Concept
Alternatives to Joint Liability
Apportionment of liability and contribution among tortfeasors
Alternatives to Joint Liability
Every apportionment rule answers one question: when part of a judgment cannot be collected from the tortfeasor who owes it, who absorbs the loss? Under traditional joint and several liability, each tortfeasor is individually liable for the full amount. The plaintiff may collect from any one of them or any group, and the risk of a co-defendant's inability to pay falls on the remaining defendants, not on her.1 The rule rests on causation. When the tortious conduct of each of two or more persons is a legal cause of harm that cannot be apportioned, each is subject to liability for the entire harm, whether the conduct was concurring or consecutive.2
Comparative fault does not logically compel a different answer. One court that kept joint and several liability after adopting comparative negligence found no persuasive authority that the one requires abolishing the other.3 Yet comparative fault supplied the argument for reform. Because a negligent plaintiff can now recover damages, most states have treated its arrival as a policy change large enough to justify a different rule.4 By one recent count, only eight states keep joint and several liability in its original form, about sixteen have adopted several liability (some with exceptions), and most of the rest use hybrids whose terms vary too widely for simple description.5
The alternatives differ in where an uncollectible share lands:
| Rule | Who absorbs an insolvent defendant's share |
|---|---|
| Joint and several liability | The remaining defendants |
| Pure several liability | The plaintiff |
| Threshold hybrid | Defendants above the line; the plaintiff when only defendants below the line remain |
| Type-of-damages hybrid | The remaining defendants for economic damages; the plaintiff for noneconomic damages |
| Reallocation | All other parties, the plaintiff included, in proportion to fault |
The Third Restatement sorts the hybrids by what triggers joint liability: a threshold percentage of fault, the type of harm suffered, or reallocation of shares that prove uncollectible.6 The sections below take pure several liability first, then each hybrid in that order, and end with a question every several-liability system must answer: may the jury assign fault to people who are not parties? These rules matter when the harm cannot be divided, because the jury's percentages rate each defendant's fault in causing a single, indivisible injury.7 Because statutes control, an exam answer should name the rule it applies and check the governing statute whenever the facts supply one.
Pure Several Liability
Under pure several liability, each defendant owes only the fraction of the damages that matches its own percentage of fault. Vermont's statute states the ratio: each defendant is liable for the proportion of the total damages that its causal negligence bears to the causal negligence of all defendants against whom recovery is allowed.8 Indiana's verdict procedure turns the ratio into arithmetic. The jury multiplies each defendant's percentage of fault by the damages and enters a verdict against each defendant for the product.9 A defendant found 60 percent at fault on $80,000 of damages therefore owes $48,000.10
Several liability swaps the causal logic of Restatement (Second) of Torts § 879 for a fault logic. Under § 879, each tortfeasor whose conduct is a legal cause of indivisible harm owes the entire harm.11 A jury that rates one defendant 15 percent at fault and another 85 percent is saying the first was far less negligent, even though each was a but-for cause of all of the plaintiff's damages. Several liability lets that comparison set what each defendant owes.12 A defendant's liability therefore shrinks when others share the blame, though its own conduct and the harm it caused have not changed.13
The illustration of Flite shows the arithmetic. The jury sets her damages at $60,000 and assigns fault of 20 percent to Flite, 50 percent to Micawber, 20 percent to Heap, and 10 percent to Murdstone. Flite's own share comes off first, leaving $48,000. Each defendant then owes a fraction of that sum equal to its fault over the defendants' combined 80 percent: $30,000 from Micawber, $12,000 from Heap, and $6,000 from Murdstone.14 Multiplying each percentage by the full $60,000 gives the same three figures. When every defendant can pay, Flite recovers $48,000 under either rule. Under joint and several liability she could collect all of it from Murdstone alone, whereas under several liability she must collect from each defendant separately, an arduous and expensive process in itself.15
The two rules part company only when someone cannot pay. Suppose Micawber is insolvent. Under several liability, Flite collects $18,000 from Heap and Murdstone and absorbs the other $30,000, because several liability casts the burden of the insolvent tortfeasor on the plaintiff.16 Under joint and several liability, Murdstone, at 10 percent fault, could be made to pay the entire $48,000 and left to seek contribution from the others.17
Walt Disney World v. Wood shows the same divergence with a defendant who is immune instead of insolvent. The jury assigned 85 percent of the fault to Daniel, 14 percent to the plaintiff, and 1 percent to the park. Daniel had married the plaintiff before trial and was therefore immune, so under joint and several liability the park owed 86 percent of her damages.18 Under proportionate liability, the park would owe only 1 percent.19 Each rule looks harsh to someone. Joint liability made a 1 percent defendant answer for 86 percent of the loss, while several liability leaves a plaintiff whom the jury held only 14 percent responsible uncompensated for 99 percent of it. The choice decides only who carries the unavoidable risk of a missing defendant.
