Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
The Bhopalization of American Tort Law PETER W. HUBER The new tort law, and the entire vision of "public law" adjuil'- canon that animates it, is politically dangerous. It removes the most controversial social issues from the arena in which controversy can be brokered and tensions diffused through Me very process of po- , . . . . . - lz~lcal partzczpanon. Since 1960, American tort law has undergone a revolution that can be accurately (if somewhat disrespectfully) described as a mad scramble by the Trial Lawyers of America to keep pace win He National Academies of Engineering and Sciences. New legal arguments have been prepared, tested, honed, and developed on a wide variety of accidents, nonaccidents, illnesses, and cures. Bendectin, Agent Orange, Love Canal, Three Mile Island, the Dalkon Shield, and the pertussis vaccine have all figured in the training. The transformation, in a nutshell, has consisted of shifting the focus of American tort law from "pnvate" risks to "public" ones. And the prime mover has been science: the radical chances in tort law are traceable to equally radical improvements in the science of hazard identification and risk assessment. But not science alone, because the courts have often not been successful in recognizing where science ends and other bodies of learning or ignorance begin. Thus, trans-sc~ence,~ conscience, and a generous dose This paper was originally presented at a National Academy of Engineenng symposium. ~&Hazards: Technology and Fairness'' (Washington, D.C . June 3-4. 1985), and will also appear in the volume based on that symposium (forthcoming from the National Academy Press, 19863. 191
192 PETER W. HUBER of sheer nonsense have also figured prominently in the tort law's expansion from the narrow world of private Hazards to the universe of public ones. But though the tort system's reach has been greatly extended, the system has yet to demonstrate that with public risks it offers useful deterrence, fair compensation, or a needed vehicle for resolving festering social disputes. Legislatures are increasingly looking outside the tort system for mechanisms to accomplish these important objectives more effectively. TORT LAW, OLD AND NEW "Public" risks are perhaps not self-defining. When I use the teen, I mean risks that are centrally produced or mass-produced, broadly distributed, often temporally remote, and largely outside the individual risk bearer's direct understanding and control the hazards of large-scale electric power plants, air transport in jumbo jets, mass-produced vaccines, chemical additives in food, hazardous wastes, and recombinant DNA technology. These risks gen- erally have a zero-infinity charactereither the likelihood of harm to any individual is minuscule but the likelihood of exposure is very large (as with pollution), or the likelihood of exposure is small but the likelihood of wide- spread harTn (should exposure occur) is large (as with a nuclear power plant accident). "Private risks," by contrast, tend to fall in more familiar ranges of the probability-consequence spectrum. They are discretely produced, lo- calized, personally controlled, or of natural origin, and often immediate in their effects the risks of cottage industries, wood~stoves, transportation by car, and the like. The "old" tort law revolved around private hazards and thus touched a relatively narrow range of human activities. It focused on correcting clear injustices and resolving festering disputes. And it maintained this focus by hewing strictly to two general requirements: a tort dispute had to be bipolar, and it had to be timely. These two overarching principles gave birth to a multitude of subsidiary rules and requirements. In the past 25 years, most of these have been either eroded beyond recognition or entirely discarded. Bipolarity Old tort law insisted, first of all, that a dispute be bipolar. Of course, a lawsuit cannot proceed with fewer Man two parties. But under the old tort law you generally could not have more than two either. There were a few well-established exceptions to cover third-party claimants, joint tortfeasors, and so on. But the general rules strongly disfavored crowded courtrooms. In addition, under the old tort law the parties had to have character. The plaintiff had to be someone special someone specially injured. Under *he old law of "nuisance," for example, the private individual simply was not
THE BHOPAI~ZaATION OF A~IERICAlI TORT LAW 193 permit-ted to sue for anything, in the nature of a "public" nuisance. If the hazard was diffuse, broadly shared by the community, then it was by defi- nition "public," and no private individual could sue. To bring a private lawsuit you had to prove a special injury to you personally- something out of the ordinary, an injury to your land, or to other private interest, an injury not largely shared by the community as a whole. If everyone in the community suffered more or less equally, the presumption was that no one suffered at all unless a government official (typically a public prosecutor) determined otherwise. Under the old tort law the defendant had to have character, too. His distinction was that he had to be the identified, demonstrated source of the plaintiff s injury. The plaintiff was required to show that this defendant more likely than not caused that particular injury. Just as the plaintiff had to distinguish himself from the grey mass of suffering humanity, so the defen- dant had to be distinguished from the mass of humanity's oppressors. If there was only a 1 percent chance that I caused your cancer, you could not sue me successfully. These rules limited the reach of tort law very considerably. They confined tort litigation to the resolution of narrow, bipolar disputes. The old tort system was quite comfortable with car accidents but not with such things as pollution. Times have changed. The "new" tort law is perfectly happy to take on multiparty disputes of