$15

  Table of Contents

I. INTRODUCTION1

II. THE DEBATE OVER PUNITIVE DAMAGES4

A. Who Cares about Punitive Damages?4

B. What Are Punitive Damages?7

C. A Brief History of Punitive Damages9

D. Out of Control?10

III. TO PUNISH AND DETER: The Case for the Criminal Law11

A. The Trouble with the Debate11

B. Why We Can't Deter Everything, and Why Only Intent to Harm and Recklessness

Should Carry Punitive Sanctions13

C. Why We Should Prefer to Punish Using the Criminal Law17

D. Some Advantages of Criminalizing Punitive Sanctions19

IV. CASE STUDIES: Dos and Don'ts24

A. Those Who Were Rightly Punished25

B. Those Who Were Wrongly Punished: Exxon and the Valdez26

C. Those Who Were Rightly Not Punished29

V. OTHERWISE, MORE DUE PROCESS PROTECTIONS32

A. Should Juries Be in this Business Anyway?33

B. Why Jury Guidance Is Essential34

C. The Supremes Speak: Haslip, TXO, BMW36

D. More Defendant Protections39

VI. SO HOW DO WE SET PUNITIVE DAMAGES?42

A. How Much Did the Defendant Benefit?43

B. The Effect of the Regulatory System43

C. A Checklist for Calculating Punitive Damages45

D. Multiple Assessments of Punitive Damages46

E. The Red Herring of Caps and Ratios48

F. When the Wealth of the Defendant Shouldn't Matter54

VII. CONCLUSION57

ABOUT THE AUTHOR59 

The week that the jury in the Exxon Valdez trial came in with their final verdict, juror Nancy Provost's granddaughter had been learning about big numbers in her fifth-grade math class. "Billions and millions and trillions," says Provost. "She goes, `I've got to find a big number, over a hundred thousand dollars.'" Provost handed her the front page of the September 17 Anchorage Daily News. The lead story was an article on the previous day's punitive damages verdict against Exxon Corporation. "$5,000,000,000," screamed the headline. "I said, `How about this?'" Provost recalls. "My son-in-law says, `You know, Grandma made that money.'" Provost and ten other jurors did indeed "make" that $5 billion verdict, a big number in a fifth-grade classroom, a big number in a corporate boardroom—even if the boardroom happens to belong to a multibillion-dollar corporation like Exxon—a big number, period.

—Emily Barker, The American Lawyer 

I. INTRODUCTION

The doctrine of punitive damages is one of the most hotly contested aspects of the judicial system. Objections to the doctrine of punitive damages go all the way to the 19th century, when Justice Foster of the New Hampshire Supreme Court said of punitive damages:

The idea is wrong. It is a monstrous heresy. It is an unsightly and an unhealthy excrescence, deforming the symmetry of the body of the law.

Justice Foster was referring to the symmetry by which tort law was to be entirely compensatory and criminal law entirely punitive. More recently, legal scholar Peter Huber has criticized punitive damages as having "an open-ended, anything-goes quality that can too easily stoke the ambitions of eager plaintiffs, the zealous advocacy of their lawyers, and the vindictive or sympathetic passions of juries." According to legal commentator Walter Olson, punitive damages are "a peculiar holdover from the law's moralistic past... [a] vestigial tailbone of the civil law."

Punitive damages in environmental cases are every bit as contested as punitive damages in the tort system at large. There are few theoretical questions that are different in environmental cases than in other cases. But questions about punitive damages are especially acute in environmental cases, especially in "toxic tort" litigation, which generally relates to injuries caused by the production or handling of hazardous materials, like asbestos or pesticides. Blurry questions of causality and unclear methods of calculating penalties—both hallmarks of environmental law—compound the problem. The largest punitive award ever levied against a corporation—$5 billion—was assessed in an environmental case—against Exxon in the Valdez case.

In today's popular debate over punitive damages, defendants in lawsuits tend to maintain that the doctrine of punitive damages is fatally flawed. Meanwhile, plaintiffs hold that punitive damages are necessary to obtain justice and that any reform of the system would put the "little guy" at the mercy of "Big Business."

This paper will argue that neither the critics nor the supporters of the punitive damages system have it quite right. The present popular debate over punitive damages (both in environmental cases and in general) is largely misconceived, painting a false dichotomy between environmental protection and corporate profits. Contrary to what defenders of the tort system say, there is a problem with punitive damages, though contrary to what tort reformers may argue, the problem is not large punitive damages awards per se. The problem exists not because punitive damages are assessed in "crisis" proportions—no one knows what "crisis" proportions are. Rather, the problem is that punitive damages are often assessed in inappropriate situations or in unjustifiable amounts.

Formulations like "Punitive damages are too high" or "Punitive damages are too low" suffer from the same problem as statements like "Too many people are convicted of murder." Such statements, based on aggregate results, lose sight of the purpose of the law, which is to establish a fair process for achieving a fair result in individual lawsuits. In the legal system, a fine that is $100 too high and one that is $100 too low do not cancel each other out. A solution to the punitive damages problem must explain what a "correct result" is and how to achieve it in individual cases. In this paper, I will argue that:

•There are important differences between civil and criminal law. Civil law is primarily a private law whose goal is to settle individual disputes arising from individual harm, and to compensate injured parties. Criminal law assigns blame and metes out punishment for transgressions against the social order and the collective moral code.

The boundaries between these two areas of the law have been blurred in recent years, but these boundaries should be reestablished. Punishment should be reserved for intentional misdeeds—recklessness, intent to harm, and intentional violations of the law. Punishment is inappropriate for accidents and merely negligent behavior, for which compensatory damages are enough.

•Morally blameworthy actions which are now punished with punitive damages in civil cases should instead be punished through the criminal law. Some advantages of the criminal law are that:

a)In criminal law, the burden of proof is higher, which is appropriate because of the moral stigma attached to a criminal conviction and the need to protect the innocent from being wrongly convicted.

b)Criminal fines go to the state and not to the injured party (though the injured party may also bring a civil suit for compensatory damages), which is appropriate because: 1) awarding punitive fines to injured parties is a bad way to encourage plaintiffs to sue and is no substitute for an attorney's fee recovery system; 2) once compensatory damages have already been paid, the defendant's wrongdoing, if it is truly wrong, is an offense against society at large; and 3) allowing the injured party to keep punitive fines can give people too much of an incentive to sue in borderline frivolous cases.

c)Criminal punishments, because they are laid out in statutes, are generally more predictable than jury-determined punitive damages awards, and are therefore fairer because they give potential wrongdoers advance notice of the likely consequences of their conduct.

d)Criminal law only punishes wrongdoers once for each offense (instead of once for each injured party), which is appropriate because the magnitude of a criminal offense lies in the reprehensibility of its intent and not the harm it caused or how many people it affected. Indeed, criminal law even allows punishment when, by some fluke, no one was injured (a case where the tort system wouldn't allow recovery).

e)Decisions to prosecute criminal violations rest with public authorities vested with the task of punishing criminal conduct, which is appropriate because criminal violations are offenses against societal norms and not just injuries to a particular person.

•However, if the civil law continues to be used to impose punitive damages, we should keep in mind that:

a)The civil law is still an inappropriate vehicle for imposing punishment. Any "punitive damages" award will still be punitive, since it will impose costs in excess of the harm caused, so procedural safeguards are necessary. But the punishment rationale should be dropped. Deterrence should be the only goal of punitive damages awards.

b)Juries themselves aren't the problem; the more fundamental problem is unlimited discretion, whether of juries or judges. Punitive damages reform must involve at least the procedural safeguards mandated by the Supreme Court in the Haslip case: clear jury instructions, post-verdict review by the trial court, and appellate review.

c)In addition, punitive damages should incorporate those features pointed out above as advantages of the criminal law. The burden of proof for awarding punitive damages should be higher; plaintiffs shouldn't keep punitive damages awards; punishments should be more predictable; and multiple awards of punitive damages for a single action should be curtailed.

d)To effectively deter, total fines—including regulatory penalties, compensatory damages, and punitive damages—should be equal to, or slightly higher than, the economic benefit to the defendant of his reckless or malicious conduct. Therefore, all penalties already incurred by defendants, including regulatory fines or compensatory damages, should be subtracted from this number to yield the optimal amount of punitive damages. Multipliers may be appropriate, in cases where the underlying conduct was hard to detect.

e)Any sort of monetary cap on punitive damages awards is arbitrary, has nothing to do with deterrence, and should be avoided. The ratio between compensatory and punitive damages should also be irrelevant. Relying on this sort of simple formula, or using a ratio as a limit on punitive damages, makes it more difficult to come up with appropriate deterrent fines, and it compounds any previous errors in the calculation of compensatory damages and regulatory fines.

f)The wealth of the defendant generally should not be a consideration in establishing damages. 

II. THE DEBATE OVER PUNITIVE DAMAGES

A. Who Cares about Punitive Damages?

Aggregate statistics on punitive damages don't tell us whether the tort system should be reformed. But a look at some of the numbers can help us understand the arguments of the tort reformers, and imagine the magnitude of the problem, to the extent there is one.

