Concept
Judgments, Releases, and Settlements
Contribution and indemnity among tortfeasors
Judgments, Releases, and Settlements
Most multi-defendant cases end before judgment, when the plaintiff settles with one defendant. The deal reaches past the two parties who made it. The defendants who stay in the case want to know what they still owe, the settler wants to know whether anyone can pull it back in, and the plaintiff wants to know what a cheap early deal costs. Three goals pull against one another in the answers, and each rule in this section decides which goal gives way.
The first goal is one satisfaction for one harm. A plaintiff injured by the joint wrong of several parties “may recover his damages against either or all; but, although there may be several suits and recoveries, there can be but one satisfaction.”1 The second goal is early settlement, which depends on finality: a defendant who pays to settle wants to be sure that its exposure has ended.2 The third goal is fairness to the defendants who do not settle. They had no voice in the bargain, and the price is likely to fall well below the settler’s equitable share, because a settlement reflects the uncertainty of trial and gives the plaintiff funds to litigate against the rest.3
A settlement below the settler’s share leaves a gap in the plaintiff’s recovery, and the law has three places to put it. Comment m to Restatement (Second) of Torts § 886A addresses a release or covenant not to sue that does not purport to be a full satisfaction of the claim. It describes three solutions and concedes that none is satisfactory.4 Under the first, the settler’s payment is credited against the claim against the others, but a tortfeasor who later pays more than its equitable share keeps its contribution claim against the settler, so the settler can end up bearing the gap. Under the second, the credit is the same, but the payment also extinguishes any contribution claim against the settler, so the nonsettling defendants bear the gap. Under the third, the claim against the others shrinks by the settler’s equitable share of the obligation, whatever the settler paid, so the plaintiff bears the gap.5 The Restatement does not choose among them. The Institute takes no position on the effect of a release or covenant not to sue one tortfeasor on the other tortfeasors’ right to contribution from him.6
The first subsection fixes what a judgment, a release, and a satisfaction each do to the plaintiff’s claim. The second asks who may seek contribution after a settlement. The last two compare the two methods of crediting a settlement against a nonsettler’s liability, and both return to Dana.
Judgment, Release, and Satisfaction
Restatement (Second) of Torts § 886 equates two events. The discharge of a judgment against one of several tortfeasors, each liable for a single harm, is treated like a release or covenant not to sue given to one of them on a claim not yet reduced to judgment.7 A release surrenders the claim, usually for payment. A covenant not to sue promises not to enforce it and leaves the claim in existence. The equation sends a paid judgment to the law of releases, and that law has changed so much that the equation means something different today than it did at early common law.
Three events must be kept apart, because the early cases ran them together. A judgment is the court’s determination that a defendant owes a stated sum, and the plaintiff may take judgments against as many tortfeasors as the plaintiff can reach.8 Satisfaction is full compensation, by payment of a judgment or of a settlement that reflects the whole claim, and the plaintiff may have it only once.9 A release is a surrender of the claim itself, and it can be given for partial compensation or for none.10 An unpaid judgment is therefore neither a satisfaction nor a release, and a release for a fraction of the loss is not a satisfaction.
An unpaid judgment shows the change most plainly. Under the original English rule, the plaintiff could obtain only one judgment against one or more “joint tortfeasors,” a term then limited to those who acted in concert or were vicariously liable for each other. Because the act of one was the act of all, there was a single cause of action, which “merged” in the judgment, so a judgment against one, even if unsatisfied, barred any later action against another.11 Courts later replaced that rule with one that lets the plaintiff sue another responsible party so long as the first judgment has not been fully satisfied.12 Section 3(e) of the Uniform Contribution Among Tortfeasors Act codifies the later rule. A judgment against one tortfeasor does not of itself discharge the others “unless the judgment is satisfied.”13
The reason is the one behind joint and several liability itself. The plaintiff should be fully compensated as long as at least one tortfeasor can pay, and a judgment against an insolvent defendant compensates no one.14 Suppose Erg wins a $16,000 judgment against Kelvin, and Volt is also liable for the loss. If Kelvin pays in full, Erg has had the one satisfaction the claim allows and has nothing left to collect from Volt. If Kelvin is bankrupt and the judgment goes unpaid, Volt remains liable. Erg may sue Volt and obtain a second judgment, though she can never collect more than her assessed damages.15 A problem that says only that the plaintiff holds a judgment against one tortfeasor has not told you the claim is satisfied.
