Who Is a Required Party

Who Is a Required Party

Rule 19(a)(1) opens by describing whom it reaches: a person who is subject to service of process and whose joinder will not deprive the court of subject-matter jurisdiction must be joined as a party if one of the conditions that follow is met.¹ Those opening words act as a gate. A court cannot make a party of someone it cannot serve, and it cannot add someone whose presence would destroy the jurisdiction on which the case rests; joining a nondiverse party in a diversity case is the usual example. Both limits belong to the feasibility question taken up in a later section. Courts usually test the conditions first, because whether the absentee’s absence causes any harm decides whether feasibility matters at all.²

Courts often phrase this first inquiry as whether the absentee is “necessary,” although Rule 19 never uses that word.³ The older vocabulary sorted absentees into “necessary” parties, who must be joined if feasible, and “indispensable” parties, whose joinder is compelled even at the cost of dismissing the action.⁴ Modern procedure drops the labels but keeps the sequence. A court orders joinder “if feasible,” and the action must be abated only if, “in equity and good conscience,” it cannot proceed without the absentee.⁵ This unit covers the first step.

The conditions share one design: each describes a harm that follows if the absentee stays out. Courts commonly reduce them to three questions. Can the court give complete relief to those already parties? Will a judgment entered without the absentee, as a practical matter, impair the absentee’s ability to protect an interest in the subject of the action? Will an existing party face a substantial risk of double, multiple, or otherwise inconsistent obligations? An affirmative answer to any one makes the absentee a person who must be joined if feasible.⁶ The rule’s text sorts the questions into two clauses. Subdivision (a)(1)(A) addresses complete relief among the existing parties. Subdivision (a)(1)(B) addresses an absentee who claims an interest relating to the subject of the action, and it divides into impairment of that interest and double or inconsistent obligations.⁷ An earlier formulation of the rule, quoted by a court of appeals, set out the same conditions in nearly identical words.⁸

The tests ask about harm, rather than about what joinder would add, because a plaintiff can almost always name someone else who could have been sued. The choice of defendants generally belongs to the plaintiff.⁹ A rule that compelled joinder whenever it would make the lawsuit more comprehensive would swallow that choice, and efficiency concerns alone do not make an absentee required.¹⁰ Tying joinder to a concrete consequence for the relief, for the absentee, or for an existing party keeps the compulsory category narrow. Unlike the liberally construed permissive joinder of Rule 20, Rule 19 draws frequent judicial reluctance to mandate joinder.¹¹ The first test asks only what the court can accomplish between the parties already before it.

Complete Relief Among Existing Parties

The first condition asks a narrow question about the lawsuit as it stands. An absentee who can be served and joined must be joined if, “in that person’s absence, the court cannot accord complete relief among existing parties.”¹² The words among existing parties fix the focus on the people already in the case. The court asks whether the judgment it can enter between them will give the plaintiff what the claim entitles the plaintiff to receive. In Janney Montgomery Scott, Inc. v. Shepard Niles, Inc., the Third Circuit stated the limit plainly: the inquiry asks only whether the court can grant complete relief to the persons already parties, and the effect a decision may have on the absent party is not material.¹³

That limit follows from the harm the clause targets. A decree that cannot be carried out leaves the plaintiff with hollow relief, and this clause guards against that result. It comes into play when the claim seeks nonmonetary relief and any court order would be ineffective or incomplete unless certain nonparties were before the court.¹⁴ Harm to the person left out is a separate problem, which the two interest-based clauses in the next sections address. The distinction has practical force. An argument that the absentee will feel the judgment, however sincere, says nothing about whether the existing parties can obtain full relief from one another.

A decree against Adams would leave the sale undone as to one seller and in force as to the other. The court cannot give Menendez the relief the claim seeks unless Vincenzo is before it. The clause has likewise been found satisfied in disputes over rights to a piece of property when one of the owners is not joined.¹⁵ Relief can also fail because it depends on an act that only the absentee can perform. In one illustration, a plaintiff seeks an order requiring Tower Financial to increase electrical service. A court could issue that order, but Tower could not obey it without the building owner’s consent. The owner, General Leasing, should therefore be joined so that the court can grant complete relief between the original parties.¹⁶