Since the 1970s, punitive damages awards have been increasing in frequency and in amount. In particular, punitive damages awards in environmental tort cases, especially toxic tort cases, have increased. A 1992 study, Punitive Damages Explosion: Fact or Fiction?, notes that total punitive damages awards in Texas, California, Illinois, and New York increased from an average of $800,000 in the 1968–1971 period to an average of $312.1 million in the 1988–1991 period—by a factor of 390, or by a factor of 117 if we adjust those numbers for inflation. The average award in the first period was only $1,080, while the average award in the second period was $778,000—nearly equal to the entire amount of punitive damages awarded during the first period.

The Supreme Court has expressed "concern about punitive damages that `run wild.'" Good numbers are hard to come by because no comprehensive reporting system exists; a great many punitive damages awards are reversed or reduced by higher courts, which further complicates the matter. But some numbers provide evidence of this trend:

•In Alabama, juries awarded over $200 million in punitive damages in 1994. From 1974 to 1978, aggregate punitive damages affirmed by the Alabama Supreme Court were $409,385. From 1979 to 1983, they were $4,239,766. From 1984 to 1988, they had risen to $33,167,497, and, finally, from 1989 to 1993, they had climbed to $89,616,379.34. Increases in punitive damages have been observed in many jurisdictions, but Alabama has reportedly earned a reputation for punitive damage awards that "grow thick and wild" and are "becoming the norm."

•A review of Jury Verdicts Weekly, which reports only a portion of California jury verdicts, reveals that California state juries handed down punitive damage verdicts totaling over $1.7 billion from 1990 to 1994 in 263 cases—for an average of about $6.5 million per verdict.

•One 1992 study analyzed state and federal Texas court decisions affirming punitive damage awards where at least one litigant was a business, to find out whether there has really been an "explosion" of punitive damage awards. While from 1968 to 1971, the total punitive damages affirmed in Texas were $85,000, that number had risen to $127,591,000 for the period 1988–1991—an increase by a factor of 1,500 in one generation. From 1992 to 1994, in state court cases alone, Texas appellate courts affirmed $186,683,294.60 in punitive damages.

•A Pacific Research Institute study of 1,024 lawsuits filed in January 1991 in San Francisco County Superior Court shows that: 1) punitive damages are demanded in 27 percent of all cases where they are conceivably recoverable; 2) business and government defendants are four times as likely as an individual defendant to face a lawsuit that demands punitive damages; 3) lawsuits that include punitive damage demands take one-third longer to resolve (21 months) than suits without these demands (15 months); 4) the probability of a punitive damage award if a case proceeds to trial, contains a punitive damages demand, and is against a business defendant, is estimated to be 14 percent.

While punitive damages have been around for a long time, they didn't "skyrocket" until the late 1970s and 1980s. According to Victor Schwartz and Mark Behrens, the increase was due to three important legal developments in the late 1960s and early 1970s:

•Courts moved away from applying punitive damages only in the area of intentional torts and started applying them in the new field of products liability (paralleling other changes in tort law, including changes in the scope of recoverable injury);

•With the advent of "mass tort" litigation, courts allowed punitive damages to be awarded repeatedly, for different plaintiffs, for what was really one act; and

•Courts began to allow punitive damages in contract actions like breach of an implied duty of good faith and fair dealing.

Courts have also come to look upon punitive damages as an entitlement of injured plaintiffs; they have therefore been reluctant to overturn punitive damages awards, even in cases where the same defendant has already paid punitive damages to many plaintiffs for the same actions. As Judge Henry Friendly put it in the 1967 case of Roginsky v. Richardson-Merrell:

We know of no principle whereby the first punitive award exhausts all claims for punitive damages and would thus preclude future judgments; if there is, Toole's judgment in California [a previous punitive damages award against the same manufacturer], which plaintiff's brief tells us came earlier, would bar Roginsky's. Neither does it seem either fair or practicable to limit punitive recoveries to an indeterminate number of first-comers, leaving it to some unascertained court to cry, "Hold, enough," in the hope that others would follow. While jurisprudes might comprehend why Toole in California should walk off with $250,000 more than a compensatory recovery and Roginsky in the Southern District of New York and Mrs. Ostopowitz in Westchester County with $100,000, most laymen and some judges would have some difficulty in understanding why presumably equally worthy plaintiffs in the other 75 cases before Judge Croake or elsewhere in the country should get less or none.

The aggregate numbers also come with tales of some seemingly arbitrary individual punitive damage awards, like the $2.9-million punitive damage award against McDonalds, in the New Mexico case where a woman spilled hot coffee on herself after, as she explained, she "put... [a styrofoam coffee cup with a plastic cover] between [her] knees and tried to get the top off that way." The trial judge reduced the award, but only to $490,000 because he, too, wanted to send a message to McDonalds and "punish and deter" its corporate coffee policy (which, by the way, was based on the company's internal surveys of how hot its consumers wanted their coffee to be). As one commentator has put it, "today, hardly a month goes by without a multi-million dollar punitive damages verdict in a product liability case."

B. What Are Punitive Damages?

Damage payments in civil cases can be divided into two components: compensatory damages and punitive damages. Compensatory damages include several components:

•Pecuniary damages, which compensate successful plaintiffs for actual out-of-pocket expenses, like medical expenses or lost wages from injuries;

•Nonpecuniary damages, which compensate people for nonmonetary losses, including "pain and suffering"; and

•Hedonic losses, for lost pleasure in life.

Compensatory damages, by definition, already fully compensate plaintiffs to the extent allowed by the law. But in addition, courts are allowed to make defendants pay punitive damages. The goal of punitive damages is not to compensate the injured plaintiff (though the plaintiff does receive the money), but to punish and deter the injurer.

Most environmental cases where punitive damages are an option are of the "nuisance" model, though many of the decisions awarding punitive damages rest on other liability theories like negligence, trespass, or strict liability (that is, where the defendant may have to pay damages regardless of whether he was negligent) for especially dangerous activities. Under the general nuisance model, the plaintiff is usually a property owner or occupant who has suffered some harm to his person or property from the actions of another, and the defendant is a past or current owner or occupier of some other land. Often, the defendant owns land adjacent to a plaintiff's land; many of the cases involve air or water pollution, water diversion, mining operations, blasting, vibrations, noise, flooding, obstructions, and the like. A subset of nuisance and strict liability litigation involving spills, releases, burial, or disposal of toxic wastes or other substances that have caused injury to people or property is called "toxic tort litigation."

A majority of jurisdictions allow punitive damages in civil cases. (Table 1 indicates the states that have eliminated punitive damages or statutorily reduced the number of situations in which they are permissible.) According to supporters of the current punitive damages system, these are some reasons for using punitive damages:

Source: Kagan, "Toward a Uniform Application of Punishment," p. 773, nn. 107–115. 

•Assume that a misdeed has occurred, which we want to discourage unconditionally. If the perpetrator is likely to repeat the offense even though he has to pay compensatory damages, extra damages may have to be assessed to discourage the activity. For instance, if the wrongdoer profits by more than the damage he causes, having to pay only compensation would still leave him ahead, unless the costs of the litigation exceed the difference.

•If the misdeed is hard to detect, or if people are unlikely to sue, then only a few victims will be compensated. The offender will only pay a portion of the costs imposed by his conduct, so some extra damages may have to be assessed.

•Because of legal costs and other considerations, compensatory damages may not be enough to motivate people to sue when they have a good case. Assume that someone is harmed by actions that are reprehensible but not criminal, and that the harmed party is reluctant to sue. Then the financial lure of a punitive damage award may be necessary to lead them to bring willful wrongdoers to justice. This is called the "private attorney general" argument.

•Punitive damage awards may provide compensation to plaintiffs whose actual damages exceed those that the law allows them to recover through compensatory damages.

These are the justifications for punitive damages that often appear in the popular debate; some are valid, and some are not. But first, some history.

C. A Brief History of Punitive Damages

The notion that total damages can exceed mere compensatory damages has a long history. The Babylonian Hammurabi Code, the Hindu Code of Manu, and the Bible all contain references to the doctrine of multiple damages. The Romans assessed multiple damages for certain offenses, and justified them by invoking the need to constrain wealthy elites. In ancient Rome, quadruple damages were a creditor's remedy against debtors who didn't pay their debts for a year; while part of the multiple damages may have been some substitute for interest, a 300 percent interest rate is a bit high, and so we must conclude that the primary purpose of this fine was punitive. Closer to home, in the 18th century, courts assessed "exemplary damages" against the government for its oppressive treatment of a dissenting newspaper in the companion cases of Wilkes v. Wood and Huckle v. Money. From then on, English courts used exemplary damages to punish and deter the misuse of wealth and power.

In America, punitive damages were first used in the 1780s and 1790s. The first such award was in Genay v. Norris, a malicious poisoning case. Punitive damages were later awarded in cases such as Coryell v. Colbaugh, a suit over breach of promise of marriage, and Boston Manufacturing Co. v. Fiske, a suit over patent infringement. Punitive damages were generally used against bullies who oppressed the weak and powerless, in cases of assault and battery, rape, and (what would now be called) sexual harassment. By the end of the 19th century, the application of the doctrine shifted from powerful individuals to large corporations. Railroads often had to pay punitive damages to women, invalids, and children who were badly treated by conductors, porters, and other railway employees. Judges and juries awarded punitive damages in cases where the defendant exhibited willful and gross (but not criminal) disregard of a plaintiff's rights.