The early law treated a release the same way. If the act of one was the act of all, a release of one released all, and confusion about who counted as a “joint tortfeasor” carried the rule over to defendants who acted independently but were liable for the same damages.16 Release also came to be confused with satisfaction. In Abb v. Northern Pacific R. Co. (Wash. 1902), a release was treated as “conclusive evidence of satisfaction” even though it expressly reserved rights against other tortfeasors.17 The original American rule was accordingly that a release of one released every tortfeasor liable for the same damages, whatever had been paid.18
The principle behind that rule was sound, since a plaintiff who has been paid in full has no loss left to recover and the rule avoided unjust enrichment.19 The error was to treat the release as proof that payment had occurred. A release given for a small part of the loss leaves the plaintiff far short of compensation, and the one-satisfaction principle gives no reason to extinguish other claims until the plaintiff has actually been paid. The rule also discouraged settlement. A plaintiff who released one of several tortfeasors released all, so could settle with any one of them only for something close to the value of the whole case.
The covenant not to sue became the plaintiff’s way around the rule. Pringle’s loss was $50,000, and she settled with Whiting for $40,000. If the settlement is a release, the strict common-law rule bars any suit against Shattuck, the other negligent party, though $10,000 of the loss is uncompensated. If it is a covenant not to sue, and the jurisdiction holds that a covenant waives no rights against other tortfeasors, Pringle may still sue Shattuck. A $50,000 judgment is then reduced by the $40,000 already paid, and Shattuck owes $10,000.20 The covenant works by a formal distinction. The plaintiff keeps the cause of action but promises not to enforce it against the settling defendant, who is protected by a counterclaim for breach of the promise.21 Courts generally treat the covenant as a procedural device invented to avoid the traditional effect of a release. Counsel simply styled the settlement as a covenant, and the release rule became a dead letter.22
The same facts have produced opposite outcomes where the governing rule differed. In Tyler v. Ahmed (11th Cir. 1987), a plaintiff who had settled with her own auto insurer for all of her injuries could not afterward bring a malpractice action against the physician who treated them. In Morgan v. Cohen (Md. 1987), the plaintiff likewise settled an auto-accident claim under a general release and then sued a physician for negligent treatment. The court held that the release did not discharge the physician as a matter of law, because Maryland had replaced the common-law rule by adopting the Uniform Contribution Among Tortfeasors Act.23 The difference lay in the rule each court applied. Tyler followed the common-law approach, under which a release of one tortfeasor ends the claim against the rest. Morgan asked whether the parties intended to release the physician and whether the settlement had compensated all of the damages.24
Decisions and statutes have changed the original rule extensively, by several routes:
- Some states distinguish a release, which discharges the other tortfeasors, from a covenant not to sue, which does not unless full compensation has been paid.25
- Some treat a release that expressly reserves rights against other tortfeasors as a covenant not to sue, whatever words of release it uses.26
- Some hold that even a release with no reservation leaves the others liable unless it shows an intention to release them or the plaintiff has been fully compensated, as in Leung v. Verdugo Hills Hospital (Cal. 2012).27
- Some have statutes. Under § 4 of the Uniform Joint Obligations Act, a release with an express reservation of rights does not release the others. Under § 4(a) of the Uniform Contribution Among Tortfeasors Act, the others are released only when the release’s terms so provide.28
The last route is the one to know best. The Uniform Act provides that 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 any of the other tortfeasors from liability for the injury or wrongful death unless its terms so provide.”29 Restatement (Third) of Torts: Apportionment of Liability § 24(b) takes the same position for a jurisdiction that has overruled the old release rule outright. A release or covenant given to one tortfeasor does not release the others unless it so stipulates.30 The shift separates two questions that the old rule had merged. What the release surrenders depends on what the parties agreed, so the document is read for its terms. How much the plaintiff has actually received is the satisfaction question, and the one-satisfaction limit continues to protect the other tortfeasors from paying for a loss already compensated. The credit rules in the next subsections carry out that protection.