A claim that the named defendants can satisfy in full presents the opposite picture. In Janney, the plaintiff sued Shepard Niles on an agreement that Underwood had also signed, and Shepard Niles argued that Underwood had to be joined. The court reasoned that if the agreement imposes joint and several liability on its co-obligors, the persons who share the duty to perform, complete relief may be granted in a suit against only one of them.¹⁷ Whether the liability was joint and several depended on how the agreement should be construed. The record did not show that it imposed joint and several liability, but it did not rule that reading out. Because the district court had to view the pleadings in the light most favorable to the plaintiff on Shepard Niles’s Rule 12(c) motion, its conclusion stood.¹⁸ The court of appeals agreed that complete relief could be afforded without Underwood.¹⁹

Set the rescission hypothetical beside Janney, and one variable explains the different results. A rescinded sale has two sellers, and a decree reaches only the seller before the court, so the relief is incomplete. A jointly and severally liable obligor owes the whole, so a judgment against the obligor sued can be collected in full, and the absent obligor would add only a second source of payment. If every possible obligor were a required party, Rule 19 would convert the permissive joinder of Rule 20 into compulsory joinder. A plaintiff who might collect more by suing both of two defendants remains free to make the tactical choice to sue one, and a defendant who fears bearing the whole judgment can implead the other.²⁰

Joint tortfeasors fall on the same side of the line. Ramirez, injured in a two-car collision, sues one driver, Bryan, and not the other, Grayson. Ramirez could have sued Grayson as well, but Rule 19(a)(1)(A) does not require it. If Bryan is found to have negligently caused the injury, the court can enter judgment against him for the full damages, because under joint and several liability any tortfeasor found liable to the plaintiff is liable for the plaintiff’s entire damages.²¹ An argument that joining Grayson would be more efficient fails for the reason it fails elsewhere: efficiency concerns alone do not make an absentee required, as courts note in citing Temple v. Synthes Corp.²²

The test measures relief against what the plaintiff is entitled to receive, so the chance that the plaintiff loses on the merits does not make the relief incomplete. Cardena sues Nashoba, claiming that Nashoba’s negligence caused a fire, and Nashoba points to Kelleher as someone who might be liable instead. If the jury finds Nashoba negligent, Nashoba pays the full damages. If it does not, the full relief to which Cardena is entitled from Nashoba is nothing. Either result can be entered without Kelleher, because Rule 19(a)(1)(A) does not make a party of everyone who might be liable to the plaintiff instead of the defendant.²³

The absentee’s position matters at the next stage as well. As one authority summarized, many courts leave the joinder of obligors to the plaintiff’s discretion, so the plaintiff may select defendants without concern about dismissal for nonjoinder. Joint obligors are thus treated as Rule 19(a) parties but are not deemed indispensable under Rule 19(b). Joint obligees, by contrast, usually have been held indispensable, and their nonjoinder has led to dismissal.²⁴ The plaintiff’s freedom to choose among obligors matches the result in Janney.

The Claimed-Interest Threshold

Rule 19(a)(1)(B) reaches an absentee who “claims an interest relating to the subject of the action” and who is so situated that deciding the case without that person may cause one of two harms.²⁶ Clause (i) addresses the risk that the absentee’s practical ability to protect the interest will be impaired. Clause (ii) addresses the risk that an existing party will face double, multiple, or otherwise inconsistent obligations because of the interest.²⁷ Both clauses depend on the same opening words, so the interest works as a threshold. An absentee with no interest relating to the subject of the action has nothing for a judgment to impair and no claim that could pull an existing party in two directions. The complete-relief clause runs on a different track, because it looks only at the relief available among the persons already in the case and asks nothing about what the absentee holds.²⁸

The threshold exists because nearly every lawsuit touches people who are not parties. A witness whose testimony will matter, a lender whose borrower will be poorer if the defendant loses, and a person with a separate claim resting on similar facts all have some connection to the dispute. If each of them were a required party, the plaintiff’s choice of defendants would give way whenever anyone else was affected. Requiring a stake in the subject of the suit itself keeps the compulsory category small enough to justify that override.

Courts tend to discount interests that are merely financial or speculative. In Cachil Dehe Band of Wintun Indians v. California, the Ninth Circuit put it this way: the interest must be more than a financial stake, and more than speculation about a future event.²⁹ Both limits follow from the structure of the rule. A financial stake exists whenever a judgment makes someone richer or poorer, and a speculative interest depends on events that may never occur, so neither shows a connection to the subject of the action itself. A fixed fund is the exception. The threatened depletion of a limited fund to which the absentee lays claim can be an interest at risk, because “[a]n interest in a fixed fund or limited resource that the court is asked to allocate may also be protected.”³⁰

The bank’s stake rises and falls with the roofer’s finances but concerns nothing the homeowner’s claim asks the court to decide, so the bank has only a financial stake. Each absent injured person, by contrast, has a claim on the fixed fund that the court is being asked to allocate.