Businesses likely to be sued and their insurers favor several liability for the reason this result shows: it limits a marginal defendant to its own share.20 Courts have also reasoned from consistency and fairness. Having adopted a rule that links liability more closely to fault, the Tennessee court thought it inconsistent to keep joint and several liability, which may fortuitously impose liability out of all proportion to fault.21 Another court put the point more bluntly: there is nothing inherently fair about a defendant 10 percent at fault paying 100 percent of the loss, and plaintiffs now take the parties as they find them.22 Courts that keep joint and several liability answer from the plaintiff's side. In return for allowing a negligent plaintiff to recover, fairness requires reducing her damages by her own percentage of fault. Eliminating joint liability would push the burden of an insolvent or immune defendant onto her and reduce her damages beyond that percentage.23
Contribution loses its function under several liability. Contribution lets a defendant who has paid more than its fair share recover the excess from the others, and a defendant who owes only its own percentage never pays an excess. The Tennessee court drew that consequence: situations in which a defendant pays more than its share would no longer arise, so the Uniform Contribution Among Tort-feasors Act would no longer apportion liability between codefendants.24 Micawber, who never pays more than his $30,000, has no claim against Murdstone, and no one has a claim against Micawber.25 Kottler v. State shows the consequence. A driver and the driver's employer settled with their passenger and then sought contribution from the State for negligent road maintenance. Washington had limited joint and several liability by statute, and its scope did not include their situation, so they were not entitled to contribution.26 Several liability also removes a procedural step. Proportional contribution under joint liability requires a suit establishing the plaintiff's recovery followed by a second action, or a motion, for contribution, and several liability needs only the first.27
Settlement bargaining changes as well. Under joint and several liability, a plaintiff could tell even a 10 percent defendant that any finding of negligence, even one percent, made it liable for the entire judgment. The argument was forceful because a right of contribution against more culpable co-defendants is often worth little when they cannot pay. Under several liability the threat disappears. Murdstone's exposure falls from the full $48,000 to $6,000, and a defendant in his position bargains harder.28 A settlement simply extinguishes the settling tortfeasor's share, and no one else needs a credit, because the others owe only their own shares. Had Flite settled Micawber's $30,000 share for $9,000, Heap would still owe $12,000 and Murdstone $6,000. Flite would collect $27,000 in all and bear the cost of the poor bargain.29 Later sections take up contribution and settlement credits where joint liability survives.
