securities fraud allegation. Id., at 85a, Amended Complaint ¶100. Vindicating that claim would not require the enforcement of a federal duty or liability. New Jersey law encompasses fraudulent conduct that does not necessarily rest on a violation of federal law or regulation. See, e.g., §49:3–49(e)(1) (West 2001) (fraud and deceit include “[a]ny misrepresentation by word, conduct or in any manner of any material fact, either present or past, and any omission to disclose any such fact”); see App. to Pet. for Cert. 84a–86a (invoking §49:3–49 et seq.). So although Count 1 refers to the Securities and Exchange Commission’s view about naked short selling, that count does not require respondents to establish a violation of federal securities law to prevail on their fraud claim. Because respondents’ cause of action in Count 1 seeks to enforce duties and liabilities created by state law and does not necessarily depend on the breach of an Exchange Act duty or liability, §27 does not provide federal jurisdiction over that claim.
Although the Court acknowledges the “natural reading” of §27, ante, at 6, it holds that §27 adopts the jurisdictional test that this Court uses to evaluate federal-question jurisdiction under 28 U. S. C. §1331. See ante, at 8–10; see also ante, at 10–18. Federal courts have the power to review cases “arising under” federal law, §1331, including those in which the complaint brings state-law claims that “necessarily raise a stated federal issue, actually disputed and substantial, which a federal forum may entertain without disturbing any congressionally approved balance of federal and state judicial responsibilities.” Grable & Sons Metal Products, Inc. v Darue Engineering & Mfg., 545 U. S. 308, 314 (2005). The Court wrongly equates the phrase “arising under” in §1331 with the phrase “brought to enforce” in §27, and interprets the latter to require that a case raising state-law claims “mee[t] the ‘arising under’ standard” for that case to proceed in federal court. Ante, at 8; see ante, at 8–9. None of the Court’s rationales for adopting that rule is persuasive.
The Court first argues that “it is impossible to infer that Congress, in enacting §27, wished to depart from what we now understand as the ‘arising under’ standard” because there was no “well-defined test” to depart from. Ante, at 10. The Court’s case law construing §1331, the Court explains, “was for many decades – including when the Exchange Act passed – highly unruly.” Ibid. (internal quotation marks omitted).
But when Congress enacts a statute that uses different language from a prior statute, we normally presume that Congress did so to convey a different meaning. See, e.g., Crawford v Burke, 195 U. S. 176, 190 (1904) (explaining that “a change in phraseology creates a presumption of a change in intent” and that “Congress would not have used such different language [in two statutes] without thereby intending a change of meaning”). Given what we know about §1331, that presumption has force here. Our §1331 case law was, as the Court notes, “highly unruly” when the Exchange Act was enacted in 1934. Given the importance of clarity in jurisdictional statutes, see Hertz Corp. v Friend, 559 U. S. 77, 94 (2010), it is quite a stretch to infer that Congress wished to embrace such an unpredictable test.
That is especially true given that §27 does not use words supporting the convoluted arising-under standard. Section 27 does not ask (for example) whether a federal issue is substantial or whether a ruling on that issue will upset the congressionally approved balance of federal and state power. Indeed, §1331 itself does not even use words supporting the arising-under standard. See ante, at 10 (acknowledging that the arising-under standard “does not turn on §1331’s text”). Rather, the Court has refused to give full effect to §1331’s “broa[d] phras[ing]” and has instead “continuously construed and limited” that provision based on extra textual considerations, such as “history,” “the demands of reason and coherence,” and “sound judicial policy.” Romero v International Terminal Operating Co., 358 U. S. 354, 379 (1959). Faced with a plain and focused text like §27, however, we should not rely on such considerations. And importing factors from our §1331 arising-under jurisprudence – such as a substantiality requirement and a federal-state balance requirement – risks narrowing the class of cases that Congress meant to cover with §27’s plain text. For these reasons, it is unwise to read into §27 a decision to adopt the arising-under standard.
The Court next relies on two prior decisions – Pan American Petroleum Corp. v Superior Court of Del. for New Castle Cty., 366 U. S. 656 (1961), and Matsushita, 516 U. S. 367. See ante, at 10–14. Neither case justifies the Court’s decision to apply the arising-under standard to §27.
