Concept
Settlement Among Tortfeasors
Apportionment of liability and contribution among tortfeasors
Settlement Among Tortfeasors
Under Restatement (Second) of Torts § 879, each of two or more persons whose tortious conduct is a legal cause of harm that cannot be apportioned is liable for the entire harm, whether their conduct was concurring or consecutive.1 The injured plaintiff may collect only once, however, because no plaintiff is entitled to more than one full satisfaction of the damages.2 A settlement with one tortfeasor sets those two principles against each other. The plaintiff keeps a claim against the others, but its size now depends on what the settlement has paid, and its survival depends on what the plaintiff gave up to get the money.
Plaintiffs often take less from one defendant than the claim may be worth, for reasons that have little to do with how much that defendant owes. Parties value claims subjectively: the settling defendant's negligence may prove hard to establish, or a sympathetic defendant may sway the jury.3 A settlement price also reflects the uncertainty of trial and gives the plaintiff funds to litigate against the remaining defendants.4 The defendants face pressure of their own. In what is described as a “whipsaw” tactic, plaintiffs usually escalate their demands with each successive defendant who agrees to settle, so the last defendant to settle generally pays much more than the first. That can create disharmony among defendants and may incite one to bid against another to be first.5
Three questions follow, and the subsections below take them in order. Does the settlement discharge the other tortfeasors, or does the plaintiff keep the claim against them? How far does the money already paid reduce that claim? And may anyone seek contribution from the settling tortfeasor, or may the settlor seek it from the others? Unless a passage says otherwise, assume that joint and several liability and the Uniform Contribution Among Tortfeasors Act apply.6
Releases and Covenants
Section 4 of the Uniform Contribution Among Tortfeasors Act sets the modern rule. A release or covenant not to sue given in good faith to one of two or more persons liable in tort for the same injury does not discharge the others unless its terms so provide, reduces the claim against them by the greater of any amount the instrument stipulates or the consideration paid for it, and discharges the settling tortfeasor from all liability for contribution to any other tortfeasor.7 This subsection takes up the first clause, which asks whether the claim against the others survives. The next two subsections measure the credit and apply the contribution bar.
At common law, a release given to one tortfeasor released every other person liable for the same harm, concurrent tortfeasors as well as those acting in concert.8 The rule rested on the principle that a victim may be compensated only once, so that releasing one wrongdoer and suing another would enrich the victim unjustly.9 Terminology helped the rule spread. Early common law counted as joint tortfeasors only those who acted in concert, and the act of one was the act of all, so a release of one surrendered the single claim they shared. The same label later attached to independent tortfeasors who could be joined in one suit because they were liable for the same damages.10 The rule also confused satisfaction, which is full compensation for the injury, with release, which surrenders the claim whether or not the plaintiff has been paid in full.11 The one-satisfaction principle is sound, but it justifies only barring a second recovery of money already received. It gives no reason to cancel the unpaid balance of a claim because part of it was paid.