D. Out of Control?

Even assuming that punitive damages are justified, are the awards in environmental cases too high? Some point to the $5-billion award against Exxon as an example in the affirmative. The punitive damage award was upheld by an Alaska federal district court on January 27, 1995, though it may still be appealed when the entire Exxon case is done. The Exxon spill was the sixth largest on record; in 1978, the Amoco Cadiz ran aground off the coast of France, spilling 68.7 million gallons of oil—six times as much as the Valdez.

Critics of punitive damages assert that:

•Compensatory damages, in many cases, are adequate to deter harm, and punishment, the traditional domain of criminal law, is not an appropriate goal for civil law.

•The winners and losers in the redistribution of wealth wrought by punitive damages awards aren't obvious. Common wisdom has it that corporations are the ones who are punished, but corporations themselves can feel no pain. The ultimate costs of punitive damages awards are borne by a company's employees, stockholders, the consumer, or all of these; excess money paid by corporations translates into potentially higher consumer prices, and lower profits translate into potentially lower dividend payments on the company's stock. One commentator presents the effect of Exxon's punitive damage assessment this way: "60 percent to 70 percent of Exxon stock belongs to individuals rather than institutions. After tax deductions, Exxon is already out $2.5 billion, $2 a share. The punitive damages could cost another $4 a share. There's that much less in assets to back the pensions of possibly millions of people."

•Offsetting a plaintiff's litigation expenses isn't an appropriate function of civil litigation (aside from provisions that explicitly allow successful parties to collect attorney fees). If the law doesn't allow plaintiffs to recover certain damages (for instance, attorneys' fees, or as-yet-unthought-of forms of pain and suffering), they shouldn't be able to recover for those damages through the catch-all proxy of punitive damages. If the plaintiffs don't like these restrictions, they can push for changes in the law, but using punitive damages as a back-door way of collecting money the law doesn't entitle them to is a blunt instrument for a delicate job.

•Multiple awards of punitive damages to multiple plaintiffs, which are often assessed, especially in consumer products liability cases, are excessive punishment.

Despite its $5-billion punitive damages award, Exxon is still prospering, but this may be misleading. Exxon reported a profit of $5.28 billion in 1993 on revenues of $111 billion, and according to Standard & Poor's, it still has a coveted triple-A credit rating (which is the highest one can get). But Exxon hasn't had to pay the $5 billion yet. According to Philip Dodge, senior vice president at Southeast Research Partners in Florida, if appeals fail and Exxon has to pay the full $5 billion, the company's debt ratio could increase by as much as 30 percent, and a big part of operating profit could be used to service the debt. Any money actually paid may have a delayed effect; the full payment period on the $1 billion settlement with the state of Alaska, for instance, extends past the year 2000. 

III. TO PUNISH AND DETER: The Case for the Criminal Law

A. The Trouble with the Debate

Defenders of the system of punitive damages point out that while punitive damages critics concentrate on products liability and medical malpractice, the expansion of punitive awards has mainly been in the areas of intentional torts and business/contract actions; the areas that get the most media attention aren't the areas with the most punitive damages awards. According to some commentators, "judges and juries award punitive damages with striking rarity to individuals in suits against manufacturers." The point is also made that statements like "the average punitive damage award increased, in inflation-adjusted dollars, from $43,000 in 1965–1969 to $729,000 in 1980–1984—a jump of 1500 percent" are misleading, in that they report mean awards, not median awards. When dealing with means, a small number of very large awards can dramatically skew the results.

Just about every substantive point made by one side in the debate is disputed by the other side. For instance, means are often more useful than medians. For instance, if a company faces n punitive damage awards, where $µ is the mean award and $m is the median, the total amount that the company will have to pay in punitive damages will approach $nµ as n increases. The median is an essential statistic, and should be used together with the mean, but alone isn't very useful to the company trying to calculate its expected liability exposure.

But to merely point out that punitive damages are high shows nothing. All punishments seem high to the person being punished.

But one can quibble with the rationale for imposing punishment in many cases. As Kenneth Adams of Dickstein, Shapiro & Morin, one of the lead plaintiffs' attorneys in the Exxon case, put it, the fine sends a signal to Exxon; "prevention is better than recklessness. What went wrong was fundamental bad management." This isn't literally true; "prevention" and "recklessness" aren't opposites. It's possible to have accidents without being reckless. Political scientist Aaron Wildavsky draws distinctions between "anticipatory" societies, which try to prevent all accidents, and "resilient" societies, which try to find better ways of managing accidents when they occur. Anticipation, Wildavsky argues, has its place but can't be pushed too far. In the first place, preventing all accidents is impossible, since, by definition, though one can predict the likelihood of an accident, one can't know them all. Accidents are inevitable, even with all possible care. Second, trying to anticipate all possible harm or risk means that resources will be misspent pursuing expected accidents, which would lower the resources available to improve technology generally, enhance safety elsewhere, or be able to respond resiliently when an inevitable but unanticipated accident occurs.

In short, there is such a thing as too much prevention.

Nonetheless, punitive sanctions may sometimes be warranted. If someone achieves ill-gotten gains through reprehensible activity, then forcing him merely to give up these gains, once discovered, may not discourage him from engaging in the activity in the first place. If compensatory damages are all one has to pay, one may perceive advantages to being reckless or malicious and just paying the money if one is discovered. But a key to these hypothetical scenarios is some notion of recklessness, maliciousness, or intent to harm.

The question "What makes gains ill-gotten?" should be the central question of the punitive damages debate. This section will argue that:

•We can't deter everything we don't like, nor do we want to.

•Deterrence gets more expensive as we have more of it, and it gets more expensive faster and faster. The cost of each marginal gain is higher than the previous one.

•When the activity we're trying to deter is an accidental by-product of an otherwise socially useful activity, the costs of getting rid of the activity can exceed the benefits of altogether eliminating the by-product. The only way to get the "optimal" amount of such activities is by making everyone bear their full costs through payment of compensatory damages.

•We should reserve punishment for cases where there are truly "ill-gotten gains"—intent to harm or violate the law, or instances of recklessness.

•In such cases, the criminal law is a more effective means of punishing bad behavior (and gives government the opportunity to punish blameworthy acts, even when they caused no injury). Tort law should be reserved for compensation.

•The criminal law will be fairer in the following respects:

1)the burden of proof is higher when imposing punishment than when requiring compensation;

2)the criminal fines go to the state and not to the injured party (though the injured party may also bring a civil suit for compensatory damages);

3)punishments are more predictable, because appropriate ranges of punishment are written into the statute establishing the criminal violation;

4)because of the guarantee against double jeopardy, the problem of multiple punishment for the same cause of action does not exist; and

5)decisions to prosecute rest with public authorities vested with the task of punishing criminal conduct and endowed with the discretion to distinguish between violations they feel are worth prosecuting and those they don't.

B.Why We Can't Deter Everything, and Why Only Intent to Harm and Recklessness Should Carry Punitive Sanctions

We can distinguish four types of damage cases:

•Intent or knowledge—where the defendant caused damage intentionally or knowingly;

•Recklessness—where the defendant knew and consciously disregarded a substantial and unjustifiable risk that the damage would occur;

•Negligence—where the defendant should have known, but didn't, of a substantial and unjustifiable risk that the damage would occur; and

•Innocent accidents—where the damage occurred, but the risk of its having happened was justifiable or insubstantial.

Pinning down just what "substantial" and "justifiable" mean in this context may involve having recourse to generally accepted industry standards, the currently existing regulatory structure, or the ever-elusive "reasonable person."

In the first three of these cases, compensatory damages must be paid. That the plaintiffs should be compensated, or "made whole," for the damage they've suffered isn't disputed. In the case of innocent accidents, where a case really can be made that no one was at fault, the costs rest where they fall. So either the law provides for the compensation of victims, or it denies them the right to recover altogether; in either case, the compensation question is taken care of. The question, though, is: When should defendants be punished for their behavior?

The first two categories, intent and recklessness, involve knowledge and usually some deliberation. In either case, the defendant intends to either harm someone or to put someone under a substantial and unjustifiable risk of harm. "From a moral standpoint," Judge Friendly tells us, "there is not too much difference between the driver who heads his car into a plaintiff and the driver who takes the wheel knowing himself to be so drunk that he probably will hit someone and not caring whether he does or not." These sorts of cases, we want to punish, because intent to harm is reprehensible. Indeed, the Model State Punitive Damages Act advocates a malice standard: "The plaintiff must establish that the defendant's actions showed malice. This burden of proof may not be satisfied by any degree of negligence including gross negligence." So far, 13 states require by statute that plaintiffs seeking punitive damages establish that the defendant acted with malice (in at least some types of lawsuits), and two states require proof of malice through case law (see Table 2).

Negligence and innocent accidents, though, are different. Accidents will always happen, even under the most responsible management. This isn't hyperbole; unless one completely avoids a particular industry, eliminating all chance of an accident is literally impossible. We can reduce accidents, though; by spending more resources in prevention, we can prevent more accidents, but these efforts cost money, and the higher the level of safety, the more it costs to prevent each additional accident. If we were to try to eliminate all accidents, we would end up going to extreme lengths, spending countless resources that would be better spent elsewhere. For example, research conducted at the Harvard Center for Risk Analysis indicates that different "life-saving interventions" can have wildly different costs per life-year saved (see Table 3). 