Threshold Hybrids
A threshold hybrid keeps joint and several liability for defendants who carry a large share of the fault and withdraws it from those who carry a small one. The statute fixes a percentage line. A defendant whose share reaches or exceeds the line, as the statute specifies, is jointly and severally liable, so the plaintiff may collect the whole judgment from it. A defendant below the line is only severally liable and owes just its own percentage.31 The line also decides who stands behind whom when a defendant cannot pay. An above-line defendant still covers a co-defendant's uncollectible share, while a below-line defendant covers no share but its own. The plaintiff therefore absorbs the loss whenever the only solvent defendants fall below the line.32
Legislatures reacted chiefly to the marginal defendant, whose tiny fraction of responsibility had required it to bear the full damages attributable to more culpable defendants who were wholly or largely insolvent.33 A threshold answers that complaint in two halves. Below the line, a defendant 10 percent at fault answers for 10 percent of the loss and no more. Above the line, joint liability survives, so the plaintiff keeps her guarantee of collection against the principal wrongdoers.34 The second half is the easier to defend, since those defendants contributed enough of the fault that it seems fairer for them than for the plaintiff to carry a co-defendant's insolvency. The first half does the contested work, because it moves the risk of some insolvencies onto the plaintiff, and where the line falls is a legislative choice that is to some extent arbitrary.35
The statutes draw the line in different places. Four show the range:
- Iowa makes a defendant found less than 50 percent at fault only severally liable, and keeps a defendant found 50 percent or more at fault jointly liable for economic damages (Iowa Code § 668.4).36
- New Hampshire allows a judgment for the full amount of the damages against a defendant found 50 percent or more at fault, and only several liability against a defendant found less than 50 percent at fault (N.H. Rev. Stat. Ann. § 507:7-e).37
- New Jersey sets a 60 percent threshold for joint and several liability, covering both economic and noneconomic damages (N.J. Stat. § 2A:15-5.3).38
- New York provides that a tortfeasor whose culpability is 50 percent or less is not jointly liable for the plaintiff's noneconomic damages, only for its proportionate share (N.Y. CPLR 1601(1)).39
Across the states, the line runs from as low as 15 percent to as high as 60.40 The wording at the line matters on an exam. Iowa and New Hampshire treat a defendant at exactly 50 percent as above the line, while New York treats a defendant at exactly 50 percent as below it for noneconomic damages.41 Iowa and New York also limit joint liability by type of damages, a second device taken up in the next section.
The park in Walt Disney World v. Wood is the marginal defendant these statutes protect. Found 1 percent at fault, it paid 86 percent of the damages under joint and several liability because the defendant who bore 85 percent of the fault was immune.42 Under a threshold statute with any line from 15 to 60 percent, the park would owe only its own 1 percent, and the plaintiff would bear the rest.43
Congress used the same sorting idea in the Private Securities Litigation Reform Act of 1995, which responded to the concern that peripheral defendants in fraud actions could be held liable for the full damages caused by the principal wrongdoer.44 The line there is knowledge, not a percentage. A defendant in a private fraud action is jointly and severally liable for the full amount only if the trier of fact specifically finds that it knowingly committed a violation. Otherwise it is liable solely for the portion of the judgment that matches its percentage of responsibility.45 The jury answers both questions by special interrogatory.46 The scheme reaches claims under Securities Act § 11, in which a purchaser may sue any enumerated defendant over a material misstatement or omission in a registration statement.47
Economic Versus Noneconomic Damages
California's Civil Code § 1431.2 is the model for hybrids that sort liability by the kind of loss. Voters adopted it by referendum in 1986 to counteract what the measure called “the deep pocket rule.” It makes each defendant's liability for noneconomic damages several only, so a defendant owes just its own percentage of the noneconomic award.49 Hybrids of this type keep joint and several liability for economic damages, such as medical expenses and lost income, and drop it for noneconomic damages, such as pain and suffering.50 The verdict therefore divides into two accounts governed by different rules, which is why the jury returns a special verdict separating them.51
The line may have two rationales. Out-of-pocket losses are the ones a plaintiff most needs paid, and keeping joint liability for them makes payment likely even when a defendant is insolvent or immune. The limit on noneconomic damages may instead reflect simple doubt that a money award can actually compensate intangible consequences.52 The cost falls on the plaintiff. A noneconomic share assigned to an insolvent or immune defendant has no one standing behind it and goes uncollected.
Iowa layers a damages line on top of a percentage line. A defendant found less than 50 percent at fault is only severally liable. A defendant found 50 percent or more at fault is jointly and severally liable for economic damages and not for any noneconomic damages.53 Crossing the percentage line therefore does not restore joint liability across the board, and treating the most culpable defendant as answerable for the whole award is the usual error.