every description. Plaintiffs' class actions, to start with, have become routine. These allow plaintiffs to consolidate numerous small injuries into one large one. It is not uncommon for class actions to embrace tens of thousands, and sometimes millions, of plaintiffs. An as- bestos-related insurance dispute in California has been conducted in a large college auditorium the space is needed simply to provide room for tables for the myriad lawyers involved. The evolution of the enormous lawsuit can be gauged, somewhat indirectly, by the large body of solemn, scholarly, and judicial literature addressing what obligations (if any) a lawyer initiating such litigation may have to notify his "clients" of the very existence of "their" lawsuit. And as the recent Agent Orange settlement is now demonstrating, distributing the damage awards that may be won in such suits can require administrative facilities and technical expertise comparable to those typically available only to a large city or state government. Things have become equally crowded on the defendants' side of the au- ditorium. In one recent settlement, more than 150 companies were involved in a tentative agreement to clean up a toxic waste dump. The California litigation involving the drug commonly called DES introduced the radical concept of "market share" liability, which allows plaintiffs to sue undif- ferenuated groups of defendants without ever working out exactly which defendant caused whose injury. Under this rule of law, defendants are simply held liable in proportion to their share of the market in the drug or over
194 PETER W. HUBER product at the time the harm occurred. A similar concept is codified in the Price-Anderson Act (42 U.S.C. § 2210(b)) (and may be significantly ex- panded when the act comes up for renewal next year)all nuclear operators win share in the liability if there is an accident at any one of Weir plants. The most radical change in the new tort law is one skill gleaming in the eyes of various legal commentators but a moderately serious prospect none- theless. Powerful forces are pushing, often to receptive audiences in the courts and legislatures, for what has been called a "proportional causation" rule of liability. This would permit me to sue you for my cancer if ~ could show even 1 percent likelihood that your power plant in fact caused it. Notwithstanding the 99 percent probability in your favor, ~ would still re- cover but only 1 percent of my actual damages. One percent, that is, plus or minus the usual slack allowed by Me tort system which seems to be several hundred percent, at least. Under the old tort law, by contrast, ~ would not collect at all unless ~ established a causal link between your conduct and my injury that exceeded 50 percent; once that threshold was exceeded, would recover my full damages. Timeliness The old tort law also had a rigid sense of timeliness. A tort action could not be brought too early. Nor too late. Premature actions were simply dis- m~ssed without prejudice to the playoffs right to sue again later. Stale actions were dismissed once and for all. Some of the details are worm · . revleu~mg. Tort law entertains and for a long time has entered prospective actions for injunctive relief. These are actions designed to cut off a course of dangerous conduct before it culminates ~ actual harm. Under Me old tort law, if you could really show Mat ~ was on Me brink of doing something terribly, imminently, and irreparably damaging to your interests, you could get an injunction to stop me. But if ~ was anywhere short of the brink, or if the consequences were anything short of irreparable, no injunction would issue. You (the prams had to wait for Me harm to occur; only Men could you sue me for money damages. The rationale was simple enough: what may appear to be a real threat of harm often is not, and litigation postponed will often be litigation avoided altogether. Thus, for example, in an 1885 English case, Fletcher v. Bealey,2 the court rejected an injunction plea from a downnver paper plant whose owner was worried about an upstream chemical manufacturer's wastes being piled at Me edge of the ever. There was no dispute Mat Me wastes would be injunous; Me only question was when, if ever, Hey would enter the river. The court ruled that Me harTn was not demonstrably imminent and that, therefore, no injunction should issue. This rule was widely followed In the American
THE BHOPALlZATION OF AMERICAN TORT LAW 195 courts, and the rationale for embracing the rule on this side of the Atlantic was clear enough. Article III of the Constitution restricts courts to deciding "cases or controversies"- disputes, in other words, that have reached a certain level of ripeness. And American courts of that era were willing to accept the idea that if the wrong lies too far in the future the court's order will necessarily be speculative and prone to error. Under the old tort law, the would-be plaintiff could not bypass the strict rules against premature injunctions by asking for premature money damages either. Negligence "in the air" was emphatically not actionable. Actual loss or damage was an essential element of the cause of action; in all but the most exceptional cases, the threat of future hann, as yet unrealized, was not enough. You could not, in other words, sue for damages alleged to be caused by exposure to risk itself. There were some minor exceptions to cover cases in which the exposure to the risk caused present damage to land values and so on, or in which the defendant's conduct created such acute and broadly shared fears in the community that the courts felt it appropriate to intervene. But these exceptions were narrow. A tuberculosis hospital, for example, located ire a proper place, was not an actionable nuisance. Finally, under the old tort law a plaintiff was not permitted to sue too late eider. Once he had been injured, a plaintiff typically had about one to three years to bring his damage action. This 1imitahon