Source: Rustad and Koenig, "The Historical Continuity of Punitive Damages Awards," p. 1279, n. 64; and other sources. 

The most expensive life-saving intervention the Harvard group looked at was chloroform private well-emission standards at 48 pulp mills, which cost $99,351,684,000 per life-year saved. That's almost $100 billion. That's 100 million times the amount of money needed to save one life-year through flu vaccinations (at $1,000 per life-year saved). A liability rule that encourages too much investment in trivial safety devices takes resources away from countless other uses, some of which may save more lives. In other words, there is such a thing as overdeterrence. The way to deter appropriately is to make people bear the full costs of their actions, and that goal is already served by compensatory damages (which include out-of-pocket costs, as well as "pain and suffering" awards).

As for negligence, it's tempting to say that all one has to do to be nonnegligent is to avoid negligence. But not acting negligently is more complicated than it seems. Negligence, after all, is when you don't know, but should know, of a substantial and unjustifiable risk that damage will occur. In other words, negligence isn't an intentional act, but is a failure of knowledge. Failures of knowledge are unintentional and therefore, in a sense, "accidents"; like innocent accidents, negligent accidents can be remedied, but only by spending resources to educate oneself and act in accordance with that knowledge. By spending more and more money, one can identify and remedy one's negligent activities—but totally eliminating accidents, negligent or otherwise, is impossible. Negligent actors should bear the costs of their actions. If they are forced to do so, they will avoid acting negligently as long as the costs of avoidance are less than the costs of the negligent accidents themselves. Thus, compensatory damages are already an adequate deterrent to negligence.

Courts and legislatures, when describing the sort of conduct justifying punitive damages, tend to use phrases like "willful, wanton misconduct," "reckless or conscious disregard for the safety of others," or "oppressive, fraudulent, malicious, or outrageous conduct." As one classic handbook on the law of damages describes the purpose of punitive damages:

Since [punitive] damages are assessed for punishment and not for reparation, a positive element of conscious wrongdoing is always required. It must be shown either that the defendant was actuated by ill will, malice, or evil motive (which may appear by direct evidence of such motive, or from the inherent character of the tort itself, or from the oppressive character of his conduct, sometimes called "circumstances of aggravation"), or by fraudulent disregard of the rights of others. "Gross negligence" is a somewhat ambiguous expression. In the sense of extreme carelessness merely, it would probably not suffice, but only when it goes further and amounts to conscious indifference to harmful consequences.

Or, as another commentator puts it more specifically, in the case of chronic environmental damage:

1. In those exceptional circumstances where an activity is engaged in for a purely improper purpose or motive, liability predicated on the actual or express malice standard is a distinct possibility.... Moreover, when there exists direct or circumstantial evidence of spite, ill-will, or revenge as the real purpose or motive behind the activity, punitive damage liability reaches its highest probability. In these cases, risk and utility considerations are largely irrelevant because the utility which the defendant gains (satisfaction from seeing harm inflicted) is not a kind of utility that is socially recognizable or legally cognizable....

2.In the decisions in which the punitive damage liability was sustained on the basis of evidence supporting a standard of recklessness, conscious indifference to the rights of others, or similar standard, there were present several common factual demonstrations.

1)Defendant possessed a knowledge or awareness that its operations were discharging or emitting some hazardous or harmful substances into the air, water or ground....

2)Defendant possessed the knowledge or awareness that its activities were producing harm to the plaintiff or invading the plaintiff's rights to the beneficial use of its land....

3)In all of the cases the defendant had knowledge of a means or method by which to reduce or eliminate or abate the risk or harm resulting from its activities....

4)After the defendant is possessed of the knowledge identified in items 1), 2) and 3) and then fails to act, the risk of punitive liability attaches.

This is the theory—intent and recklessness can support a punitive damages decision; lesser wrongs can't. In practice, though, these conditions are often observed in the breach. Courts often stray from these rules of thumb. The standard varies from "reckless indifference and disregard of the law" to a standard just short of actual intent to commit the specific act in question. Different courts differ on to what extent one can imply the defendant's state of mind from his conduct.

Some courts divorce the availability of punitive damages from the precise nature of the defendant's conduct, by considering violations of laws (even without intent) to be grounds enough for considering punitive damages. The boundaries set up by environmental laws may demarcate the line that someone has to cross for his actions to exceed simple carelessness. In Tant v. Dan River, homeowners sued the corporation Dan River for negligence in allowing its boiler system to emit a black, sooty material which damaged their homes and property. The South Carolina Supreme Court upheld the punitive award, saying that:

The testimony and evidence at trial indicated Dan River was aware of emission problems with its boiler system prior to receiving complaints from the homeowners. Internal memoranda indicated Dan River had knowledge the boiler system was in "poor mechanical shape because of age," and was "on the ragged edge" relative to compliance with state regulations. Approximately three weeks after the homeowners initially complained of damage from emissions, a Dan River employee filed an internal report confirming that emissions from the boiler chimney were "at times well above the state and federal air pollution control standards" and reporting he had recorded as many as ten violations in a single day. A jury question as to punitive damages is presented when there is evidence of a statutory violation.

This is how, in fact, punitive damages are often applied. But the actual circumstances where punitive damages may be appropriate are not quite as broad as indicated in the above standard from the Dan River case:

•If someone is harmed as a result of malice or recklessness, punishment is appropriate.

•If someone deliberately violates a law, for instance, by engaging in criminal activities such as fraud, punishment is appropriate.

•If the harm is a result of an accident or mere negligence, punishment isn't appropriate (though compensatory damages and statutory fines may be appropriate).

Sometimes, determining whether something was intentional, reckless, negligent, or accidental is more difficult than it seems. In most asbestos litigation, the plaintiff worked in a shipyard during World War II, in an atmosphere permeated with asbestos dust. In most jurisdictions, juries determine whether the defendant's conduct during the 1930s and 1940s can be considered "conscious disregard for safety." But it's difficult for a jury to forget its modern notions of what is currently common knowledge or acceptable, reasonable behavior—including its modern awareness of environmental pollution, health hazards, and medical technology. Half a century ago, responsible business practice was a lot different, and much of today's knowledge about health and environmental risks simply didn't exist. As one commentator put it, "It is a laborsome task to take a jury back those same 20 or more years, arm them with the information then available, and ask them to plot the course of conduct for a defendant manufacturer, disregarding the medical state of the art as it exists [today]." And yet it must be done.

C. Why We Should Prefer to Punish Using the Criminal Law

The criminal law already exists as a way of punishing intentionally bad behavior. In 1958, Henry Hart, law professor at Harvard University, described different proposed distinctions between criminal and civil law:

•"Can crimes be distinguished from civil wrongs on the ground that they constitute injuries to society generally which society is interested in preventing?" No; society's also interested in people honoring contracts and avoiding traffic accidents.

•"Does the distinction lie in the fact that proceedings to enforce the criminal law are instituted by public officials rather than private complainants?" No; the government brings all sorts of civil enforcement actions—for instance, for an injunction or to recover a civil penalty.

•Are crimes any things that are called crimes? "So vacant a concept is a betrayal of intellectual bankruptcy. Certainly, it poses no intelligible issue for a constitution-maker concerned to decide whether to make use of `the method of the criminal law.' Moreover, it is false to popular understanding, and false also to the understanding embodied in existing constitutions. By implicit assumptions that are more impressive than any explicit assertions, these constitutions proclaim that a conviction for crime is a distinctive and serious matter—a something, and not a nothing. What is that something?"

The key—the "something," as Hart puts it—is society's moral condemnation. We condemn murder, for instance, in a way that we don't condemn run-of-the-mill (non-alcohol-related) traffic accidents. According to John Coffee, professor of law at Columbia University, what most distinguishes criminal law from civil law (and particularly from tort law) is:

its operation as a system of moral education and socialization. The criminal law is obeyed not simply because there is a legal threat underlying it, but because the public perceives its norms to be legitimate and deserving of compliance. Far more than tort law, criminal law is a system for public communication of values. As a result, the criminal law often and necessarily displays a deliberate disdain for the utility of the criminalized conduct to the defendant. Thus, while tort law seeks to balance private benefits and public costs, criminal law does not..., possibly because balancing would undercut the moral rhetoric of the criminal law. Characteristically, tort law prices, while criminal law prohibits.

This generally ought to be the case, but in practice the distinction is sometimes blurred. One big exception to the rule is the practice of awarding punitive damages in civil cases. This is, in fact, what Justice Foster of the New Hampshire Supreme Court meant when he said that punitive damages "deform[ed] the symmetry of the body of the law."

But the distinction between civil law and criminal law isn't merely academic. There are real differences between the two sorts of law. For one thing, civil law requires a plaintiff. This is problematic for punitive damages. Compensation centers on the harm to the plaintiff, but punishment centers on the inherent reprehensibility of the act. How, then, does one punish reprehensible behavior that (fortunately) didn't harm anyone? The criminal law already has a way of dealing with this. In criminal law, it's the bad act, not the injury, that creates the cause of action, so the government can punish wrongdoers for their intent, whether or not they harmed anyone.