In Pan American, the Court held that Delaware state courts had jurisdiction over state-law contract claims that arose from contracts for the sale of natural gas. 366 U. S., at 662–665. The Court reached that decision even though a provision of the Natural Gas Act provided exclusive federal jurisdiction over suits “ ‘brought to enforce any liability or duty created by’ ” that Act. Id., at 662 (quoting statute). Pan American lends some support to the Court’s view today. It applied the Court’s arising-under precedents and “explained that [the Natural Gas Act’s] use of the term ‘brought to enforce,’ rather than ‘arising under,’ made no difference to the jurisdictional analysis.” Ante, at 11; see Pan American, supra, at 665, n. 2; see also ante, at 10–13.
But Pan American does not require the Court to engraft the arising-under standard onto §27. Pan American did not carefully analyze the Natural Gas Act’s text or assess the contemporary meaning of the central phrase “brought to enforce.” Instead, the Court relied on legislative history, reasoning that “authoritative [congressional] Committee Reports” implied a limitation on the Natural Gas Act’s jurisdictional text. 366 U. S., at 665, n. 2. That reasoning does not warrant our respect. That is especially true because Pan American’s holding is consistent with the Natural Gas Act’s “brought to enforce” language. The complaint in that case did not “asser[t]” any “right .... under the Natural Gas Act” and instead asked the court to adjudicate standard state-law “contract or quasi-contract” claims. Id., at 663, 664. The Court’s disposition in Pan American rests as comfortably on the statutory text as it does on the arising-under standard.
Matsushita provides even less support for the Court’s holding today. In that case the Court held that Delaware courts could issue a judgment approving a settlement releasing securities-law claims even though the settlement released claims that were (by virtue of §27) “solely within the jurisdiction of the federal courts.” 516 U. S., at 375; see id., at 370–372. The Court explained that, “[w]hile §27 prohibits state courts from adjudicating claims arising under the Exchange Act, it does not prohibit state courts from approving the release of Exchange Act claims in the settlement of suits over which they have properly exercised jurisdiction, i.e., suits arising under state law or under federal law for which there is concurrent jurisdiction.” Id., at 381. Because the complaint in that case “assert[ed] purely state-law causes of action” and the state courts did not issue “a judgment on the merits of the [exclusively federal] claims,” §27 did not deprive state courts of jurisdiction. Id., at 382.
The Court relies on Matsushita because in that case we three times “described” §27 “as conferring exclusive jurisdiction of suits ‘arising under’ the Exchange Act.” Ante, at 13 (citing 516 U. S., at 380, 381, 385). But Matsushita did not decide whether §27 adopts the arising-under standard, so its passing use of the phrase “arising under” cannot bear the weight that the Court now places on it. To be sure, Matsushita does support the Court’s judgment today: Matsushita emphasized that state courts could adjudicate a suit involving securities-law issues where the complaint “assert[ed] purely state-law causes of action” and did not require the state courts to issue “binding legal determinations of rights and liabilities under the Exchange Act” or “a judgment on the merits of” an Exchange Act breach. Id., at 382, 384. But those statements are more consistent with §27’s text than they are with the arising-under standard. See supra, at 2–3 (invoking Matsushita).
Finally, the Court argues that its interpretation “serves the goals” that our precedents have “consistently underscored in interpreting jurisdictional statutes” – affording proper deference to state courts and promoting administrable jurisdictional rules. Ante, at 14; see ante, at 14–18. But hewing to §27’s text serves these goals as well as or better than does adopting the arising-under standard.
First, the text-based view preserves state courts’ authority to adjudicate numerous securities-law claims and provide relief consistent with the Exchange Act’s design. See supra, at 1–3. As explained above, that view places all of respondents’ state-law causes of action in state court. See supra, at 3–5. The text-based view thus “decline[s] to construe [a] federal jurisdictional statut[e] more expansively than [its] language, most fairly read, requires.” Ante, at 14.
Second, the textual test is also more administrable than the arising-under standard. The arising-under standard “is anything but clear.” Grable, 545 U. S., at 321 (Thomas, J., concurring). The standard involves numerous judgments about matters of degree that are not readily susceptible to bright lines. As noted, to satisfy that standard, a state-law claim must raise a federal issue that is (among other things) “actually disputed,” is “substantial,” and will not “distur[b]” a congressionally approved federal-state “balance.” Id., at 314 (opinion of Court). The standard “calls for a ‘common-sense accommodation of judgment to [the] kaleidoscopic situations’ that present a federal issue, in ‘a selective process which picks the substantial causes out of the web and lays the other ones aside.’ ” Id., at 313 (quoting Gully v. First Nat. Bank in Meridian, 299 U. S. 109–118 (1936)). The arising-under standard may be many things, but it is not one that consistently “provides ready answers” to hard jurisdictional questions. Ante, at 17. The text-based view promises better. I would adopt that view and apply it here.
