Declining Supplemental Jurisdiction

Declining Supplemental Jurisdiction

Satisfying § 1367(a) gives a federal court the power to hear a related claim, but having power does not oblige the court to use it. Subsection (c) addresses that gap. It opens by saying that the district courts "may decline to exercise supplemental jurisdiction over a claim under subsection (a)" in four listed situations.¹ The word "may" matters, because it makes the decision the court's and gives the plaintiff no right to have the claim heard. This rule applies to every exercise of supplemental jurisdiction, whether the claim is a pendent state claim attached to a federal claim or an ancillary claim arising out of a joinder device.²

The discretion comes from United Mine Workers v. Gibbs, which described pendent jurisdiction as a doctrine of discretion rather than a right belonging to the plaintiff.³ The reason follows from why the power exists in the first place. A federal court may hear a state claim because trying related claims together saves effort for the court and the parties. Gibbs put the justification in terms of judicial economy, convenience, and fairness to litigants, and said a federal court should hesitate to exercise jurisdiction over state claims when those considerations are missing.⁴ Comity, meaning respect for the state courts as the primary interpreters of state law, adds a second reason. Needless federal decisions on state law risk error and give the parties a less sure reading of the law than a state court would.⁵ Section 1367(c) confirms that supplemental jurisdiction remains discretionary and carries over the situations Gibbs identified.⁶

Courts disagree about how much of the Gibbs reasoning still governs. Congress did not repeat the references to economy, convenience, and fairness in the text of (c). Many courts nonetheless consult those factors in deciding whether to decline, while the Ninth Circuit has treated them as relevant only to the "exceptional circumstances" category.⁷ A decision to decline also leaves the federal claim untouched. Suppose a plaintiff sues a police officer under 42 U.S.C. § 1983 and adds a state battery claim arising from the same arrest. If the court declines the battery claim, it must still decide the § 1983 claim, because it has jurisdiction over that claim and a duty to exercise it.⁸ The plaintiff keeps the federal forum for the federal claim, and only the related state claim goes elsewhere.

Novel or Complex State Law

The first ground is that the claim raises a novel or complex issue of state law.¹⁰ The reason is federalism. When a state's highest court has not resolved a question, a federal court that decides it must guess what the state court would hold. Keeping the claim in federal court also denies the state supreme court the chance to answer the question itself, unless the federal court certifies it.¹¹ Gibbs made the same point when it said that avoiding needless decisions of state law gives the parties a surer-footed reading of the law.¹²

Suppose a plaintiff sues in federal court under a federal consumer statute and adds a state-law claim that depends on a newly enacted state statute no state appellate court has construed. That claim fits (c)(1), and the court may send it to state court. Change one fact so that the state claim rests on a long-settled common-law rule, and the federalism concern largely disappears, so the court has little reason to decline on this ground.

Predominating State Claims

The second ground applies when the state claim substantially predominates over the claim or claims within original jurisdiction.¹⁴ Gibbs measured predominance in three ways: the proof required, the scope of the issues raised, and the comprehensiveness of the remedy sought.¹⁵ The concern is that a litigant may attach a thin federal claim to what is really a state case. Once the state claim is the real body of the case and the federal claim only an appendage, the state claim may fairly be dismissed.¹⁶ Predominance may become apparent only after pretrial work or trial, so the question stays open throughout the litigation.¹⁷

The boundary is shown by Gibbs itself. The state claim there was broader than the federal one, and only the state claim ended in recovery. The Court still found no abuse of discretion in keeping it, because the federal issues were not so remote that the state claim alone was effectively tried, jury confusion could be reduced with a special verdict form, and a preemption question made the federal forum especially suitable.¹⁸ Now imagine a plaintiff who pleads one small federal claim and five state-law theories, with nearly all the discovery, witnesses, and damages directed at the state theories. That pattern points toward declining under (c)(2). The difference is how much of the case the federal claim actually occupies.

Dismissal of All Federal Claims

The third ground applies when the district court has dismissed all claims over which it had original jurisdiction.²⁰ Supplemental jurisdiction exists only because some anchor claim brings the case into federal court, so when the anchor falls away the reason for a federal forum falls with it. Gibbs said that if the federal claims are dismissed before trial, the state claims should be dismissed as well.²¹ The usual course is dismissal without prejudice, which leaves the plaintiff free to refile in state court.²² Section 1367(d) then tolls the limitations period while the claim was pending and for 30 days afterward, unless state law gives longer, so the dismissal does not cost the plaintiff the claim.²³

The discretion is strongest early in the case. Suppose a plaintiff sues the United States under a federal statute and adds state-law claims against two private defendants, with no independent basis for federal jurisdiction over them. If the court dismisses the federal claim under Rule 12(b)(6) soon after the pleadings close, it will probably decline the state claims against the other two.²⁴ Now suppose the federal claim is dismissed only after extensive discovery and trial preparation on both claims, and the state law involved is settled. Subsection (c)(3) still permits dismissal, but economy and fairness now favor keeping the state claim, because the court has already invested in it and a state court would have to start over. Subsection (c)(3) says the court may decline, not that it must, so retention in that setting is within its discretion.²⁵