period applied regardless of when he discovered that he had been injured. If the discovery came after the expiration of the period, it was simply too bad for the plaintiff. In short, the old tort law saw timing as a critical factor in litigation. Except in very rare cases, the law embraced two, very limiting presumptions: first, that only time would tell if a risk was real enough to be worm a lawsuit; and, second, that time would heal all wounds without the help of the legal system. Today, time isn't what it used to be. Through one legal vehicle or another, risks are being litigated earlier and earlier. And also later and later. We have established, to start with, comprehensive administrative li- censing systems for such things as drugs, food additives, nuclear power plants, hazardous waste dumps, aircraft, new chemicals, and the like. In a series of creative interpretations of the law, judges have declared that such statutes as the Administrative Procedure Act and the National En- vironmental Policy Act grant the courts broad authority to review licensing decisions critically as soon as they are made. For example, a panel of judges decided that they had authority to block research on the fast breeder reactor not the reactor itself, mind you, but the research program going on behind it because in the very long term such research might shape energy policy and so have a "significant impact" on the environment.3 Moreover, every time a standard-setting agency such as the Environmental Protection Agency (EPA) or the Occupational Safety and [Iealth Admin-
196 . PRIER W. HUBER castration (OSHA) sets a new health-based standard, the underlying risk assessment (if any) is immediately subject to challenge in the courts, regardless of how temporally remote the harm at which the standard is directed. There is still a concept of "ripeness'' in administrative law litigation of this type, but it has grown pathetically weak. Similarly, tort plaintiffs seeking injunctions against what they view as unacceptably risky activities are no longer required to establish that hanns are imminent. There is not the slightest doubt that an American court would be prepared to issue the Fletcher v. Bealey injunction today. Judges and lawyers may still pay lip service to "imminent harm'' standards of yesteryear, but in fact almost any activity, no matter how remote the harm that it threat- ens, can qualify. The "window of opportunity" for bringing damage actions has opened even wider. Such actions can now be brought very much later than was formerly allowed, and the evolving trend is to entertain them very much earlier as well. At the front end, prominent legal commentators and judges are proposing to make the exposure to risk itself a compensable injury. In a nutshell, I could sue you for operating a nuclear power plant or a chemical factory in my neighborhood, and I could recover. How much? Well, as I understand these (in my view ludicrous) proposals, I would recover We present actuarial value of the injury you might some day cause me. We would take the Rasmussen Report on reactor safely, 4 say, multiply He 10-7 estimated risk of accident during the plant's life by the 10~ dollar value of the consequences and distribute the proceeds among the 104 who might collectively suffer those consequences. No, not quite: the recovery would be distributed among 104 plus one recipients The lawyer who brought this inspired action would surely get a healthy cut, too. If this seems utterly fanciful, the same scheme can be dressed up in different clothes, and there are many in my learned profession who are eager to serve as the tailors. Plaintiffs, it is argued, should be allowed to recover for the anxiety that exposure to risk can cause. In practical terms, this has much the same effect as suing for the risk itself the suit can be brought at any convenient time, long before the risk is translated into actual injury, indeed, long before anyone is sure whether the risk is even real. And this highlights the one big advantage of suing, for anxiety instead of for the risk itself: anxiety levels- unlike risk levels are within the exclusive control of the plaintiff. It has not escaped notice in the profession that clever legal work and good (i.e., bad) publicity can create anxiety where none existed previously, and where none could be rationally justified by the hazard at hand. Damage actions are being brought later and later as well. The other half of the plaintiffs' bar has been arguing (with considerable success) not that
THE BHOPAr JZATION OF AMERICAN TORT LAW 197 the injury occurs earlier than had been previously thought, but that it occurs much later. Thus, an increasing number of jurisdictions have adopted "dis- covery" statutes of limitation. These start the limitation clock running when the plaintiff discovers his injury, rather than when he was actually injured or exposed to the toxic agent. A variety of other new legal doctrines based, for example, on the defendant's alleged concealment of information relevant to the hazardhave been successfully developed to extend by decades the period during which damage actions may be filed. The World in the Oyster It is these changes that have permitted courts to move from resolving "private" disputes to regulating "public" risks. Tort law's invasion of the enormous "public nsk" territory permits almost everyone to sue almost everyone any time and makes the chances of winnin;, something, somewhere, better and better. The new tort system, to start with, can sweep into any particular risk controversy large fractions of a community, or the population of a state, or even the population of the country. With asbestos, Agent Orange, the Dalkon Shield, the whooping cough vaccine, and so on, tort lawyers can deal directly with hundreds of defendants and millions of plaintiffs; indirectly, they can affect millions more. If, for example, the vaccine industry folds under the current legal onslaught, the largest consequence will