There are other differences between civil and criminal law. For example:

•The standard of proof required to impose punishment is higher in criminal law—which is appropriate, since punishment is a more serious matter than compensation, and we should therefore maintain high levels of protection to avoid mistakes in punishment;

•The injured party doesn't keep criminal fines—which is appropriate, since:

1)punitive damages are a blunt and inappropriate way of compensating plaintiffs for taking the trouble to sue;

2)the defendant's conduct should be properly thought of as an offense against society at large, not an offense against the plaintiff in particular (since the plaintiff has already been compensated as far as the law allows), so the damages awards should go to the state; and

3)if kept by the plaintiff, punitive damages awards can introduce an element of moral hazard on the plaintiff's side, since they can give him too large an incentive to sue in borderline frivolous cases.

•Criminal punishments are usually more predictable than civil fines, since criminal violations generally have fixed, definite penalty ranges. While judges, as a general rule, have had broad discretion, they are bound by the minimum and maximum penalties in the statute. This is appropriate, because fairness requires that potential offenders be put on notice as to the specific consequences of their actions.

•The guarantee against double jeopardy ensures that no one is criminally punished twice for the same cause of action.

•Decisions to prosecute criminal violations rest with public authorities vested with the task of punishing criminal conduct and endowed with the discretion to distinguish between violations they feel are worth prosecuting and those they don't. This, too, is appropriate; since the defendant's conduct is a harm to society at large, the decision of whether and how to prosecute him should rest with representatives of society at large—that is, the executive branch.

As the following sections will explain, these are important safeguards to maintain because of the serious nature of punishment. By contrast, the current, ad hoc, system of determining punitive damages has significant costs. According to lawyer Theodore Olson of the Civil Justice Reform Group,

Despite their similarity to criminal fines, punitive damages are generally imposed without any of the systemic protections afforded by our form of government to shield defendants from arbitrary and extreme punishments in criminal cases.... In criminal cases, arbitrary action is checked to a significant degree by the division of the power to punish among the three branches of government that typically wield power at the federal and state levels. The legislative branch has the responsibility prospectively to define proscribed conduct and suitable levels of punishment. The executive branch has the duty to serve as the "disinterested prosecutor with the unique responsibility to serve the public, rather than a private client, and to seek justice that is unfettered." The judicial branch customarily imposes a penalty within legislative constraints after considering the recommendations of the prosecutor, and administers a panoply of procedural safeguards to ensure that alleged wrongdoers are treated fairly.

Thus, before criminal punishment may be imposed, an individual has the right to foreknowledge of a generalized but clear legislative statement of what is prohibited and what shall be the penalty for deviation from the prescribed norm; a "prosecution" by a public official sworn to uphold the public welfare, bound by an oath to comply with the Constitution and held accountable to seek justice and not retribution or private gain; and a judicially supervised enforcement proceeding to ensure that punishment may be imposed only for violating clear standards and that punishments will be consistent, predictable and not excessive.... These important institutional controls... are almost completely lacking in the punitive damage system.

D. Some Advantages of Criminalizing Punitive Sanctions

1. Higher Burdens of Proof

For compensatory damages, the burden of proof is not very stringent. For criminal prosecutions, on the other hand, the burden of proof is substantial, as anyone who has watched the O.J. Simpson trial now knows. As Olson puts it, "Criminal defendants receive numerous procedural rights not available to punitive damage defendants [in a civil suit]. For example, a criminal defendant may be entitled to a specific finding of criminal intent beyond a reasonable doubt by a unanimous jury before he can be fined $500, but the same individual or corporation may be fined $5 million or $5 billion for the same conduct in a civil suit with none of these protections." Moreover, rights like the privilege against self-incrimination usually don't apply in civil cases.

Why are higher burdens of proof necessary for punishment?

Let's first consider the pitfalls of awarding compensatory damages. When awarding compensation, courts can make two sorts of mistakes. One is to deny a compensatory award to a deserving plaintiff, unjustly leaving the defendant with his money. Another mistake is to award compensatory damages to an undeserving plaintiff, unjustly depriving the defendant of his money. In the compensatory damages context, we usually feel that either of these injustices are equally bad, and so we feel no need to guard against one of them more vigilantly than against the other.

In the criminal context, we think differently; as the saying goes, "Better to release n guilty men than to convict an innocent man." The greater we feel n should be, the more willing we are to guard against unjustly convicting the innocent, and the more stringent our burden of proof should be.

How does this apply to the punitive damages context? The punitive damages verdict, as its name implies, is punitive; it comes with the stigma of blameworthiness attached to it, and it doesn't serve to compensate anyone. If someone's unjustly assessed a punitive damages award, that person loses a lot of money and reputation. But if the opposite mistake is made—if a punitive damages award is not awarded where it should be, there may be not enough deterrence, but none of the parties to the present case are actually hurt. All plaintiffs have already been compensated by compensatory damages.

Since the costs of unjustly awarding punitive damages are much greater than the costs of mistakenly not awarding them, it makes sense to use higher standards of proof for punitive damages verdicts. One obvious way of doing this is by actually prosecuting intent to harm and recklessness cases criminally, and calling punitive damages "criminal fines."

2. The Plaintiff Wouldn't Keep the Money

According to Olson:

The punitive damage system... is driven exclusively by private litigants and their lawyers, who have a personal, private interest in the outcome of the litigation and are unfettered by executive duties and responsibilities or by the accountability imposed by the democratic process. Indeed, plaintiffs' lawyers are bound to serve their private clients, not the public, and almost always have a direct, material and purely personal interest in inflicting the greatest possible punishment in every case.

Of course, this only describes the actions of those who set the machinery in motion. Just because lawyers want the greatest punishment doesn't mean that punishment is inevitably rendered. The lawyers have to convince a jury, and the defendants have an opportunity to convince the jury otherwise; the decision is still the jury's. But allowing private plaintiffs to keep punitive fines does carry with it some problems.

First, since punitive damages punish offenses to society at large, not to the individual injured party (who has already been compensated by compensatory damages), allowing the injured party to keep the money is inherently inappropriate. These sorts of fines ought to go to the government (in its capacity as guardian of public peace).

Second, allowing injured parties to keep punitive awards increases the incentive for frivolous suits.

Note that while the British "loser pays" system makes frivolous suits less likely (as it makes meritorious suits more likely), neither the British system nor the American system can bar frivolous suits. The American plaintiff will bring a weak case if his potential judgment is large relative to his own attorney's fees; the British plaintiff will bring a case just as weak if his potential judgment is large relative to total attorney's fees. Anything that increases the size of the potential judgment—such as the availability of punitive damages—will increase the probability of any lawsuit, including a frivolous one. Conversely, anything that decreases the size of the potential judgment—such as removing the availability of punitive damages—will decrease the probability of any lawsuit, including a meritorious one.

But a "loser pays" system is the best way to encourage meritorious suits and discourage frivolous suits. If such a system is adopted, the problem of poor plaintiffs not having the means to sue rich corporations should become a nonissue. The problem of frivolous environmental suits, however, is especially acute in a world where standing, mootness, ripeness—and other doctrines that once kept frivolous suits out of court entirely—have been significantly eroded.

3. Punishments Would Be More Predictable

At present, says Olson, "most courts give juries exceedingly wide latitude in calculating the amount of the award." The result is that:

Punitive damages are assessed and spontaneously imposed by juries randomly selected for a single case with little more guidance than a vague and general admonition to consider the primary goal of deterrence and the secondary goal of retribution. But juries cannot be expected to evaluate punishment levels based upon consideration of society's priorities and standards the way that legislatures do when they establish criminal penalties. Juries lack the information, training and time to do anything like that. Thus, individual juries establish their own punishment regime, applicable only to a single case, but often with far-reaching implications for society, without the tools, information, competence, experience or leavening features of the legislative process.

Supporters of punitive damages sometimes argue that the unpredictability of punitive damages is, in fact, their strength; anything predictable would simply be considered a "cost of doing business." But costs of doing business have a deterrent effect, just like any other costs, and have the added advantage that outside observers can estimate what that deterrent effect will be for any corporation. In addition, overly broad jury discretion raises due process concerns. Fairness requires that potential offenders be put on notice as to what the punishment will be for their misdeeds, if they should commit them. Potential offenders already know their potential exposure to compensatory damages; it's the amount of actual damage they cause. But no one can predict what a jury will award in punitive damages; punitive damages awards tend to vary greatly. As Judge Lee Sarokin commented on punitive damages, "No statute would be permitted which failed to set the maximum possible penalty faced by a defendant. Although the penalty imposed in a civil matter may far exceed that provided for under a criminal statute for the same conduct, none of the same safeguards are provided." If every jury is different, the uncertainty undercuts this ideal.

And not all courts follow the general rule of thumb of limiting punitive damages awards to cases of intentional infliction of harm or recklessness. "Penalties are regularly assessed pursuant to constantly evolving common-law tort theories that arise ad hoc from court decisions, often after the conduct for which the punishment is imposed. The standards of liability not only are vague, highly subjective and retroactive, but vary widely from state to state, ranging from `malice' at one end of the spectrum to `rudeness' at the other with everything from `recklessness' to `negligence' in between," according to Olson.

If, instead, recklessness and intent to harm were made into criminal violations, the statute that established these violations would say what the punishment was going to be. And if plaintiffs in cases of negligence or accidents were limited to compensatory damages, then potential defendants would know that their liability would be limited to the harm that they caused.