Pringle shows what the strict rule costs. Suppose Whiting and Shattuck both caused her $50,000 loss, and she releases Whiting for $40,000. Under the strict common-law rule, a party who released any one tortfeasor surrendered the right to sue every possible defendant on that claim.12 Her suit against Shattuck is therefore dismissed.13 She has received four-fifths of her loss and is barred from the rest, while Shattuck, who is liable for the same harm, pays nothing. The rule also blocked partial settlements, because a plaintiff could not take part-payment from a willing tortfeasor without giving up the claim against everyone else.14
Lawyers answered with the covenant not to sue, in which the plaintiff keeps the claim and promises not to enforce it against the settling defendant.15 The settling defendant still gets real protection, because a plaintiff who sues anyway faces a counterclaim for breach of the promise, with damages equal to what the plaintiff recovers plus the cost of defending the suit.16 A variation, the covenant not to enforce a judgment, protects the defendant against collection in the same way.17 Courts saw that the device answered formalism with formalism, since the plaintiff gave up the claim against the settling defendant for money just as a release would have required. Many nevertheless sanctioned it because both sides gained: the plaintiff could settle with one tortfeasor without waiving the claims against the others, and the defendant could resolve a legitimate claim without paying more than a fair share.18 Had Pringle taken a covenant in a jurisdiction that honors the device, her claim against Shattuck would have survived. A $50,000 judgment, less the $40,000 Whiting already paid, leaves Shattuck owing $10,000.19
Other jurisdictions attacked the underlying rule instead of relying on the covenant. The changes, made by decision and by statute, differ in how much weight they give the plaintiff's drafting:
- Some states distinguish a release from a covenant, so that a release discharges the other tortfeasors while a covenant does not unless the plaintiff has been fully compensated.20
- Some states treat a release that expressly reserves rights against the other tortfeasors as a covenant not to sue, whatever words of release it uses.21
- Some states hold that even a release without a reservation leaves the others liable unless it shows an intent to discharge them or the plaintiff has been fully compensated, as in Leung v. Verdugo Hills Hospital, a California decision.22
- Statutes elsewhere provide that a release with an express reservation does not discharge the others, as the Uniform Joint Obligations Act does, or that the others are discharged only when the release's terms so provide, as the Uniform Contribution Among Tortfeasors Act does.23
Restatement (Second) of Torts § 885 went furthest. It provides that a release does not discharge others liable for the same harm unless the agreement expressly says so, and it describes a requirement of express reservation as a “booby-trap for the unwary.” Restatement (Third) of Torts: Apportionment of Liability § 24 now supersedes § 885 and reaffirms the rule, and most jurisdictions follow it.24 The modern rule works because it gives two jobs to two devices. The credit preserves the one-satisfaction principle, and the instrument's terms decide whether the claim continues.25 A plaintiff can settle with one defendant for part of the loss and pursue the rest from the others without being paid twice.
Because the modern tests turn on intent and payment, the title of an instrument counts for less than its substance. Whether a document is a release or a covenant is a question of construction, and the parties' intent as shown by its substance controls; the form of designation carries no independent weight.26 A payment of the full damages satisfies the claim and operates as a release whatever the instrument is called.27 Cox v. Pearl Investment Co. shows how a title can mislead. Mrs. Cox fell on property owned by Pearl Investment. The tenant, Goodwill Industries, paid the Coxes $2,500, and they signed a document titled “Covenant Not to Proceed with Suit” that reserved “the right to sue any other person or persons” with a claim arising from the fall. The trial court nevertheless called the document a release and, applying the common-law rule, entered summary judgment for the owner.28
The amount paid and the words of the instrument decide what survives, and the title decides nothing by itself.29 Broad words can still cost the plaintiff the claim, because the Act's exception covers an instrument whose terms so provide. In Douglas v. United States Tobacco Co., the Eighth Circuit read “all other persons” in a general release as satisfying the Act's condition that a release discharges others “if the release so provides.” Any later tortfeasor was therefore discharged regardless of the compensation actually received, and a majority of jurisdictions follow that reading.30
Releases given to an original tortfeasor raise the same question about a physician who later treats the injury negligently, and the results turn on the governing rule. In Tyler v. Ahmed, a plaintiff who had settled all of her injury claims could not afterward bring a malpractice action against the physician who treated those injuries, because she had knowingly compromised every claim whether or not she thought herself fully compensated. In Morgan v. Cohen, a Maryland case under the Uniform Act, a general release did not discharge the physician as a matter of law, and the result depended on the parties' intent and on whether the satisfaction covered all of the damages.31 Courts also divide on whether a plain release of the driver who caused a crash bars a claim against the physician who later treated the injury negligently. Gallagher v. Mercy Medical Center applied the one-satisfaction rule to bar the claim, while Lujan v. Healthsouth Rehabilitation Corp. declined to apply the rule to successive tortfeasors.32
One class of defendants sits outside the modern rule in many jurisdictions. A release of a tortfeasor also releases a party who is liable solely because the law makes it vicariously liable for that tortfeasor's conduct.33 The usual explanation is that the principal's exposure has no source other than the agent's wrong, and a principal who pays holds a right of indemnity against the agent, so a plaintiff who released the agent and then collected from the principal would push the loss back onto the very person the release was meant to protect.34 The authority on employment still describes the older contrast, under which a release of one joint tortfeasor releases all while a covenant not to sue does not, as the customary rule in most jurisdictions when employer and employee are the defendants.35
The courts do not agree on this exception, and the disagreement is wide. Booth v. Gades held that a release of a firefighter also discharged the city that employed him, Christiana Care Health Services v. Carter reached the same result when a patient released a surgeon and then sued the hospital, and J&J Timber Co. v. Broome discharged an employer along with its driver even though the release expressly reserved rights against the employer. JFK Medical Center v. Price, by contrast, held that a release of a doctor does not release the hospital that employs the doctor, and McCrosky v. Carson Tahoe Regional Medical Center refused to let a settlement with a doctor extinguish a hospital's liability based on apparent agency. Woodrum v. Johnson, after collecting the cases and noting the diverse disagreement, found no release.36 An answer involving a principal and an agent should therefore state the governing jurisdiction's position and, if the facts do not supply one, note that the courts split.