The Jingle verdict puts both accounts in one judgment. Jingle's damages total $80,000, of which $20,000 is economic and $60,000 noneconomic. The jury finds Jingle 20 percent at fault, Weller 60 percent, and Snubbin 20 percent.54 Applying Iowa's statute, Jingle's own share reduces the economic damages to $16,000. Weller, above the line, owes all of it, including the $4,000 slice that matches Snubbin's fault, plus 60 percent of the noneconomic damages, or $36,000, for a total of $52,000. Snubbin, below the line, owes only 20 percent of the $80,000, or $16,000.55 If Snubbin cannot pay, Jingle loses nothing on the economic side, because Weller has already covered Snubbin's slice. She loses the $12,000 that matches Snubbin's 20 percent of the noneconomic damages, because the statute names no one to pay it. Under Indiana's pure several statute, which draws no line by type, Weller would owe 60 percent of $80,000, or $48,000. The extra $4,000 under Iowa's statute is Snubbin's economic slice.56
Other statutes use different numbers. Oregon tried one formula for a while and later replaced it with another complex provision, which shows how often legislatures redraft these lines.57 New York's protection for a tortfeasor 50 percent or less at fault does not extend to a tortfeasor liable for a tort requiring proof of intent or to a reckless one, since some wrongs are serious even when the comparative percentage is small.58 In Lubecki v. City of New York, a hostage was killed during a police confrontation with a bank robber. The court allowed the plaintiff to plead the recklessness exception against the city, which, if proved, would strip the city of its noneconomic protection.59
Reallocating Uncollectible Shares
Reallocation keeps joint liability's promise of collection and repairs its harshest result. Under Restatement (Third) of Torts: Apportionment of Liability § 21(a), when a defendant establishes that a judgment for contribution cannot be collected fully from another defendant, the court “reallocates the uncollectible portion of the damages to all other parties, including the plaintiff, in proportion to the percentages of comparative responsibility assigned to the other parties.”61 An uncollectible share is the portion assigned to a defendant who cannot pay, usually through insolvency, and every apportionment rule must decide who absorbs it. Joint and several liability leaves it with the solvent defendants, pure several liability with the plaintiff, and reallocation with everyone who bears responsibility for the loss.62 A very similar result comes from starting with several liability and adding a reallocation step for shares that prove uncollectible, so the starting point matters less than where the shortfall finally lands.63
Reallocation answers an objection to each pure rule. Under § 879, one defendant's insolvency falls on whichever solvent defendant the plaintiff collects from.64 That is easy to defend when the plaintiff is faultless. It is harder to defend when she was also at fault, because she then shares responsibility for the loss, and it seems more logical for her and the solvent defendant to carry the shortfall in proportion to their fault.65 Several liability fails the mirror-image test by assigning the whole shortfall to a plaintiff who may have been entirely without fault. Reallocation uses one ratio for both objections. The plaintiff's portion of the missing amount grows with her own percentage of fault and falls to nothing for a faultless plaintiff.
Justice Clark's dissent in American Motorcycle supplies a verdict that compares the rules: the plaintiff is 30 percent at fault, the first defendant 60 percent, and the second defendant 10 percent.66 Put the loss at $100,000 and suppose the 60 percent defendant is insolvent. American Motorcycle seems to leave the whole risk of an insolvent codefendant with the remaining defendants. Evangelatos v. Superior Court, a later California decision, applied a different rule: when tortfeasors prove insolvent, the shortfall “should be apportioned equitably among the remaining culpable parties—both defendants and plaintiffs.”67 The solvent 10 percent defendant's exposure then depends on the rule the jurisdiction follows:
- Under joint and several liability, as in American Motorcycle, the plaintiff holds a judgment for 70 percent of the loss against each defendant. The 10 percent defendant owes $70,000 and must seek contribution from a co-defendant who cannot pay.68
- Under reallocation, as in Evangelatos and the Third Restatement, the missing 60 percent is divided between the plaintiff and the solvent defendant in proportion to their fault of 30 and 10. The defendant absorbs 15 points of the 60 and the plaintiff absorbs 45, so the defendant owes its own 10 plus 15, or $25,000.69
- Under pure several liability, the 10 percent defendant continues to bear only 10 percent of the loss, or $10,000, and the plaintiff bears the rest.70
The reallocation figure comes from a ratio built only from the parties who remain. The insolvent defendant's 60 percent drops out, because the task is to divide that missing amount among those who can still bear it, and their percentages are 30 for the plaintiff and 10 for the solvent defendant. The solvent defendant takes 10/40, or one-quarter, of the missing 60 points, which is 15. The plaintiff takes the other 45.71 She absorbs three times as much of the insolvent defendant's share as the solvent defendant does because her fault was three times as great. A student who spreads the missing 60 points over all three parties, or drops them on the solvent defendant by reflex, has applied a different rule.