4. The Problem of Multiple Punitive Damages Would Be Alleviated

Even if one accepted the concept of assessing punitive damages in civil cases, the practice of awarding punitive damages to several plaintiffs for the same cause of action is inappropriate. In product-liability cases, companies can be sued by hundreds or thousands of separate people, and uncoordinated juries may award punitive damages in each of the cases. Any individual award may be reasonable, but if so, then a hundred such awards combined clearly aren't. The cumulative effects of these awards aren't reviewable by a court, since each award comes from a separate case. Moreover, multiple punitive damage awards can bankrupt a company or otherwise deplete the assets that would otherwise go to pay future plaintiffs' compensatory damages. "If there is a limited fund, priority should be given to compensating those who have been injured rather than conferring windfalls on those who have already been compensated," according to Judge Sarokin. Or, in the words of Judge Friendly,

The legal difficulties engendered by claims for punitive damages on the part of hundreds of plaintiffs are staggering.... We have the gravest difficulty in perceiving how claims for punitive damages in such a multiplicity of actions throughout the nation can be so administered as to avoid overkill. [One could instruct] the jury that it "may consider the potentially wide effect of the actions of the corporation and, on the other hand..., the potential number of actions similar to this one to which that wide effect may render the defendant subject." Yet it is hard to see what even the most intelligent jury would do with this, being inherently unable to know what punitive damages, if any, other juries in other states may award other plaintiffs in actions yet untried.

Judge Friendly went on to imagine a system in which the jury is allowed to assess one punitive award to be held and appropriately distributed among all successful plaintiffs, but conceded that "even as to this the difficulties are apparent." Most plaintiffs in mass toxic tort actions will sue separately, independently of one another, and no one knows ahead of time how many plaintiffs will exist. Multiple claims could be consolidated into one larger lawsuit, but that can generally only be done under limited circumstances, and it wouldn't preclude other plaintiffs from coming along later. In reality, then, a defendant will probably have to deal with many separate plaintiffs, each with its own jury which will decide how much to assess in punitive damages.

Some manufacturers have claimed that multiple awards of punitive damages violate their fundamental due process and double jeopardy rights, but these arguments have mostly failed in court; such guarantees generally protect people against abuses by government, and are not applied to problems of private litigation. The litigation against the pharmaceutical manufacturer Richardson-Merrell illustrates the problem; after fraudulent behavior was discovered that led to the marketing of the hazardous drug MER-29, over 1,500 personal injury actions were brought against the manufacturer. Only three cases actually resulted in awards of punitive damages, but some courts recognized that awarding punitive damages to multiple plaintiffs for the same cause of action could cause the bankruptcy of the company.

Whether punitive damages can lead to bankruptcy is, in itself, an irrelevant question. If I, whose net worth is less than $1 million, save $1 million by engaging in reprehensible conduct, it may be appropriate for plaintiffs to drive me into bankruptcy. Indeed, since I can't pay the full $1 million, such punishment may not be harsh enough. The real problem occurs when defendants are being punished more than once for a single cause of action, which, on its face, is inappropriate.

If we were to deal with this situation using the criminal law, on the other hand, all plaintiffs would be adequately compensated with compensatory damages, while a separate, unique criminal trial for the separate offense of inflicting harm intentionally or recklessly would decide the question of blameworthiness and inflict one punishment. (How to deal with the problem if punitive damages continue to be awarded by juries in civil cases is addressed later in this paper.)

5. Other Protections

These aren't the only protections that criminal law offers. Under criminal law, thanks to the privilege against self-incrimination, defendants have slightly more protection against having to give evidence that would lead to the imposition of punishment—though this may be of little benefit, as the privilege against self-incrimination generally doesn't apply to corporations (which are the most affected by punitive damages), and in any case doesn't protect written documents. Also, under criminal law, it is more difficult for the government to amend its accusation in mid-trial than it is for a private plaintiff to amend his pleading in a civil case—though here, too, the differences may well be slight in practice. For a fuller discussion of these, the interested reader is referred to Walter Olson's discussion in The Litigation Explosion.

Perhaps more importantly, under criminal law, decisions to prosecute rest with public authorities; these authorities are vested with the task of punishing criminal conduct, and endowed with the discretion to distinguish between violations they believe are worth prosecuting and those that are not. Judgments of social blameworthiness are reflections of a certain societal morality, which should be enforced not at the whim of individuals with their own views of who should and shouldn't be prosecuted, but at the discretion of representatives of the community. The threat of private "vigilantism" is already especially acute in a world where plaintiffs do not need to be harmed to bring suit; in a previous day, because of the doctrines of standing, ripeness, and mootness, such claims generally wouldn't even see the inside of a courtroom. When, through citizen suits, plaintiffs can bring environmental cases under actual environmental laws without having standing (i.e., actually being harmed), and where courts read unclear laws broadly enough to create legal requirements where none existed previously, allowing punishment to proceed at the whim of individual plaintiffs would mean giving important law enforcement functions to courts, who must hear the cases that plaintiffs bring. This, in turn, would lead to a usurpation of executive authority by the judiciary.

Over the centuries of its evolution, criminal law has come to embody a host of protections for innocent defendants. These protections are appropriate for situations—like cases where punitive damages are being sought—where people are being punished. These rules would make it more difficult to impose a punitive damages award, but that's not why the protections are valuable. Rather, they are valuable because of the serious nature of punishment. First, since mistakenly punishing the innocent is considered much more serious than mistakenly not punishing the guilty, punishment should embody a great deal of defendant protection. Second, the innocent as well as the guilty have legitimate privacy interests and deserve to know what they're being accused of and how much they stand to lose if convicted—all the more so since defendants should be presumed innocent until proven guilty. And third, even the guilty should be punished fairly and predictably. 

IV. CASE STUDIES: Dos and Don'ts

Even though this paper argues that punishment should be reserved for the criminal law, the same general principles set forth earlier in the paper still apply if punitive damages continue to be awarded by courts. Intent to harm or recklessness merits punishment; mere negligence or accidental harm doesn't.

Later in this paper, I will argue that if punitive damages continue to be awarded in civil cases, their purpose should be to deter and they should bear some relation to the amount of money the defendant saved (if any) by his behavior.

These principles allow us to evaluate specific punitive damages awards to determine whether or not they were justified. The following examples illustrate cases where defendants were rightly punished (the cases of asbestos manufacturers and Ramsey Associates), where defendants were wrongly punished (the case of the Exxon Valdez), and where defendants were rightly not punished (the case of Ashland Oil and the Love Canal case involving Hooker Chemical and Occidental Petroleum). Keep in mind that punishment (is the defendant morally blameworthy?) must be distinguished from concepts such as liability (should the defendant pay?): a firm may rightly be required to pay compensatory damages, but also not be subject to punitive damages or other punishment.

A. Those Who Were Rightly Punished

1. Asbestos Manufacturers

Who should get punitive sanctions? One example in the environmental arena is the set of failure-to-warn product-liability cases involving the Johns-Manville Corporation and other manufacturers and distributors of asbestos products. In general, when a manufacturer doesn't warn consumers about the dangers of its product, it will appropriately have to pay damages to injured plaintiffs. In the asbestos cases, plaintiffs have alleged, some successfully, that Johns-Manville knew about the dangers of inhaling airborne asbestos fibers in 1947 and didn't tell anyone. This element of intentional failure to warn brings the asbestos cases into the realm of justified punitive liability.

Kenneth Smith, Johns-Manville's medical director, had researched and written on asbestos inhalation in the late 1940s and early 1950s, and in the early 1950s recommended to high-ranking Johns-Manville corporate officials that appropriate warnings be placed on all asbestos containers. Smith's repeated recommendations were ignored, and no warnings were given until 1964. One court concluded that this conduct justified a punitive damages award because "Johns-Manville engaged in outrageous conduct by exhibiting a reckless indifference to the health and well-being of plaintiff."

Asbestos manufacturers have been punished countless times for the same deceptive conduct. Since each punitive damage award tried to be a complete punishment, the aggregate amount of punitive damages may have been vastly in excess of the reasonable amount necessary for deterrence. Of course, we don't know how harshly they would have been punished if they had been prosecuted criminally. Their punishment, which would have been determined by statute, may have been greater than the sum of the punitive damages that they've actually had to pay; they may even have had to serve time in prison. And we don't know how many times they would have been punished; their offense could be made a crime by each of the 50 states, as well as the federal government, and the same broad cause of action could be defined as, say, two different crimes. If all this were the case, they could have been punished up to 102 times—twice for each state and the federal government—assuming they engaged in punishable activities in every jurisdiction. But if the facts, as alleged in the cases, are correct, then the cause of action, in general, did merit punishment.