Dollar Credit Rule
Under the dollar credit rule, a good-faith settlement reduces the claim against the remaining tortfeasors by the amount the instrument stipulates or the consideration paid, whichever is greater, and discharges the settlor from contribution to the others.38 The rule is also called the pro tanto rule, from the Latin for “to that extent”: the remaining defendants receive credit for exactly the dollars the plaintiff has already received. The preceding subsection asked whether the claim survives. This one asks how much of it is left and who may seek reimbursement from whom.
The credit follows from two premises already in place. Each tortfeasor owes the entire indivisible harm, so a co-tortfeasor's settlement does not shrink what the plaintiff may demand from the others.39 But the plaintiff is entitled to one full satisfaction only, so whatever the settlement has paid must come off the remaining claim.40 Erg's claim shows both at work. Erg settles with Kelvin for $7,000, and a jury later fixes Erg's damages at $16,000. Volt, the remaining defendant, owes the whole judgment, receives credit for the $7,000, and pays $9,000.41
The Rogers fire shows how the credit rewards or punishes the settlor as the settlement price moves above or below what the jury later finds. Dunne and Thurber negligently start a fire that causes Rogers $42,000 in loss. Dunne settles; Thurber goes to trial and loses, and the jury sets damages at $42,000. Thurber remains liable for the whole judgment but receives a credit for what Dunne paid. If Dunne paid $12,000, Thurber owes $30,000, well above the $21,000 that half the loss would be. If Dunne paid $35,000, Thurber owes only $7,000. A good deal for Dunne shifts more than half of the damages to Thurber, and a bad deal leaves Dunne paying more than half himself.42
Contribution cannot repair either result. Section 4(b) of the Uniform Act discharges Dunne from liability for contribution to any other tortfeasor, so Thurber cannot make Dunne pay more than the $12,000.43 Section 1(d) closes the other side. A tortfeasor who settles cannot recover contribution from another tortfeasor whose liability the settlement did not extinguish.44 Dunne bought a release of his own liability only, so a Dunne who paid $35,000 cannot undo his bargain by claiming the excess from Thurber.45 The reason lies in what contribution is for. It reimburses a person who paid a debt that others also owed, and a settlor who buys only its own release has paid only its own debt. The other defendants still owe the plaintiff the remainder, already reduced by the credit, so the credit has passed the benefit of the settlement to them.