Because the statutory test turns on the release’s terms, the wording of the document carries the result. In Douglas v. United States Tobacco Co. (8th Cir. 1982), the court read the phrase “all other persons” in a general release to satisfy the “if the release so provides” condition of the local Uniform Act. Any later tortfeasor was discharged regardless of the compensation actually received, and a majority of jurisdictions follow that position.31 A release also need not be a separate document. Insurers often print a general release on the face of a settlement check, and in Cupidon v. Alexis (Md. 1994) that language operated as a release once the plaintiff cashed the check.32
With the release rules in view, § 886 does its work. A paid judgment ends the plaintiff’s claim against the tortfeasor who paid. Whether it also ends the claim against the others turns on the analysis a release would receive: what the discharge provides and, in states that ask, whether the plaintiff has been fully compensated.33 Putting the paid judgment on the same footing as a release keeps the two ways of ending a claim consistent and avoids the old error of treating a judgment or a document as proof of payment. What a nonsettling defendant still owes after a partial payment, and who may seek contribution after a settlement, are the subjects of the subsections that follow.
Settlement and Contribution Rights
Restatement (Second) of Torts § 886A(2) allows contribution only to a tortfeasor who has discharged the entire claim by paying more than its equitable share, and it limits recovery to the excess. No tortfeasor must contribute beyond its own equitable share.35 A settlement is one way to pay, so the discharge requirement turns the question into what the settlement bought. A defendant who pays and obtains a release that ends the plaintiff’s claim against every tortfeasor has done what a defendant who pays a judgment does. A defendant who buys peace for itself alone has discharged only its own debt. A second question points the other way: whether a defendant who did not settle may collect contribution from one who did. The two questions have different answers, and the settler comes first.
The general rule, absent a statute to the contrary, lets a tortfeasor who settles the plaintiff’s entire claim before trial, and so has no judgment entered against it, obtain contribution from the other tortfeasors whose liability the settlement discharged.36 Two burdens come with the claim. The settler must prove that the others were liable to the plaintiff, and it must prove that the settlement was reasonable.37 Each burden fills a gap that a judgment would have closed. A judgment fixes the defendants’ liability to the plaintiff and binds them in the contribution action, while a settlement fixes nothing. The reasonableness requirement guards the price. The other tortfeasors had no voice in the bargain, and without a check on the figure a settler could overpay to end the case quickly and bill the surplus to defendants who never agreed to it.
The release must actually reach the tortfeasor from whom contribution is sought. Section 1(d) of the Uniform Contribution Among Tortfeasors Act says that a tortfeasor who settles with the claimant “is not entitled to recover contribution from another tortfeasor whose liability for the injury or wrongful death is not extinguished by the settlement.”38 The provision applies the discharge requirement of § 886A(2) to settlements. Contribution reimburses a payer for relief its payment gave to someone else, and a settlement that leaves the other tortfeasor liable to the plaintiff gave that tortfeasor nothing. The Act also provides that a good-faith release discharges the other tortfeasors only if its terms so provide, so the document itself decides whether the settler has a claim.39 A settler that wants contribution rights needs a release that discharges the others by its terms.
Twain’s settlement with Nash shows the cost of a release that stops at the settler. Twain accepts $16,000 from Nash and releases Nash alone, then tries the case against Benchley, and the jury assesses damages at $20,000. Benchley receives a dollar credit for Nash’s $16,000 and owes $4,000. Nash, who paid $6,000 more than half of the damages the jury found, cannot recover that sum from Benchley, because the release did not extinguish Benchley’s liability.40 The result looks harsh in hindsight but reflects what Nash bought. He avoided the expense of a trial and the risk of a much larger verdict: an award of $100,000 would have cost him at least $50,000, and perhaps the whole amount if Benchley were insolvent.41 A settler that paid for certainty cannot demand a refund when the risk it avoided never materialized.