A faultless plaintiff stands at the other end of the range. With a percentage of zero in the ratio, the solvent defendant bears the full cost of the other's insolvency, which is the traditional result.72 Reallocation therefore departs from joint and several liability only to the extent the plaintiff was at fault. A solvent defendant owes at least its own share and never more than traditional joint liability would impose.
The National Commissioners on Uniform State Laws have supported the approach through the Uniform Apportionment of Tort Responsibility Act, and some states adopted the reallocation provision of the 1977 Uniform Comparative Fault Act, which the 2002 Act also contains.73 The Restatement excepts four categories from reallocation: intentional tortfeasors (§ 12), persons acting in concert (§ 15), vicarious liability (§ 13), and persons who fail to protect the plaintiff from the specific risk of an intentional tort (§ 14).74 Joint and several liability is the norm for intentional joint tortfeasors.75
Federal securities law offers a statutory relative of the idea. Under the Private Securities Litigation Reform Act, a defendant who did not knowingly violate the securities laws is generally liable only for its percentage of responsibility. If other defendants cannot pay because of insolvency, each remaining defendant must make an additional payment, up to 50 percent of its own liability, to cover the shortfall.76 A plaintiff whose net worth is under $200,000, and whose recoverable damages exceed ten percent of that net worth, can instead hold a non-knowing defendant jointly and severally liable for any uncollectible shares.77 The comparison clarifies the Restatement rule. The securities provision spreads the shortfall among the defendants and caps what each must add, while the Restatement spreads it among all other parties, the plaintiff included, in proportion to their percentages of fault.
Fault of Nonparties
Several liability raises a question the percentages cannot answer for themselves: whose fault goes on the verdict form? A wrongdoer can sit outside the lawsuit for many reasons. The person may be known only as a hit-and-run driver, may be immune from suit, may have settled with the plaintiff before trial, may be beyond the reach of process, may be the plaintiff's employer in a workplace injury, or may have been left out because collecting from that person looked hopeless.79 Jurisdictions split on whether the factfinder may give any of these absentees a percentage of the fault.80
The stakes follow from the arithmetic. Under several liability each defendant owes the share matching its own percentage, and the jury has only 100 percent to distribute. Every point assigned to an absentee is a point that no party pays, so the plaintiff holds that share uncollected. Leaving the absentee off the form has the opposite effect. The remaining percentages must then fill the whole 100 percent, and the parties who are present pay for fault that was never theirs. Counting the absentee shifts the loss to the plaintiff, and ignoring the absentee shifts it to the defendants.
Under traditional joint and several liability, the question matters far less to defendants, because each tortfeasor already owes the whole indivisible harm less only the plaintiff's own share.81 It can still matter to a partly negligent plaintiff, because her percentage depends on what it is measured against. An injured worker who sues only the manufacturer, for example, may have a smaller percentage when her fault is compared with the combined fault of the manufacturer and a negligent employer than when it is compared with the manufacturer's alone. The Restatement counts all tortfeasors, joined or not, in computing the plaintiff's share, although not every state agrees.82
Where joint liability is gone, the number of actors on the form changes what each defendant owes. A jury distributes at most 100 percent of fault, so two negligent actors average 50 percent each and twenty average 5 percent each. A defendant gains by finding more negligent actors even though its own fault is exactly what it was before they appeared.83 Defendants therefore argue that all fault must be counted, including that of nonparties, insolvent and immune tortfeasors, and intentional tortfeasors. Plaintiffs argue that only the fault of parties and of persons who can be sued should count, since otherwise the plaintiff collects less from the remaining defendant and nothing from the absent actor.84
Statutes sometimes settle the matter by limiting the comparison to parties. The choice is defended on the ground that, once joint liability is gone, counting nonparties would force a plaintiff to join many doubtful defendants lest the trier conclude that part of the fault belongs to someone not before the court.85 The contrary rule tells the trier to compare all fault that caused the harm, whether or not the actor is a party. Indiana's statute takes this course. The jury finds the percentage of fault of the claimant, the defendants, and any nonparty. It considers every person who caused or contributed to the injury, whether or not that person was or could have been named, and the parties' percentages may total less than 100.86