Exceptional Circumstances

The fourth ground is a catch-all: in exceptional circumstances, there are other compelling reasons for declining jurisdiction.²⁷ Because the first three grounds cover the usual situations, this one is reserved for unusual cases, and the words "exceptional" and "compelling" signal that the court needs a strong reason. The Gibbs considerations of economy, convenience, fairness, and comity are the natural source of such reasons, and many courts consult them here, though the Ninth Circuit confines them to this ground alone.²⁸ Gibbs also recognized a separate reason, the likelihood of jury confusion in trying divergent legal theories together, which could justify separating the claims for trial under Rule 42(b).²⁹

The example below is invented. Suppose a federal claim and a state claim apply different standards of proof, and the court concludes that no instruction can keep a jury from blending them. That risk of confusion is a compelling reason of the kind (c)(4) contemplates. In Gibbs the same risk existed, but a special verdict form could reduce it, and the Court found no error in keeping the state claim.³⁰ The difference is whether the problem can be managed inside a single federal trial.

Novel or Complex State Law

The first listed reason for declining a supplemental claim is that the claim "raises a novel or complex issue of State law."³² The ground rests on federalism. A federal court that hears a state claim must apply state law to it, and when the state's highest court has never addressed the question, the federal judge has to predict how that court would rule. A prediction can be wrong, and even a careful one lacks the authority of an answer from the state's own court. Retaining the claim also keeps the state supreme court from ever speaking on the point, unless the federal court takes the extra step of certifying the question to it.³³

This ground applies a theme that Gibbs had already set out. The Court there said that needless decisions of state law should be avoided as a matter of comity, meaning respect for state courts as the primary interpreters of state law. It added that avoiding them also gives the parties a surer-footed reading of the law that governs them.³⁴ Subsection (c)(1) turns that concern into a named ground for declining, and so it is the natural choice when the state question is unsettled and the answer matters beyond the case at hand. The ground is aimed at the difficulty of the state-law question, not at the size of the claim or the amount of money involved.

Two contrasting examples show where the line falls. Suppose a plaintiff sues in federal court on a federal statutory claim and adds a state claim that turns on an unresolved question of state constitutional law. The state's courts have not spoken on the issue, so the district court may send the claim to state court under (c)(1). Now suppose the added claim is a routine battery claim governed by well-settled state law. The federal judge has no real guessing to do, the federalism concern largely disappears, and (c)(1) gives little reason to decline. The only fact that changed is whether the state's own courts have yet had the chance to settle the question. Even when (c)(1) applies, the court still has discretion, because the statute says only that the court "may" decline.³⁵

Predominating State Claims

The second ground in § 1367(c) lets a district court decline a supplemental claim when it "substantially predominates over the claim or claims over which the district court has original jurisdiction."³⁷ The reason lies in why supplemental jurisdiction exists at all. A related state claim rides into federal court because an anchor claim already justifies the federal forum and trying the two together saves effort. When the state claim becomes the main event, that logic runs backward, and the federal claim serves only as the entry ticket. A plaintiff would otherwise have an incentive to attach a thin federal claim to what is really a state-law dispute in order to litigate it in federal court.³⁸

Gibbs supplied the measuring sticks. Predominance may be judged in terms of the proof the claims require, the scope of the issues they raise, and the comprehensiveness of the remedy sought, and a state claim that predominates may be dismissed without prejudice and left to the state courts.³⁹ The Court also gave the idea a vivid image: once a state claim is the real body of the case and the federal claim is only an appendage, the state claim may fairly be dismissed.⁴⁰ The inquiry does not end at the pleadings, because pretrial work or the trial itself may reveal a state-law dominance that no one could see at the outset.⁴¹

Two contrasting situations show where the line falls. Imagine a plaintiff who pleads one small federal claim alongside five state-law theories, with nearly all of the witnesses, documents, and damages directed at the state theories. The state claims are the real case, and the court may decline them under (c)(2). Now imagine that the state and federal claims rest on the same evidence and the federal claim is the heart of the dispute. Nothing predominates, and the economy that justifies joint trial remains intact. Gibbs itself sits near the boundary. The only recovery there was on the state claim, yet the Court found no abuse of discretion in keeping it, because the federal claims were substantial, both sets of claims arose from the same nucleus of operative fact and sought alternative remedies, and the federal issues were not so remote that the state claim alone was effectively tried.⁴² Even when predominance exists, the statute says only that the court "may" decline, so the decision stays within its discretion.⁴³

Dismissal of All Federal Claims

The third ground for declining supplemental jurisdiction looks at what has happened to the anchor claim. Section 1367(c)(3) permits a district court to decline a supplemental claim when "the district court has dismissed all claims over which it has original jurisdiction."⁴⁵ The word "all" matters. As long as one claim within original jurisdiction survives, the case still belongs in federal court and the first three grounds do not apply in this form. The ground opens only when the last claim that carried the case into federal court is gone.