Two limits from the earlier sections still apply once the release reaches the other tortfeasor. The settler recovers only the excess over its own equitable share, so a settlement at or below that share leaves nothing to collect even under a broad release. And the tortfeasor from whom the settler collects never owes more than its own equitable share.42
Whether a defendant that did not settle may collect contribution from one that did is the harder question, because the Restatement leaves it open. The Caveat to § 886A states that the Institute takes no position on the effect of a release of one tortfeasor, or a covenant not to sue him, on the other tortfeasors’ right to contribution from him.44 The Caveat blocks a tempting inference. A student who knows that § 886A gives contribution among persons liable for the same harm may assume that a released tortfeasor stays in the pool, but the Restatement does not say so, and the answer comes from the jurisdiction’s statute or decisions. As the introduction showed, only the first of comment m’s three solutions leaves the settler exposed to contribution.45
Legislatures have divided along the same line. The 1939 Uniform Act let a nonsettling tortfeasor seek contribution from the settler.46 The 1955 Act provides that a release or covenant not to sue given in good faith “discharges the tortfeasor to whom it is given from all liability for contribution to any other tortfeasor.”47 Massachusetts follows the 1955 approach, Restatement (Third) of Torts: Apportionment of Liability § 23, comment i endorses it, and many states now hold that contribution cannot be obtained from a settling joint tortfeasor.48 The position is not universal. The Maine court in Lavoie v. Celotex Corp. refused to let the policy of encouraging settlement operate to the detriment of nonsettling defendants.49
The reason for the bar is the settler’s need for finality. A defendant settles to exchange a payment for certainty. If a nonsettling defendant could later sue the settler for contribution, the settler would still face paying more after a trial it did not attend, and its money would have bought nothing. A defendant in that position would do better to wait for the verdict, and the contribution claim would become the reason that no one settles.50 The bar removes that risk. In the jurisdictions that adopt it, a defendant who pays a settlement is sure that its exposure has been extinguished, even though the resulting allocation of the loss may be disproportionate to fault.51
The cost of the bar falls on the defendant that stays in the case. Rogers’s claim against Dunne and Thurber is worth $42,000. Dunne settles for $12,000 and obtains a release of himself alone. Thurber remains liable for the whole judgment but receives a dollar credit for Dunne’s payment, so he pays Rogers $30,000.52 Dunne’s equal share was $21,000, so Dunne paid $9,000 less than his share, while Thurber pays $9,000 more than his. Because the bar discharges Dunne from contribution, Thurber cannot make Dunne pay anything beyond the $12,000. His remaining recourse is to challenge the settlement’s good faith, a point taken up below.53 The bar also gives every defendant a reason to settle first, since a defendant who persuades the plaintiff to accept less than half the likely damages avoids paying its full share.54
The two directions share one design: the settlement price fixes the settler’s final burden. A defendant who strikes a favorable bargain shifts more than half of the loss to the nonsettler, and one who strikes an unfavorable bargain keeps more than half of it. Had Dunne paid $35,000 and left Thurber unreleased, Thurber would owe only $7,000 after the credit, and Dunne could not recover the excess from Thurber, because § 1(d) denies contribution to a settler whose settlement did not extinguish the other’s liability.55
The bar protects only a good-faith settlement, and the requirement answers a danger the bar creates. A discharged settler has no reason to care how much the nonsettler ends up paying, so a plaintiff and a friendly defendant could agree on a token price and leave the plaintiff to collect the rest from the defendant who stayed. The settlement must therefore be in good faith.56 Courts divide on what that means. Some ask only whether the deal was collusive, and others also ask whether the settlement was adequate.57 An adequacy inquiry has to be cautious. The parties’ valuation of a claim turns on subjective factors, such as the risk that the settler’s negligence cannot be proved at trial or the effect of a sympathetic defendant, and a court is unlikely to find bad faith merely because the settler paid less than a pro rata share.58
Where the court finds good faith, the nonsettling defendant’s relief comes through the credit rules discussed next. Where it does not, the statutory discharge fails and the settler remains exposed to contribution.