Ownership of disputed property is the clearest source of an interest. In Haas v. Jefferson National Bank, Haas claimed half of a block of stock in a suit against the bank, and all of the stock had been issued in Glueck’s name. Glueck, who was not a party, therefore had an interest in the very property the court would be dividing, and the court called that interest “at least apparent.”³¹ A judgment awarding Haas half could not bind Glueck, who might later assert ownership of the whole against the bank, and the same judgment could affect his ability to protect his claim to all of the shares.³² Both clauses of subdivision (a)(1)(B) are therefore in play, and the interest in the stock opens the door to each.

An absentee whose only link to the case is a role in the story stays outside the threshold. Xavier sues Apex Corporation for the negligence of its driver, Belinsky, and Apex asks that Belinsky be joined because his negligence will be litigated. Belinsky’s conduct will be evidence, but he claims no interest in the subject of Xavier’s claim, which is the liability of Apex. As one court put it, a party’s plan to offer evidence that a nonparty behaved improperly does not, by itself, make the nonparty necessary, because a contrary view would greatly expand the universe of required parties.³³ Glueck and Belinsky both have a part in the dispute. Glueck’s name is on the shares in contention, while Belinsky holds nothing that a judgment against Apex will award or take away.

Contracts divide along the same line, between an agreement the suit puts at issue and one the suit merely touches. In MasterCard International v. Visa International Service Ass’n, the underlying litigation involved the MasterCard contract and whether MasterCard had a right of first refusal to World Cup sponsorship rights. Visa’s separate contract with FIFA might be affected, but it was not the contract at issue. Even if MasterCard prevailed, the Visa contract would not become invalid; at most FIFA would likely have breached a warranty in that contract, and Visa could sue FIFA for the breach. The court distinguished Crouse-Hinds Co. v. InterNorth, Inc., where the absent nonparty was a party to the contract at issue.³⁴ The court framed that contrast as one of impairment, which the next section develops. The starting observation, though, concerns the interest: a contract the suit puts at issue is part of the subject of the action, and a contract the suit merely touches is not.

Impairing the Absentee’s Interest

Once an absentee has an interest in the subject of the action, Rule 19(a)(1)(B)(i) asks what the judgment will do to that interest. The clause reaches an absentee who is so situated that disposing of the action without that person may “as a practical matter impair or impede the person’s ability to protect the interest.”³⁶ The operative words are as a practical matter, because a person who was never a party is not legally bound by the judgment. Issue preclusion reaches only persons who were parties to the prior action or shared the interest of those who were.³⁷ A clause that demanded legal binding would apply to almost no one, so the rule directs attention to a different question: whether the absentee, though free to ignore the judgment, will be left unable to protect what the absentee holds.

One court explained the point this way. That a judgment is not res judicata as to, or enforceable against, a nonparty means only that. It does not mean a court may never enter a judgment that in practice affects a nonparty, and it does not allow a court to proceed without weighing that effect merely because the nonparty is not bound in the technical sense. Rule 19(a) instead requires the court to consider the extent to which the judgment may “as a practical matter impair or impede” the absentee’s ability to protect its interest.³⁸ Three kinds of practical effect recur in the decisions.

The first is a decree aimed at one party whose consequences land on another. In Torrington Co. v. Yost, a manufacturer sued a former employee, Yost, for an injunction limiting his employment at his new employer, INA, for eighteen months.³⁹ The order would run only against Yost, so INA would not be bound in the strict sense. Yet if the court enjoined Yost from working for INA, INA would lose his services without having had a chance to oppose the injunction. INA might also sue Yost for breach of contract, and a court in that action might find that he breached by obeying the injunction.⁴⁰ Helzberg’s Diamond Shops v. Valley West Des Moines Shopping Center follows the same logic. An injunction that might prevent a jewelry retailer from operating its store in a shopping mall in the manner the retailer originally intended made the retailer a party to be joined if feasible.⁴¹

The second kind of effect is the depletion of a fund the absentee shares. When a court is asked to allocate a limited resource among claimants, an increase for one claimant comes out of the shares of the others, and that threatened depletion can impair the absentees’ ability to protect their claims.⁴² The tribal fishing dispute in Makah Indian Tribe v. Verity is the familiar illustration: a tribe sought a larger share of a limited harvest, and the other tribes that shared it stood to lose what the plaintiff gained.