Procedural safeguards accompany the broader rule. Colorado requires a defendant to give formal notice of a nonparty's fault within 90 days of the start of the suit, presumably so the plaintiff can join that person, and Indiana puts the burden of proving nonparty fault on the defendant.87 Tennessee's court in McIntyre v. Balentine treated nonparty fault as an affirmative defense and directed the jury to assign the nonparty a percentage. The plaintiff can recover a judgment against that person only after a timely amendment of the complaint and service of process.88 Where nonparty fault counts, the nonparty usually must be someone the plaintiff had a legal right to sue, who at least prima facie owed her a duty, breached it, and proximately caused the harm.89 The federal securities statute also counts nonparties. It directs the jury to take account of persons claimed to have contributed to the loss, including those who settled with the plaintiff.90
The immune actor draws the sharpest disagreement. Tennessee lets a jury consider an immune employer's conduct in deciding whether a manufacturer is liable but forbids assessing fault against the employer, while Mississippi in Mack Trucks, Inc. v. Tackett permitted fault to be allocated to an immune employer.91
A jurisdiction that refuses to compare a nonparty's fault leaves the defendant one other route. The defendant may still argue that a divisible portion of the injury was caused by the nonparty and not by the defendant, which reduces its liability by causal apportionment instead of fault apportionment.92 That argument depends on the divisibility analysis taught earlier in this chapter, and it fails when the harm is a single injury no one can divide.
A nonparty whose conduct was intentional presents the problem at its sharpest. A negligent defendant whose fault is real but modest looks small beside a criminal assailant. Under several liability, the defendant may be charged with a sliver of the damages while the assailant, who is often judgment-proof, owes the rest, and the plaintiff absorbs the shortfall.93 A stylized verdict shows how much turns on the choice. Hamilton, the plaintiff, failed to repair a lock, Spencer, the defendant, failed to patrol, and Sly, a burglar, committed the burglary. If the jury looks only at Hamilton and Spencer, it might find Hamilton 25 percent and Spencer 75 percent at fault. If it also assigns fault to Sly, it might find Hamilton 10 percent, Spencer 20 percent, and Sly 70 percent.94 Two results follow from those figures:
- Under joint and several liability, Spencer owes 90 percent of Hamilton's damages if Sly is counted and 75 percent if Sly is ignored. Counting Sly lowers Hamilton's own share, and Spencer answers for everything but that share.95
- Under several liability, Spencer owes 20 percent if Sly is counted and 75 percent if Sly is ignored. Hamilton, who probably cannot collect Sly's 70 percent, loses heavily by his inclusion, which is why plaintiffs argue against counting intentional tortfeasors.96
Courts divide on whether negligent and intentional conduct may be compared at all, and the division often tracks statutory wording. In Welch v. Southland Corp., a thief shot a patron of a convenience store, and the Washington court held that the comparative-fault statute did not permit the jury to compare the store's negligence with the thief's intentional conduct.97 A reason supports that view. Comparative negligence replaced contributory negligence, which was traditionally no defense to an intentional tort, so legislatures enacting comparative negligence statutes probably did not intend intentional conduct to be compared.98 Other courts see the exclusion as artificial and reason that a more culpable party should bear the financial burden of its own intentional act, a position taken in a California decision, Weidenfeller v. Star & Garter.99
The Restatement answers the concern with a specific rule. A person who is liable to the plaintiff for failing to protect her from the specific risk of an intentional tort is jointly and severally liable for the share of comparative responsibility assigned to the intentional tortfeasor, in addition to the share assigned to that person.100 The rule follows from what the duty is. The defendant was bound to guard against exactly the intentional wrong that occurred, and a rule that shrank its percentage as the wrongdoer's culpability grew would reward the defendant when the very event it had to prevent took place. Full liability puts the risk of the assailant's insolvency on the negligent defendant, who may seek indemnity or contribution from the assailant, and spares the innocent plaintiff.101 Applied to the Hamilton verdict, Spencer would owe its own 20 percent plus Sly's 70 percent, or 90 percent of the damages, leaving only Hamilton's own 10 percent with Hamilton. Whether the rule is available depends on each jurisdiction's law, particularly whether a statute permits it.102