Dollar Credit Rule
Under the dollar credit rule, also called the credit or pro tanto rule, the award against the nonsettling defendant is reduced by the amount the settler actually paid. If one defendant settles for $30,000 and the plaintiff wins $100,000 against the other, the plaintiff collects $70,000.60 The Uniform Contribution Among Tortfeasors Act reaches the same result by reducing the claim against the others by the amount stipulated in the release or paid for it, whichever is greater.61 The Restatement endorses no treatment, so each jurisdiction chooses.62 Where tortfeasors remain jointly and severally liable, so that the plaintiff may collect a judgment from any one of them, the dollar credit is one of the two main answers. The proportionate credit, taken up next, is the other.63
Return to Dana. Frost owes the $100,000 verdict less the $20,000 Ellis paid, or $80,000, and Dana collects $100,000 in all, the one satisfaction the law allows.64 Ellis’s equal share was $50,000, so he paid $30,000 less than his share, and Frost bears $30,000 more than his. The settlement’s discount fell on the defendant who stayed in the case. In a jurisdiction that discharges a good-faith settler from contribution, Frost cannot shift any of that excess back to Ellis.65
The plaintiff’s side of the rule is clearest when the settlement is small. Twain’s damages are $20,000, and he settles with Nash for $5,000. Benchley remains liable for the entire judgment and receives credit only for the $5,000 Nash actually paid, so Benchley owes $15,000 and Twain ends with the full $20,000.66 A jurisdiction that credits the settler’s proportionate share treats Twain as having given up half of his claim when he settled, so Benchley owes only $10,000 and Twain ends with $15,000.67 That is why plaintiffs have less incentive to settle under the proportionate approach than under the dollar credit.
| Approach | Settlement with Nash | Jury award | Benchley pays | Twain receives in all |
|---|---|---|---|---|
| Dollar credit | $5,000 | $20,000 | $15,000 | $20,000 |
| Proportionate credit | $5,000 | $20,000 | $10,000 | $15,000 |
The same arithmetic explains why the rule invites early and modest settlements. Justice Stevens explained in McDermott, Inc. v. AmClyde & River Don Castings, Ltd., 511 U.S. 202 (1994), that “the settlement figure is likely to be significantly less than the settling defendant’s equitable share of the loss, because settlement reflects the uncertainty of trial and provides the plaintiff with a ‘war chest’ with which to finance the litigation against the remaining defendants.” Under the credit, that discount lands on the defendants who stay in the case.68 The rule has the advantage of producing one precise number for the second defendant to pay. Its disadvantage is that it encourages each defendant to settle early so that the other is left holding the bag.69 A defendant who persuades the plaintiff to accept less than half the likely damages avoids paying a full share.70
The credit comes off the judgment. In Pringle’s case, a $50,000 judgment less the $40,000 Whiting paid leaves $10,000 for Shattuck.71 It follows that a settlement larger than the verdict leaves nothing owed, and the settler cannot claw back the surplus, because it chose the price when it bought peace.72 Because the credit measures dollars and not fault, Frost’s burden moves with the price of Ellis’s bargain. Had Ellis settled for $60,000, Frost would owe $40,000. Had Ellis settled for $5,000, Frost would owe $95,000. Frost’s exposure swings by $55,000 with no change in anyone’s fault.