The third kind is a finding that a later court would treat as settled against the absentee. In Janney, the Third Circuit accepted that issue preclusion can satisfy the clause, but it required a real showing. Mere presentation of an argument that preclusion is possible is not enough. Instead, it must be shown that some outcome of the federal case that is reasonably likely can preclude the absent party on an issue material to that party’s rights or duties, under standard principles governing the effect of prior judgments.⁴³ Underwood, the absent co-obligor, could not make that showing. For preclusion to apply against it, a state court would have to find Underwood in privity with Shepard Niles, because Underwood controlled or directed the earlier litigation or Shepard Niles was sued in a representative capacity.⁴⁴

The weaker the connection between the judgment and the absentee, the less the clause helps. The effect of a district court decision on Underwood’s rights was speculative, because the court could not say with assurance that a decision in Janney’s action against Shepard Niles would be likely to impair Underwood’s ability to defend itself in the state action or to obtain indemnity or contribution. The decision’s “precedential” effect therefore did not justify treating Underwood as necessary. The court set that case against one in which determining title to real estate in which the absent party had an interest appeared inevitable if the dispute was to be resolved.⁴⁵ Courts are not uniform on the point. Some have found that the persuasive effect of a prior judgment can supply the practical impairment, as where a federal judgment might set precedent in a later state-court action involving the absent party and similar arguments, which puts Janney at the demanding end of the range.⁴⁶

A contract the suit leaves intact does not carry the same risk. In MasterCard, the Visa contract was not the contract at issue, and a MasterCard victory would not make it invalid. FIFA would likely have breached a warranty in that contract, and Visa would have the right to sue FIFA for the breach. In Crouse-Hinds, by contrast, the absent nonparty was a party to the contract at issue, so its ability to protect its interest in that contract would have been seriously impaired.⁴⁷ Visa would not be bound by any injunction in MasterCard’s suit, but FIFA would be, and a later proceeding would have to recognize and respect that injunction.⁴⁸ What separates the two absentees is whether a remedy remained open to them after the judgment.

Impairment also fails when the absentee is adequately represented. Courts have indicated that there is no risk of impairment if the nonparty is adequately represented in the action.⁴⁹ In Janney, however, representation mattered only if impairment were first shown, and because it was not, the adequacy of Shepard Niles’s representation of Underwood was immaterial.⁵⁰ Moran v. Household International, Inc. reached the same conclusion under Chancery Rule 19 for holders of rights issued under a rights plan. Because Household, a party, fully protected the rights holders’ interests, they need not be joined even if feasible, and requiring joinder of every rights holder would likely deny judicial review of the plan altogether.⁵¹ Joining an absentee whose position a party already protects adds expense without adding protection, and a requirement to join a numerous group can keep the case from being decided at all.

Finding impairment establishes only that the absentee is a person to be joined if feasible. The harm to the absentee is weighed again under Rule 19(b)(1), which asks the extent to which a judgment rendered in the person’s absence might prejudice that person or the existing parties.⁵² In Helzberg, the retailer was a party to be joined if feasible, but it was not subject to personal jurisdiction in the forum. The court went on to the Rule 19(b) factors and held the retailer not indispensable, because it retained all of its rights under its lease and none of its rights or obligations would be adjudicated in a proceeding to which it was not a party.⁵³ In Torrington, INA could not be joined, and the court concluded that dismissal was the better option.⁵⁴ In Republic of the Philippines v. Pimentel, the parties conceded that the Republic and its Commission were required because, without them as parties in the interpleader action, their interests in the subject matter were not protected under Rule 19(a)(1)(B)(i). The dispute concerned whether the action could go forward without them under Rule 19(b).⁵⁵ Requiring the absentee and dismissing the case are separate questions, and the rule answers them at separate stages.

Double or Inconsistent Obligations

The second branch of Rule 19(a)(1)(B) protects a different person from the one the impairment clause protects. Clause (i) asks what a judgment will do to the absentee, while clause (ii) asks what the absentee’s absence will do to a party already in the case. An absentee who claims an interest relating to the subject of the action must be joined if feasible when deciding the case without that person would “leave an existing party subject to a substantial risk of incurring double, multiple, or otherwise inconsistent obligations because of the interest.”⁵⁷ The danger follows from a plain fact about judgments: a person who was not a party is not bound by the result. A defendant who loses to one claimant may therefore still answer to another claimant who is free to ask a different court for something the first judgment cannot accommodate.⁵⁸