The settlor also receives something for the risk it accepted. Suppose Nash pays $16,000 to settle with Twain, and the jury later sets Twain's damages at $20,000. Benchley owes only $4,000 after the credit, and Nash cannot recover his overpayment. Nash has still saved the expense of trying the case and has avoided the risk, always difficult to assess, that the jury would award much higher damages, say $100,000. Without the settlement he would have paid at least $50,000, and perhaps $100,000 if Benchley were insolvent.46
The settlor's protection depends on good faith. In most dollar credit jurisdictions, the settling tortfeasor is immune from contribution only if the court finds the settlement an arm's-length, bona fide bargain that was not collusive.47 Courts divide on whether the inquiry looks only for collusion or also asks whether the amount was adequate.48 Thurber might argue that Dunne's $12,000 settlement was so low that it shows a plan with Rogers to load the loss onto Thurber. The argument is unlikely to succeed, because many subjective factors shape what a claim is worth to a settling defendant, and a payment below the pro rata share is not bad faith by itself.49 A collusive settlement, in which the plaintiff and one defendant agree on a token payment to push the loss onto the others, falls outside the protection.50 The 1955 Uniform Act's bar on contribution from a settlor is also endorsed by Restatement (Third) § 23, comment i. The 1939 Uniform Act took the opposite position and let nonsettling defendants seek contribution from a settlor, a rule that can discourage settlement because the settlor can never be sure its deal is final.51
The settlor's own bar has an edge. In the absence of a statute requiring otherwise, a tortfeasor who settles the plaintiff's entire claim, and so discharges other tortfeasors' liability along with its own, may obtain contribution from those others if it proves their liability to the plaintiff and the reasonableness of the settlement.52 Dunne obtained a release of his own liability only, which is why § 1(d) leaves him without a claim against Thurber.53
A three-defendant verdict shows where the gap between a settlor's fault and its payment lands. The plaintiff settles with T1 for $50,000 before trial, tries the case against T2 and T3, and wins $300,000, and the jury assigns one-third of the fault to each defendant. The court enters a joint and several judgment against T2 and T3 for $250,000. They cannot seek contribution from T1, and between themselves each owes $125,000, so whichever pays more may recover the excess from the other. T1's one-third share of the loss was $100,000, yet T1 paid $50,000. The missing $50,000 falls on the two remaining defendants, each of whom ends up $25,000 above its own one-third.54
The plaintiff's side of the arithmetic is more favorable. Twain, whose damages the jury sets at $20,000, can settle with Nash for $5,000 and still collect $15,000 from Benchley, $20,000 in all, because the remaining defendant owes the full judgment less only what was paid.55 A settlement larger than the jury's eventual verdict reverses the benefit. The credit then covers the whole judgment, the remaining defendant pays nothing, and the settlor recovers nothing, because it bought its peace at a price it chose.56
The jury ordinarily hears nothing about the settlement. Its task is to fix the plaintiff's damages and decide whether the remaining defendant is liable, and a settlement figure invites jurors to read it as the value the plaintiff places on the claim, which may reflect no such thing.57 The judge applies the credit afterward, so a $24,000 verdict in a case where $7,000 has already been paid becomes a $17,000 judgment against the remaining defendant.58 A tortfeasor's liability insurer pays on the tortfeasor's behalf, so its payments earn the credit. Payments from sources collateral to the defendant, such as the plaintiff's own health insurance, do not.59 The order of subtraction matters when the plaintiff was also at fault. Suppose damages are $100,000, the plaintiff is 20 percent at fault, and the settlement is $30,000. Subtracting the settlement first and then applying the plaintiff's percentage leaves $56,000, while applying the percentage first and then subtracting leaves $50,000. Whalen v. Kawasaki Motors Corp. is a New York decision that addressed that order.60
The credit hands each remaining defendant a precise figure to pay, and that precision is its chief advantage.61 Its cost is an incentive problem. A good-faith settlor is shielded from contribution, and the credit counts dollars instead of fault, so each defendant gains by settling early and cheaply and leaving the others with the balance. Restatement (Second) of Torts § 886A comment m names the disadvantage: the rule encourages each defendant to settle early so that the other is left holding the bag, which produces inversions between relative culpability and what each defendant ends up paying.62 The dissent in American Motorcycle Association v. Superior Court worked through the inversion with a plaintiff 30 percent at fault, one defendant 10 percent, and another 60 percent. If the 60 percent defendant settles for 20 percent of the loss, the plaintiff's judgment against the 10 percent defendant is 70 percent of the loss less a 20 percent credit, so the least culpable defendant owes 50 percent.63 The most culpable defendant escaped a 60 percent share for a third of that sum, and dollar liability runs opposite to fault. The next subsection takes up the alternative that measures the credit by the settlor's share of fault.