* * *
For these reasons, I concur in the judgment.
 Compare 772 F. 3d 158 (CA3 2014) (case below) with Barbara< v New York Stock Exchange, Inc., 99 F. 3d 49, 55 (CA2 1996) (construing §27 more narrowly), Sparta Surgical Corp. v National Assn. of Securities Dealers, Inc., 159 F. 3d 1209, 1211–1212 (CA9 1998) (construing §27 more broadly), and Hawkins v National Assn. of Securities Dealers, Inc., 149 F. 3d 330, 331–332 (CA5 1998) (per curiam) (same).
 Section 27 also grants federal courts exclusive jurisdiction of “violations of [the Exchange Act] or the rules and regulations thereunder.” 15 U. S. C. §78aa(a). Manning argues that the “violations” language applies only to criminal proceedings and SEC enforcement actions. See Brief for Respondents 28. Merrill Lynch, although not conceding that much, believes the “violations” clause irrelevant here because, in private suits for damages, it goes no further than the “brought to enforce” language quoted in the text. See Reply Brief 1, n. 1. Given that both parties have thus taken the “violations” language off the table, we do not address its meaning.
 See Securities Act of 1933, 15 U. S. C. §77v(a); Federal Power Act of 1935, 16 U. S. C. §825p; Connally Hot Oil Act of 1935, 15 U. S. C. §715i(c); Natural Gas Act of 1938, 15 U. S. C. §717u; Trust Indenture Act of 1939, 15 U. S. C. §77vvv(b); Investment Company Act of 1940, 15 U. S. C. §80a–43; Investment Advisers Act of 1940, 15 U. S. C. §80b–14(a); International Wheat Agreement Act of 1949, 7 U. S. C. §1642(e); Interstate Land Sales Full Disclosure Act of 1968, 15 U. S. C. §1719.
 The concurrence adopts a slightly different approach, placing in federal court Exchange Act claims plus all state-law claims necessarily raising an Exchange Act issue. See post, at 2–3 (Thomas, J., concurring in judgment). In other words, the concurrence would not ask, as the “arising under” test does, whether the federal issue embedded in such a state-law claim is also substantial, actually disputed, and capable of resolution in federal court without disrupting the congressionally approved federal-state balance. See post, at 6–7; Grable & Sons Metal Products, Inc. v Darue Engineering & Mfg., 545 U. S. 308, 314 (2005). But this Court has not construed any jurisdictional statute, whether using the words “brought to enforce” or “arising under” (or for that matter, any other), to draw the concurrence’s line. For as long as we have contemplated exercising federal jurisdiction over state-law claims necessarily raising federal issues, we have inquired as well into whether those issues are “really and substantially” disputed. See, e.g., Hopkins v Walker, 244 U. S. 486, 489 (1917); Shulthis v McDougal, 225 U. S. 561, 569 (1912). And similarly, we have long emphasized the need in such circumstances to make “sensitive judgments about congressional intent, judicial power, and the federal system.” Merrell Dow Pharmaceuticals Inc. v Thompson, 478 U. S. 804, 810 (1986). At this late juncture, we see no virtue in trying to pull apart these interconnected strands of necessity and substantiality-plus. Indeed, doing so here – and thus creating a gap between our “brought to enforce” and “arising under” standards – would conflict with this Court’s precedent and undermine important goals of interpreting jurisdictional statutes. See infra, at 10–14 (discussing our prior decisions equating the two tests), 14–17 (highlighting the need to respect state courts and the benefits of using a single, time-tested standard).
 Section 22 grants federal courts exclusive jurisdiction “of all suits in equity and actions at law brought to enforce any liability or duty created by .... [the NGA] or any rule, regulation, or order thereunder.” 52Stat. 833.
 The Romero Court continued: “A reluctance which must be even more forcefully felt when the expansion is proposed, for the first time, eighty-three years after the jurisdiction has been conferred.” 358 U. S., at 379. The Exchange Act was passed a mere 82 years ago, but we believe the point still stands.
 As the Court explains, the parties have not pressed us to construe §27’s language conferring jurisdiction over “violations” of the Exchange Act, its rules, or its regulations. See ante, at 5, n. 2. Like the Court, I focus on §27’s “brought to enforce” language.
all rights reserved