Justice Clark, dissenting in American Motorcycle Association v. Superior Court, pressed that point to its conclusion. A credit measured in dollars may turn the principle that liability follows fault upside down, because the extent of dollar liability may end up in inverse relation to fault.73 Take his percentages: the plaintiff is 30 percent at fault, the first defendant 10 percent, and the second 60 percent, on a $100,000 loss. The plaintiff’s judgment is $70,000, and if the 60 percent defendant settles for $20,000, the credit leaves the 10 percent defendant owing $50,000, half of the entire loss. A rule that carves the settler’s share out of the judgment would let the plaintiff pursue the 10 percent defendant for only 10 percent of the loss, however much the plaintiff obtained from the settler.74 The inversion arises because the two halves of the rule point in different directions. Joint and several liability makes the defendant who stays answer for the whole, and the credit subtracts only the dollars paid, so a lightly at-fault defendant can absorb what a heavily at-fault settler left unpaid.
A verdict may carry other reductions besides the credit, and the order in which they are applied changes the answer. The credit must be sequenced with any reduction for the plaintiff’s own fault and with any statutory cap on damages, and courts do not all sequence them alike. Whalen v. Kawasaki Motors Corp. is a New York example of the problem.
The credit has work to do only where the nonsettler is liable for more than its own share. In a state that has abolished joint and several liability, each remaining defendant answers only for its own share, so there is nothing for a credit to offset. Where joint liability survives only for some damages, such as economic damages, the credit applies only to that part of the judgment. The settler’s protection from contribution carries its own check, since the settlement must be in good faith, as the preceding subsection explained.75
Proportionate Credit Rule
The third solution in comment m to Restatement (Second) of Torts § 886A is the proportionate share rule. Under it, the settlement extinguishes the plaintiff’s claim against the settler and diminishes the claim against the others by the settler’s equitable share of the obligation.77 The dollar credit subtracts what the settler paid, and this rule subtracts what the settler owed. The credit is therefore the same whether the settlement was generous or stingy. The Restatement endorses neither approach, because the Institute took no position on how a release of one tortfeasor affects the contribution rights of the others, so each jurisdiction chooses.78
The rule is easiest to see on a verdict whose percentages are known. Nell sues Fagan and Twist and settles with Fagan for $25,000. At trial against Twist, the jury finds her damages are $100,000 and assigns fault of 20 percent to Nell, 70 percent to Fagan, and 10 percent to Twist. With Nell’s own share removed, the judgment is $80,000, for which Twist is jointly and severally liable. A dollar-credit jurisdiction subtracts the $25,000 Fagan paid, and Twist pays $55,000. A proportionate-credit jurisdiction treats Fagan as having sold his proportion of the judgment to Nell for the settlement amount. Twist’s credit is 70/80ths of the $80,000, or $70,000, and he pays $10,000 instead of $55,000.79 The credit is figured against the judgment already reduced for Nell’s fault, which is why Fagan’s 70 sits over 80 and not over 100.80 The percentages that fix how much of the loss Twist finally bears also fix his credit, and the $25,000 never enters his arithmetic.
The Supreme Court chose among comment m’s alternatives in McDermott, Inc. v. AmClyde, an admiralty case. McDermott, the crane owner, sued AmClyde, the manufacturer and general contractor of the crane, River Don Castings, the maker of the crane’s main hook that broke, and three parties involved in the manufacture and sale of steel cable slings. The sling defendants settled for $1 million the weekend before trial. The jury assessed McDermott’s damages at $2.1 million and allocated 32 percent of the responsibility to AmClyde and 38 percent to River Don.81
The Court adopted the proportionate share approach, under which “no suits for contribution from the settling defendants are permitted, nor are they necessary, because the non-settling defendants pay no more than their share of the judgment.”82 River Don therefore owed its own 38 percent, or $798,000, whatever the sling defendants had paid. AmClyde had been found immune under a provision of its contract with McDermott, and the Court characterized that immunity as a quasi-settlement in advance, so River Don’s liability stopped at its own percentage even under established principles of joint and several liability.83 The rule has continued to gain ground in admiralty. The Ninth Circuit applied it in Foss Maritime Co. v. Corvus Energy Ltd., 878 F.3d 1144 (9th Cir. 2017), observing that strict product liability damages are apportioned by proportionate fault in the personal injury context and that the same should hold for property damage.84