The credit does its work only where the remaining defendants are jointly liable for the amount credited. Its job is to keep the plaintiff from collecting the same debt twice, and a defendant who owes only its own several share has a debt the settlement never touched. Georgia held in McReynolds v. Krebs that no setoff applies once a statute has eliminated joint and several liability, and Texas reasoned in Duncan v. Cessna Aircraft Co. that a settlement does not measure the harm the remaining defendants caused and so should not affect their liability.64 Where joint liability survives only for economic damages, the setoff reaches only those damages, and Florida's decision in Wells v. Tallahassee Memorial Regional Medical Center divided the settlement between economic and noneconomic loss in the same proportion as the jury's award. Florida also allowed no setoff in Gouty v. Schnepel when the jury found the settling defendant not liable, because the setoff statute presumes joint liability.65
Proportional Credit Rule
Restatement (Second) of Torts § 886A comment m lists three ways to treat a payment from one tortfeasor who takes a release or covenant that is not a full satisfaction of the claim, and the third is the proportionate share rule.67 Under that rule the plaintiff is treated as holding two divisible claims, each with its own set value. A settlement with one defendant extinguishes that defendant's portion and leaves the other responsible only for the rest. If each defendant bears half of a $100,000 loss, a $30,000 settlement makes half the claim disappear, and the second defendant owes $50,000 however much or little the first paid.68 Once a jury has assigned percentages of fault, the same idea becomes a credit measured by the settlor's share of the judgment. Restatement (Third) of Torts: Apportionment of Liability § 16 views the settling defendant as selling its proportion of the ultimate judgment to the plaintiff for the agreed price, so the remaining defendants are credited with that proportion and the price drops out of their calculation.69
The Nell illustration puts numbers on the difference. Nell sues Fagan and Twist and settles with Fagan for $25,000, then tries her case against Twist in a jurisdiction that keeps joint and several liability and requires the jury to assign percentages to every tortfeasor, the settlor included. The jury finds her damages to be $100,000 and fixes fault at 20 percent for Nell, 70 percent for Fagan, and 10 percent for Twist, so Twist is liable for $80,000, her damages reduced by her own share.70 A dollar credit would subtract the $25,000 Fagan paid and leave Twist owing $55,000, more than five times his own share of the loss. The proportional credit instead gives Twist 70/80ths of the $80,000 judgment, which is $70,000, and he owes $10,000.71 The credit is the settlor's percentage put over the combined percentages of all the defendants, applied to the judgment rather than to the raw damages.72 Because the plaintiff's own percentage cancels out of that fraction, the credit also equals the settlor's percentage of the total damages, here 70 percent of $100,000.
Once the credit no longer depends on what Fagan paid, the settlement price moves Nell's total recovery. Nell holds $25,000 from Fagan and a $10,000 claim on Twist, a total of $35,000 against an $80,000 judgment, and the $45,000 gap is the difference between Fagan's $70,000 share and the $25,000 he paid. Under the dollar credit the gap landed on Twist, since Nell's total stayed at $80,000 and he paid the difference for Fagan's cheap exit. The proportional rule places the risk of the bargain on the plaintiff and the settlor, the two parties who set the price, and takes it off the defendant who had no say in it.
The same verdict with the roles reversed shows the symmetry. Had Nell instead settled with Twist for $3,000 and tried the case against Fagan, Fagan would receive a credit of 10/80ths of the $80,000 judgment, or $10,000, and would owe $70,000, exactly his own share, where a dollar credit would have charged him $77,000.73 Nell's total would then be $73,000, and she would bear the $7,000 by which Twist's $10,000 share exceeded his $3,000 payment. A cheap settlement by the least culpable defendant therefore cannot enlarge what the most culpable defendant owes.
The bargain can also favor the plaintiff, and the plaintiff keeps that advantage. In the Flite illustration the jury sets damages at $60,000 and assigns 20 percent of the fault to Flite, 50 percent to Micawber, 20 percent to Heap, and 10 percent to Murdstone, so the judgment is $48,000. Suppose Heap settles for $22,000 before trial. The credit against the defendants who remain is 20/80ths, or one quarter, of the $48,000, which is $12,000, leaving a $36,000 judgment. That sum is Micawber's $30,000 share plus Murdstone's $6,000 share, so the credit removes Heap's share and nothing else.74 Flite therefore collects $58,000 in all, which is $10,000 more than the jury decided she should receive, because Heap paid well in excess of his share. The Third Restatement approves that result on the ground that the plaintiff bears the risk of an inadequate settlement and should also obtain the benefit of a favorable one.75 Had Heap paid only $5,000, Flite would have collected $41,000, which is $7,000 less than the judgment.