2. Ramsey Associates

Punitive sanctions may also be appropriate in certain nuisance cases, when the nuisance is caused intentionally. For example, in Coty v. Ramsey Associates, Ramsey Associates had applied to build a motel on some land that they owned, but because of opposition by its neighbor Coty, local authorities had approved a much smaller motel than Ramsey Associates wanted to build. As a result, Ramsey Associates applied for a zoning permit to operate a pig farm. Gerald Boston describes the case:

Defendant [Ramsey Associates] fenced in the area and then dumped rusty storage tanks, followed by sixteen truckloads of chicken manure directly across from plaintiffs' properties. One of the drivers of the manure-carrying trucks told the police that defendants had finally "gotten even" with plaintiffs. While defendants contended that the manure was for crop fertilizer, an expert at trial opined that it was so greatly in excess of recommended amounts that it would destroy any attempted growth. Following this, defendants brought in ten junked automobiles and one hundred pigs and cows. The animals were fed at the location closest to the plaintiffs' homes and during the winter died of starvation and their decomposing carcasses were left lying there with accompanying stench and later infestation of flies. The trial court found that the defendants had used the farm as a pretext to abuse and kill animals which had no purpose other than to intentionally annoy and harass plaintiffs and cause them economic injury.

The court said that the defendants had "no purpose" other than annoyance in mind—because the defendants neither made nor tried to make money from their actions. The court awarded a total of $380,000 in punitive damages to plaintiffs. The defendants, on appeal, held that they were legally using their property, and even an admittedly "improper motive" didn't change that. But legality is no defense to the creator of a nuisance. The Vermont Supreme Court held that "where a defendant has acted solely out of malice or spite, such conduct is indefensible on social utility grounds and nuisance liability attaches."

B. Those Who Were Wrongly Punished: Exxon and the Valdez

Who should be exempt from paying punitive damages? Let's consider the case of Exxon. The punitive damages verdict in the Exxon case was out of line with punitive damages verdicts in general. It is larger than any award of punitive damages ever made by any jury, and five times larger than the largest award affirmed by an appellate court (the $1-billion award in Texaco v. Pennzoil). Apart from Texaco, no award greater than $25 million had survived appellate review before the Exxon award, and the Exxon award is 200 times larger than $25 million. Even in proportion to Exxon's wealth, the award is aberrantly large; this award was 14 percent of Exxon's net worth, and punitive damages are almost always below 1 percent of net worth.

Of course, large doesn't mean wrong. Maybe the other awards were too small, or the Exxon case could have had its own peculiar characteristics. But we should keep in mind that most cases where large punitive damage awards have been upheld are cases of intentional tortious conduct where the defendant profited by his conduct, and all such cases have involved companies that weren't already punished by a criminal fine.

These circumstances are absent here.

First, Exxon's punitive damage award was levied on account of recklessness (not removing a drunk captain), not actual intent to harm, which was present in most of the other cases.

Second, if we want punitive damages to deter, we should look to how much defendants benefited by their behavior as a guide to how much it'll take to prevent them from doing it again. But what if a defendant didn't benefit at all? In that case, the losses the defendant has already sustained may be adequate to prevent him from repeating the offense. "Ladies and gentlemen," Exxon trial lawyer Jim Neal told the jury, "we didn't build [a] $130-million state-of-the-art tanker, put $16 million worth of crude oil on it, and recklessly and callously turn it over to a drunk. We can make mistakes, but we ain't that stupid." Whether Exxon was "that stupid" is difficult to determine; what is clear, though, is that Exxon lost a lot of money, even without the punitive damages.

•The jury was instructed that punitive damages couldn't be awarded in an amount greater than that necessary to punish and deter. But the goal of punishment may already have been served through criminal proceedings, not even counting the substantial damage to Exxon's reputation that resulted from the spill. Exxon had already been prosecuted criminally in a separate case, and its criminal sanction was already the largest criminal sanction for an environmental crime in history—$150 million, of which $125 million was remitted because Exxon accepted responsibility for the spill and the cleanup. (Exxon also paid $100 million in restitution to state and federal governments as part of the criminal sentence.) The court found that this sentence contained "an appropriate amount of punishment," and contained "an appropriate element of encouragement of respect for the law."

One justification for punitive damages, therefore—that they are necessary to punish behavior not covered by criminal law—is absent in the Exxon Valdez case.

•The goal of specific deterrence (deterring Exxon) was served by Exxon's criminal fines, civil liabilities, and remedial expenses, which are estimated to have come to $4 billion, the largest single cost in Exxon's history.

Besides the $25-million criminal fine, this includes $100 million in restitution to state and federal governments imposed as part of the criminal sentence, $900 million in settlement of government claims for natural resource damages (with the possibility of a $100-million reopener), $2.1 billion in cleanup expenditures through 1992, $304 million in settlement of private damage claims paid through the Exxon claims program, $22.5 million in settlement of native subsistence claims, $287 million in compensatory damages awarded to fishermen, $9.7 million in compensatory damages awarded to the Native corporation and municipality plaintiffs in a state court trial, $46 million in casualty losses for the vessel and cargo, and potential additional liability of up to $309 million. Exxon also changed its policies after the spill to greatly reduce the probability of the spill happening again—though this, in itself, doesn't resolve whether punitive damages are necessary, since Exxon may have changed its policies under the expectation that it would have to later pay punitive damages.

So did Exxon save any money by its misdeeds? They lost a great deal of money in lost oil, a lost ship, remedial expenses, civil liabilities, and criminal fines. These amounts clearly overwhelm any possible savings from not taking enough care in removing alcoholic captains from duty. In fact, when plaintiffs claimed in the trial that Exxon wanted to save the small amount of money that would have been required to remove tanker captains with alcohol histories from duty, they were inadvertently supporting the position that Exxon saved very little and was therefore adequately deterred by its previous $4 billion payments. The court, too, in denying Exxon's post-trial motion to set aside the punitive damage award, argued that "the evidence established that with relatively small expense, when compared to the enormous risk, Exxon could have [e]nsured that its supertanker crews were rested and not captained by relapsed alcohol abusers."

•The goal of general deterrence (deterring other companies from doing what Exxon did) may also already have been served. Other oil companies, like Exxon, changed their policies after the spill. This fact alone doesn't tell us whether they would have been deterred absent the prospect of punitive damages. But it's likely that this is in fact the case; $4 billion is greater than the net worth of most American corporations and would have bankrupted Arco, which is about a tenth the size of Exxon. The federal Oil Pollution Act of 1990, which was enacted in direct response to the Valdez oil spill, increased tanker operators' liability from what it was before the spill. So even without any punitive damages, other companies are already deterred more than they would have been without the oil spill.

C. Those Who Were Rightly Not Punished

1. Ashland Oil

Another group of people who shouldn't have to pay punitive damages are those who, while they may have been negligent, didn't cross the line into recklessness, intent to do harm, or intent to violate the law.

In Arnoldt v. Ashland Oil, residents of an area near Ashland Oil's refinery in Catlettsburg, Kentucky, brought a private nuisance action against Ashland, alleging that the refinery's air emissions interfered with the use and enjoyment of their property. The circuit court awarded the plaintiffs $10.3 million, to be divided among four plaintiffs, selected at random from a list of more than 200 plaintiffs. Compensatory damages totaled $1.3 million and punitive damages totaled $9 million.

Under Kentucky law, punitive damages could be awarded "only upon proving, by clear and convincing evidence, that the defendant from whom such damages are sought acted toward the plaintiff with oppression, fraud, or malice," where malice, under the relevant Kentucky statute, was defined as "either conduct which is specifically intended by the defendant to cause tangible or intangible injury to the plaintiff or conduct that is carried out by the defendant both with a flagrant indifference to the rights of the plaintiff and with a subjective awareness that such conduct will result in human death or bodily harm." In other words, "malice" means, roughly, recklessness or intent to harm.

The appellate court found that the trial court had misstated the burden of proof required for punitive damages, sometimes referring to "clear and convincing evidence," sometimes "a preponderance of the evidence," and sometimes "a clear preponderance of the evidence." (In fact, "clear and convincing evidence" is more stringent than "a preponderance of the evidence," but less than "beyond a reasonable doubt.") The court overturned the punitive damage award on the grounds that even though the damage may have been great, Ashland's conduct wasn't "sufficient to evidence conscious wrongdoing." The original trial court had allowed the jury to consider the question of punitive damages without evidence of conscious wrongdoing. "Because appellees," the appellate court concluded, "did not introduce evidence which demonstrated a specific intent to cause bodily harm or injury, they likewise failed to demonstrate fraud or oppression toward appellees."

2. Hooker Chemical and Love Canal

Yet another group of people who shouldn't have to pay punitive damages are those who, on the whole, acted responsibly—like Hooker Chemical, the company associated with the infamous Love Canal. From 1942 to 1953, Hooker placed 21,000 tons of chemical waste in the abandoned canal. In 1953, the land was going to be condemned and seized under eminent domain for use by the local school board. In anticipation of the condemnation, the school board was planning to build its school two years before Hooker deeded it the land. According to Hooker, while it could have let the school board condemn its land and pay compensation, it instead sold the property for $1 so that it could insert warning language into the deed. A year later, an elementary school was built there and a neighborhood grew up in the area.

In the 1970s, toxic chemical-laden groundwater began seeping into neighborhood yards and basements; state health officials declared an emergency at Love Canal in 1978 and relocated about 2,500 residents. Love Canal became a symbol of the evils of industrial pollution and corporate irresponsibility.