The reason follows from why supplemental jurisdiction exists. A related state claim is allowed into federal court because an anchor claim already justifies the forum and trying the two together saves effort for the court and the parties. Once the anchor is dismissed, the state claim sits in a federal court that has no remaining federal reason to hear it, and the state law questions are the only matters left. Gibbs addressed the situation directly: if the federal claims are dismissed before trial, even though they were not insubstantial in a jurisdictional sense, the state claims should be dismissed as well.⁴⁶ The dismissal says nothing about the merits of the state claim, so the usual course is to dismiss it without prejudice. That leaves the plaintiff free to refile in state court, where the claim can be heard as a state case. Section 1367(d), taught later in this chapter, deals with the limitations problem that refiling would otherwise create.

The word "should" in Gibbs states an expectation. It does not state a command, and the Court made that clear by calling pendent jurisdiction a doctrine of discretion, not of the plaintiff's right.⁴⁷ Dismissing the federal claim does not strip the court of power over the state claim, because the power was fixed when the case began. The court keeps the power and decides whether to use it. In River Park, Inc. v. City of Highland Park, the Illinois Supreme Court relied on that point. The plaintiffs there had lost their federal civil rights claim in federal court and later argued that their state claims were not barred by claim preclusion because the federal court would have had to dismiss them too. The court rejected the argument: a district court is not required to dismiss pendent state claims after dismissing the claim that supplied original jurisdiction, and § 1367(c) leaves the choice to the court.⁴⁸ Because the plaintiffs never put the state claims before the federal court, nobody could know whether it would have declined them, and federal courts had retained similar claims in similar circumstances.⁴⁹ A plaintiff with related state claims therefore does better to plead them in the federal action than to count on a later state suit.

How far the case has progressed explains why the choice varies. Early in the case, the federal court has invested little in the state claim, so sending it away costs little. Suppose an employee sues an employer under a federal discrimination statute and adds a state-law contract claim, and the court grants a Rule 12(b)(6) motion on the federal claim soon after the pleadings close. Dismissing the contract claim without prejudice is the natural result. This is the situation in which the discretion is especially likely to be exercised.⁵⁰ Now change the timing. Suppose the federal claim is lost on the eve of trial, after extensive discovery and motion practice on both claims, and the state law involved is settled. A state court would have to start over, and the parties have already paid for a federal-court record. Economy, convenience, and fairness, the considerations that justify supplemental jurisdiction in the first place, now favor keeping the claim.⁵¹ The court may still dismiss, but it is not compelled to, and the late stage gives it good reason to retain.

Exceptional Circumstances

The fourth ground in § 1367(c) is a catch-all. A district court may decline a supplemental claim when, "in exceptional circumstances, there are other compelling reasons for declining jurisdiction."⁵³ The first three grounds name the usual situations: unsettled state law, a predominating state claim, and the loss of every federal anchor claim. This one catches what Congress could not list in advance. An open-ended clause carries a risk, though, because a court that could decline whenever it saw some reason would hollow out the broad grant in subsection (a). The words "exceptional" and "compelling" guard against that. They require a reason that is unusual in kind and strong in weight, so ordinary inconvenience or a crowded docket does not qualify. Those burdens attend almost every case, and if they sufficed, a court could decline nearly any supplemental claim.

The Federal Circuit's decision in Voda v. Cordis Corp. shows the ground at work. Voda, an Oklahoma resident, held three U.S. patents on catheter use in heart surgery, along with a European patent and separate British, Canadian, French, and German patents. He sued Cordis, a Florida corporation, in federal court in Oklahoma. The U.S. patent claims fell within the court's original jurisdiction under 28 U.S.C. § 1338, and he added claims for infringement of the foreign patents. The district court exercised supplemental jurisdiction over those added claims on the theory that they arose from the same transaction as the U.S. claims and so formed part of one Article III case.⁵⁴ The Federal Circuit held that this was an abuse of discretion. The Paris Convention and TRIPS leave each country's patent system independent of the others, and nothing in those treaties suggests that one nation's courts should adjudicate another nation's patents. Exercising jurisdiction would therefore harm international comity.⁵⁵ The district court's power under subsection (a) was not in doubt. The question was whether to use that power, and the answer was no.

Voda also shows where the considerations from Gibbs (judicial economy, convenience, fairness, and comity) still matter. Subsection (c) does not mention them, and courts disagree about how far they reach. Many courts continue to consult them across all four grounds.⁵⁶ The Ninth Circuit confines them to the exceptional-circumstances ground, on the view that the first three grounds are clear on their face and merely codify what courts already did.⁵⁷ Everyone agrees that (c)(4) is the place where those considerations most naturally supply the "compelling reasons." The court in Voda relied on exactly this list.

A matched pair shows where the line falls. A federal court asked to decide foreign patent claims alongside related U.S. patent claims would be ruling on rights that another sovereign granted and governs, and that intrusion into another nation's system supports declining under (c)(4). Now suppose a district court has a heavy docket, and a supplemental state-law claim would add months to the schedule. The court is inconvenienced, but no interest beyond its own administration is at stake, and congestion of that kind is not exceptional. As with every ground in (c), the statute says only that the court "may" decline, so even a qualifying circumstance leaves the choice to its discretion.⁵⁸