The reasoning starts from how the rule pictures the plaintiff’s claim, as two divisible claims, each with a set value. If each of two defendants bears half the loss, a $30,000 settlement with the first makes half of a $100,000 claim disappear and leaves the second owing $50,000, however much or little the first paid.85 That picture answers a weakness of the dollar credit. Under the dollar credit, the settlement discount falls on the defendants who stay in the case, so each defendant has a reason to settle first and leave the other holding the bag. Under the proportionate rule, the discount falls on the plaintiff alone, and the settlement price no longer shifts any of the loss to the other defendant. Justice Stevens put the point this way: “a plaintiff’s good fortune in striking a favorable bargain with one defendant gives other defendants no claim to pay less than their proportionate share of the total loss.” One virtue of the rule, he added, is that it does not make a litigating defendant’s liability depend on the amount of a settlement negotiated by others without regard to its interests.86
Three considerations guided the Court. The crucial one was consistency with the proportionate-fault rule of Reliable Transfer, which had displaced the older admiralty practice of dividing the loss equally.87 If each defendant’s final burden is set by its percentage of fault, the credit for a settler should follow the same percentage and not an unrelated dollar figure. The Court regarded the other two considerations, the effect on settlement and judicial economy, as ambiguous.88 On settlement, the rule removes the reward for settling cheaply and first, which discourages early deals, but it also frees the settler from contribution claims without any fairness inquiry, which encourages them. On economy, the nonsettling defendant must prove the settler’s fault at trial just as it would in a contribution claim, which adds to the trial, but later contribution suits and fairness hearings disappear.89
The plaintiff gets the benefit of the settlement but must accept its detriment.90 In Dana’s case, Ellis’s equal share removes half of the $100,000 claim, so Frost owes $50,000 whatever Ellis paid. Dana ends with $70,000, bearing the $30,000 by which Ellis’s $20,000 fell short of his $50,000 share. Had Ellis paid $60,000, Dana would end with $110,000. Nell’s case shows the same thing. Fagan’s 70/80ths was worth $70,000, she sold it for $25,000, and she collects $35,000 in all against an $80,000 judgment. Restatement (Third) of Torts: Apportionment of Liability § 16 approves the allocation. “Since the plaintiff bears the risk of an inadequate settlement (in which case the plaintiff will recover less than the damages determined by the factfinder), the plaintiff should also obtain the benefit of a favorable settlement.”91 The price of the principle is visible in Twain’s case. A $5,000 settlement of a $20,000 claim leaves him $15,000 here against $20,000 under the dollar credit, so a plaintiff is unlikely to sell half a claim for much less than half its value.92
Removing the settler’s share from the judgment does not turn the rest into several liability. The Court reiterated joint and several liability in the same decision, so each defendant who stays in the case remains answerable for the whole judgment that is left. Those defendants, and not the plaintiff, carry the risk that a co-tortfeasor in the case cannot pay. Settlers are off the hook in both directions, because the nonsettlers have no contribution claim against them and a settler cannot seek contribution from the nonsettlers.93 Jurisdictions that reallocate uncollectible shares do it differently. The Third Restatement reallocates the uncollectible portion to all other parties, plaintiff included, in proportion to their comparative responsibility, with exceptions for intentional tortfeasors, persons acting in concert, vicarious liability, and persons who fail to protect the plaintiff from the specific risk of an intentional tort.94 The risk that remains with the defendants who stay is also a strong motivation for a defendant to settle, since the settler escapes it entirely.95
The bar on contribution also closes a route a settling defendant might otherwise try. In the Fifth Circuit, a defendant that settled with the injured party and took an assignment of the injured party’s claims could not then sue the other responsible parties, because that would be an easy end run around the prohibition on contribution claims between a settler and a co-tortfeasor and would undermine the proportionate share rule.96 The court distinguished a Mary Carter agreement, in which the settling defendant stays in the suit and is reimbursed out of part of the plaintiff’s recovery. That is a reimbursement of a partial interest and not an assignment of the whole claim.97