Hooker executives were keenly aware of the dangers of the chemicals at Love Canal. In 1945, Hooker analyst R.H. Van Horne wrote in a memo, "Eventually we will have a quagmire at Love Canal which will be a potential source of lawsuits in the future." A year later, Hooker attorney Ansley Wilcox II expressed concerns about "contaminated water" in the canal, which children were using as a swimming hole. Wilcox suggested that a fence be built around the area, but no fence was built. Plaintiffs' lawyers alleged that Hooker Chemical had been reckless in a number of ways:

•knowingly dumping toxic chemicals in an area used for recreation by children;

•failing to fence in the swimming area or to institute other warning procedures;

•abandoning the property and its buried chemicals while knowing that it was becoming an increasingly popular neighborhood for families;

•giving "insufficient information" about the hazards to the school board before transferring the dump in 1953; and

•failing to assume responsibility for the dump after the health dangers became known in the late 1970s.

Some of Hooker Chemical's actions, such as not building the fence, may have been irresponsible; other actions, such as not taking responsibility twenty years after transfering the property, are more debatable. There are good reasons, though, to believe that on the whole, Hooker Chemical acted responsibly in the Love Canal affair:

•A private engineering firm hired by the city of Niagara Falls in 1979 to evaluate the Love Canal dumpsite concluded that Hooker's practices met and exceeded the relatively stringent standards of the Resource Conservation and Recovery Act, which wasn't enacted until 1977.

•Hooker took special care to sell the land for $1 instead of just letting the school board take its property. The entire purpose of this move was to put warning language in the deed.

•The last paragraph in the deed of Love Canal to the school board advises the school board that the property contains "waste products resulting from the manufacturing of chemicals," and warns the school board that it assumes "all risk and liability incident to the use thereof." An attached condition demands that as a condition of the property transfer, "no claim, suit, action or demand of any nature whatsoever shall ever be made" against Hooker, "including death... or loss of or damage to property caused by, in connection with or by reason of the presence of said industrial wastes."

•Hooker wanted to require that the property "be used for park purposes only, in conjunction with a school building to be constructed upon premises in proximity to" them. And it wanted the board to agree that, should the property ever stop being used as a park, title to it would revert to Hooker. The school board refused to accept these restrictions, and Hooker had to settle for the warning language in the deed.

•In 1957, Hooker's attorney, Arthur Chambers, reminded the board that chemicals were buried under the surface of the land, explaining that this "made the land unsuitable for construction in which basements, water lines, sewers and such underground facilities would be necessary." Chambers conceded that his company "could not prevent the Board from selling the land or from doing anything they wanted to with it," but stated that he felt the property shouldn't be divided for the purpose of building homes and hoped that no one would be injured.

The Love Canal case was, in the end, correctly decided. The New York state attorney general's office was seeking a punitive damage award of $250 million. But on March 17, 1994, U.S. District Judge John T. Curtin ruled that Occidental Chemical Corp., which took over Hooker Chemical in 1968, should pay no punitive damages. The court found Hooker negligent on a number of occasions, and criticized Hooker for turning its property over to the Niagara Falls Board of Education—"but," Curtin wrote, "a finding of outrageous conduct and reckless disregard of the safety of others requires more. And the conduct must be judged by the law in force at the time. Occidental argued that its actions had to be judged from the context of industry practice of the time. While the advances in science and engineering made since the 1940s would lead contemporary environmentalists to condemn many of the practices used by Hooker at Love Canal, it would be unfair to judge the company by the application of knowledge obtained after the disposal and transfer were completed."

Plaintiffs in the Love Canal case were, understandably, distraught that Occidental paid no punitive damages. "I'm devastated," said Joann Hale, a former resident of the Love Canal neighborhood. "I can't believe this. Was the judge watching the same trial I watched? I lived at the Love Canal. They made the chemicals that seeped into our backyards. That company made millions and billions of dollars. Who is going to be held responsible? [...] My husband and I both have bone tumors. Lisa was conceived and born when we were living at Love Canal. Her teeth decalcified when she was about three years old. All her teeth had to be either removed or recapped. Now she has problems with her bones, she's broken seven or eight different bones. Carrie had an eye tumor and now, she has arthritis. Are all these things related to living near the Love Canal? I can't say. But a husband and wife, both getting bone tumors. What do you think?"

Hale's comments neatly encapsulate much of the case against Hooker Chemical and Occidental Chemical—the company caused contamination and medical problems, it made a lot of money, it should be held responsible. But none of these issues speaks to the question of intent to harm or recklessness. Whether it caused contamination and medical problems is a factual matter to be resolved by courts, that it made a lot of money may well be true, and that it should be held responsible for the damage it caused (through compensatory damages) is undeniable. At the time of the punitive damages decision, the court had already found Occidental at least partly liable for cleanup costs, which Occidental's attorneys estimated at $325 million. The punitive damages award (or lack thereof) has no effect on a pending state case involving medical claims or a previous settlement in 1985 (in which over 1,300 residents received $20 million). As Richard J. Lippes, an attorney for the Love Canal Homeowners Association (which is bringing the state case for medical claims), put it, "As far as the homeowners' case goes, I think Judge Curtin's decision actually helps us. We're trying to prove negligence by Hooker, and Judge Curtin states again and again that he feels Hooker was negligent. Occidental is trying to portray this decision as them winning the whole ball game. But in the big picture, that isn't true. Judge Curtin's decision focused on a very narrow issue, punitive damages."

The comments of Lois Gibbs, the housewife-turned-activist who led Love Canal homeowners' protests in the 1970s, similarly miss the point. "[The decision] sends that same old message to corporate America. If you want to do something, like pollute the environment, just look at the cost tables and see if it is going to be worth the profit you make. You'll wind up paying medical costs and cleanup fees, but no punitive damages. How much is an arm worth, or how much is somebody's life worth? Figure out the cost tables and see if it's worth it. It sets a double standard. If I went into somebody's house and destroyed it, I would be thrown in jail. If it's a corporation, it's O.K." If Lois Gibbs destroyed someone's house, she would be acting intentionally, and would rightly deserve to be thrown in jail. But not even the plaintiffs in the Love Canal case claimed that Hooker Chemical intended to contaminate the neighborhood and cause health problems. 

V. OTHERWISE, MORE DUE PROCESS PROTECTIONS

Treating intentional infliction of harm or recklessness as criminal violations, and leaving the tort system for compensation, is a desirable outcome. However, the doctrine of punitive damages may not be reformed in the near future, at least not in the ways proposed here. Punitive damages have been around long enough, and have a large enough constituency, that they will probably have some staying power. Still, if punitive damages remain within tort law, there are various steps we can take to make the process more sensible:

•Juries have been known to be under tremendous pressure to act sympathetically toward the plaintiff who is part of their community. But juries, in themselves, are not the problem. They can be smart and responsible, if given the chance. The basic problem is unlimited jury discretion, which can result from insufficient jury guidance.

•Punitive damages reform must involve at least the procedural safeguards mandated by the Supreme Court in the Haslip case: clear jury instructions, post-verdict review by the trial court, and appellate review.

•In addition, even if cases of recklessness or intentional infliction of harm aren't made criminal, certain levels of protection for the defendant should still apply. The burden of proof for awarding punitive damages should be higher than for other civil penalties, the plaintiffs shouldn't receive punitive damages proceeds, punishments should be more predictable, and multiple awards of punitive damages should be curtailed.

A. Should Juries Be in this Business Anyway?

Some have suggested that punitive damages would be applied in a more principled manner if the judge, not the jury, determined the award. (This, in fact, was done in the Love Canal case, which was heard without a jury.) The jury would remain the trier of fact—in this case, they would still be making the decision of whether the company was liable at all for punitive damages. This would parallel the current criminal practice in most states, in which the sentence is imposed by the trial judge after the jury has convicted. Judge-determined punitive damage awards are said to have the following advantages:

•There would be less chance that passion or prejudice would result in an inflated verdict or in overkill.

•The judge has greater experience in criminal proceedings involving punishment, is potentially more aware of the social policies and economics involved in meting out punitive awards, and is more likely to achieve consistency in the magnitude of punitive awards from one case to the next.

•The judge might review evidence which wouldn't normally be admissible because of its prejudicial effect.

This case for the determination of punitive damages awards by judges is interesting, but not decisive.

Juries can be impassioned, prejudiced, or swayed by ideological or political opinion, but judges can be, too. (In a world where judges are political appointees, it is naive to expect that they would be without passion or prejudice, or that they wouldn't owe their positions to particular passions or prejudices.) Part of the reason that juries are drawn from the population at large is that the passions and prejudices of a jury are more likely to be representative of the passions and prejudices of society as a whole than are the passions and prejudices of individual judges.

Social policies and economics are indeed important in meting out punitive awards. If punitive awards were restricted to criminal law, they would be determined democratically, through criminal statutes. In the absence of a statute, though—if the democratic process provides no input on how to set an award—or if there is no other form of guidance, the judge is as much in the dark as the jury. Again, the judge will likely set a punitive award according to his own passions and prejudices, which may be less democratically acceptable than the passions and prejudices of a jury. As Judge Easterbrook put it in Zazú Designs v. L'Oréal, "A judge sanctioning misconduct may not draw a number from the Æther but must explain the choice by reference to its role in compensating the wronged party or deterring conduct that injures the judicial system." (One thing the judge could do is set an award that's consistent with similar awards elsewhere, but the Supreme Court already requires post-verdict review by the trial court, as well as appellate review, for precisely that reason.)

And it's unclear whether we want a jud