The Supreme Court adopted this approach for maritime cases in McDermott, Inc. v. AmClyde.76 A crane hook failed during a heavy lift, and the owner, McDermott, sued AmClyde, the crane's maker, River Don Castings, the maker of the hook, and other defendants, who settled before trial for $1 million.77 The jury put McDermott's damages at $2.1 million and assigned 38 percent of the fault to River Don.78 The Court rejected the dollar-for-dollar approach and chose the proportionate share approach, so River Don owed its adjudged share of $798,000 whatever the settling defendants had paid.79 The deciding consideration was consistency with the fault-based apportionment the Court had adopted in United States v. Reliable Transfer Co., because the Court regarded the effects on settlement and on judicial economy as ambiguous.80 Outside admiralty the credit depends on which rule a jurisdiction has adopted, so an exam answer should identify the rule from the facts. The Third Restatement endorses the proportional approach, and the AmClyde rule continues to gain ground.81
Because the settlor's share has already been removed from the judgment, contribution has no work to do in either direction. The remaining defendants cannot seek it from the settlor, since they never pay any part of the settlor's share, and the settlor cannot seek it from them, since its payment bought only its own release.82 Where a dollar credit applies, the settlor's protection from contribution depends on a finding of good faith, as the preceding subsection explained, whereas under AmClyde those claims are barred without any fairness finding.83 The credit leaves the rest of joint and several liability in place, however, so the remaining defendants still carry the risk that one of them cannot pay.84 If Micawber were insolvent, Murdstone could be made to pay the whole $36,000 and would be left with a worthless claim against Micawber, though no one could charge him with Heap's share.
The case for crediting by fault begins with what the dollar credit does to the defendant who goes to trial. A settlement price reflects the uncertainty of trial and gives the plaintiff funds to finance the case against the remaining defendants, so it usually runs below the settlor's equitable share, and a dollar credit passes the whole discount to the litigating defendant.85 Each defendant then gains by settling early and cheaply and leaving the other holding the bag, which produces inversions between culpability and payment. The proportionate share rule removes that gain, because no defendant can lower what another owes by settling for less than its share, and the plaintiff has little reason to discount the claim, knowing the larger fraction cannot be recovered from the second defendant.86 The Supreme Court made the same point in AmClyde, observing that a plaintiff's good fortune in striking a favorable bargain with one defendant gives the other defendants no claim to pay less than their proportionate share. The rule also does not make a litigating defendant's liability depend on a settlement negotiated by others without regard to its interests.87 A New Jersey decision, Kiss v. Jacob, states the same allocation of risk: when the settlement exceeds the share the jury attributes to the settlor, the plaintiff benefits by the excess, just as the plaintiff runs the risk of a low settlement.88
The cost of the proportional credit is administrative. Someone must determine the settlor's share, and the rule does not say clearly when that determination should occur.89 Because the settlor has usually left the case, the remaining defendant must present evidence of the settlor's liability at trial just as it would in a contribution claim, and the jury must be permitted to assign a percentage to a tortfeasor who is no longer a party.90 A credit that divides the claim equally among the tortfeasors is easier to run, because it requires only a head count, and the Rogers fire shows how it behaves. Rogers settles with Dunne, one of two negligent tortfeasors, and the jury later finds damages of $42,000. Rogers is treated as having sold half his claim, so Thurber owes $21,000 whatever Dunne paid, and Rogers ends with $33,000 if Dunne paid $12,000 and $56,000 if Dunne paid $35,000. A dollar credit on the same facts would have charged Thurber $30,000 and $7,000, leaving Rogers at $42,000 both times.91 Because this credit fixes the value of the remaining claim in advance, the plaintiff bears the whole risk of a low first settlement, which makes plaintiffs cautious about settling with the first tortfeasor.92 Plaintiffs therefore have less incentive to settle than under the dollar credit.93