Haas shows the danger in its plainest form. Haas claimed half of the stock in a suit against the bank, and all of the stock had been issued in Glueck’s name. The court reasoned that if Haas prevailed in Glueck’s absence and were adjudicated owner of half the stock, Glueck, not being bound by res judicata, could theoretically succeed in later litigation against the bank in asserting ownership of the whole.⁵⁹ The bank would then stand between two judgments that describe the same shares differently, and it could not honor both. The exposure came from Glueck’s absence. Had he been a party, one judgment would have settled the claims of both men, and the bank would have dealt with the shares once.⁶⁰

An order to pay Carena would leave the bank facing a likely second suit by Jaquith, in which a different court might order the same funds paid to Jaquith. The bank would then have to disobey one order or pay twice. Jaquith’s absence creates that exposure, because a judgment in Carena’s suit would not bind Jaquith. In Illinois Brick Co. v. Illinois, the Supreme Court recognized the same pattern on a larger scale. Purchasers at different levels of a distribution chain could each claim that the whole overcharge fell on them, so a treble-damages suit by one implicated the defendant’s interest in avoiding multiple liability for that single fund. That interest is one of those that traditionally support compulsory joinder of absent, potentially adverse claimants.⁶¹

Clause (ii) does not reach every case in which a defendant could see different results in different courtrooms. Inconsistent obligations arise when a party cannot comply with one court’s order without breaching another court’s order concerning the same incident. Inconsistent adjudications arise when a party wins on one claim in one forum and loses on another claim from the same incident in another forum.⁶² The rule’s text speaks of obligations, and the line tracks a real difference in harm. A party who wins in one court and loses in another can obey both judgments, paying one plaintiff and owing the other nothing. A party ordered to hand the same property to two people has received two commands that cannot both be carried out.

The ordinary tort case sits on the safe side of that line. Ramirez sues Bryan for injuries from a three-car collision, and another person hurt in the same collision is not a party and may sue Bryan separately. Two juries could disagree about whether Bryan was negligent, but nothing in the second judgment orders Bryan to do what the first forbids. If he is held liable to two injured plaintiffs in two different suits, he can pay them both. He may not like it, but he can do it.⁶³ The escrow bank faced two demands on one fund, while Bryan faces separate awards for separate injuries, so only the bank is in the position the clause describes.

A second limit protects the plaintiff’s freedom to sue one of several obligors. In Janney, Shepard Niles argued that Underwood’s absence exposed it to double or inconsistent obligations. Janney’s answer, which the court accepted, was that regardless of who won in federal court, the outcome would be res judicata or collateral estoppel between Janney and Shepard Niles, resolving that dispute once and for all.⁶⁴ A plaintiff who recovers in full from one obligor, moreover, cannot double that recovery by maintaining a state court action against the other.⁶⁵ Paying the whole judgment is a different event from paying twice. A defendant who claims a right of contribution or indemnity from a third person can protect itself by impleading that person under Rule 14, but it is not required to do so, and if it does not, its right to bring a separate action for contribution or indemnity is unaffected. Another court of appeals has therefore held that a defendant’s right to contribution or indemnity from an absent party does not make the absentee indispensable.⁶⁶

Haas and Janney look alike at first, because in each an absent person had a legal tie to the defendant and a later suit was possible. The direction of the later claim separates them. Glueck would sue the bank for the same shares the first judgment had treated as half Haas’s, so the bank would answer to two claimants for one asset. Underwood stood on the other side of the later claim, because Shepard Niles would be the one seeking money by suing for contribution or indemnity, and the debt would be paid only once. A defendant who may later look to the absentee for reimbursement has a remedy in hand. A defendant who may later answer to the absentee for what a court has already awarded to someone else has a problem that only joinder removes.

Exposure that arises from a separate contract is a weaker candidate. A person does not become indispensable to an action to determine rights under a contract simply because that person’s rights or obligations under an entirely separate contract will be affected by the result, as the court held in Helzberg.⁶⁷ MasterCard adds that a later tribunal is unlikely to order a party to perform the very obligations a prior court has prohibited. FIFA was bound by any injunction, a subsequent proceeding would have to recognize and respect it, and FIFA, the party supposedly facing the predicament, had not argued that it would be prejudiced by Visa’s absence.⁶⁸ A useful practical question follows from these cases: would joining the absentee actually remove the exposure? Joining Glueck would have bound him to a single ruling on ownership. Where a later court would have to honor the first order in any event, joinder adds little.

Satisfying the clause establishes only that the absentee is to be joined if feasible. In Haas, the court went on to Rule 19(b), where it treated the first factor, prejudice to the absentee or the existing parties, as tracking clause (ii) and as supplying weighty reason for a finding of indispensability.⁶⁹