be felt by tomo~Tow's childrenindividuals who have never been injured by a vaccine and who have never chosen to go to court. The new tort law's reach is greatly extended in time as well. It concerns itself with activities dating back to the 1930s and reaching forward for one or (in controversies involving teratogens and mutagens) even several generations. The Dr'~ir~g Force The enormous expansion in the reach of tort law might seem somewhat surprising, considering that the hazards of daily life have been decreasing steadily and significantly for reasons quite independent of the tort system. But the anomaly is easily explained. Tort litigation is definitely a supply- side industry. Its growth has been driven by We availability of information about hazards, not by the severity of the hazards themselves. Well-understood though comparatively trivial hazards are much more actionable Ban poorly understood but graver ones. Tort litigators' practices, in short, have expanded to embrace the science available: science's growing awareness and under- standing of diffuse, low-level hazards; its steadily improving ability to quan- tify very low probability events; its increased capacity to link old causes with
198 PETER W. HUBER new effects many decades later. This means that some excellent scientists have a lot to answer for. Bruce Ames, for one. Ames, who developed a "quick-and-dirty" lab test for identifying mutagenic chemicals, has had a particularly mischievous im- pact in He legal world. It is very convenient for a lawyer representing a plaintiff who has (or fears he one day may have) cancer to be able to label a toxin as an Ames-proven mutagen (and thus probably a carcinogen). Then the defendant's lawyer must begin "quibbling" (or so it will seem to the lay jury) about dose-response limits, nonzero thresholds, the ubiquity of all- natural carcinogens, and so on. And of course, Ames is not the only culpnt. Many other scientists who have unraveled the etiology of cancer and venous chronic illnesses and revealed the long lag times between exposure to a toxin and the onset of disease are responsible, far more than any lawyer, for the law's effective abandonment of statutes of limitation for damage actions and the "imminent hann" prerequisites for injunctions. Norman Rasmussen has been another major accessory before the fact in the creation of new tort law. Give me a scientist who is willing to put a numberany number on a risk of dying, and I will give you a plaintiffs' lawyer who is willing to wave that number in court. Then the defendant's lawyer must once again begin "quibbling" (or so it will seem) about the inherently self-negating aspects of risk assessment, the real purpose of con- ducting probabilistic risk assessment in the first place, the conservatism in the "source terTn," and so on. And again, Rasmussen alone should not shoulder all the blame. As the Nuclear Regulatory Commission (NRC) has systematically identified, analyzed, and quantified the hazards of nuclear power plants, the plaintiffs' bar has licked its collective chops in anticipation of great meals to come. CAN THE LEGAL SYSTEM COPE? Understandably enough, many lawyers recognize the legal flood precipi- tated by these scientific storms as great progress for the profession. Opening up the tort law hunting grounds (and extending the season to boot) gives lawyers and the courts enormous power to reshape the political and economic landscape. A separate question, however, is to determine who has benefited by this dramatic expansion in Be demographic and temporal reach of tort law. Notwithstanding He most pious and self-serviIlg claims of He plaintiffs' bar, and despite He stapling support that tort lawyers have received from venous "consumer protection" and "public interest" pressure groups, it increasingly appears Hat the principal and perhaps exclusive beneficiaries have been lawyers themselves, accompanied by only the tiniest group of self-selected or randomly selected consumers.
THE BHOP~TION OF WENCH TORT LAW Regressive Incentives 199 The common wisdom among those who would defend the operation of the new tort system is that while the system is a considerable inconvenience to corporate and technological Amenca, it benefits consumers and the public at large. Tort law, it is said, allocates accident costs so as to promote the efficient ordering of the free market and so as to affirm the individual's right in personal secunty. The traditional judicial assumption is that the tort system protects the public by detemng risky activities. But science has taught that risk is everywhere, in absolutely everything we do or choose not to do. So undiscriminating deterrence of risk is now useless; the tort system regulates progressively only if it picks its targets accurately. lathe new tort system does not, and this, on reflection, is hardly . surpnslug. The first and surely the largest problem is that the tort law delegates difficult decisions to simple people. There is something magnificently right in leaving questions of criminal culpability to 12 good persons and true, pulled off the street more or less at random. But there is something profoundly silly in consulting this same group about the safety of drugs or nuclear waste repro- cessing plants. To pick one recent and extreme example, the plaintiffs' bar has come close to saving Americans from the `' scourge" of vaccinations because judges and juries (unlike an almost unanimous American medical community) are still undecided about whether the Sabin polio vaccine is preferable to the Salk vaccine, or whether whole-virus pertussis vaccine is preferable to a vaccine prepared from virus extract. This ambivalence has, of course, proved enormously expensive to manufacturers who followed Food and Drug Administration (FDA) requirements and sold the "wrong" alter- nat~ve. But the more general lesson is that public-risk analysis is difficult, and lawyers, judges, and lay juries are not well qualified to decide which technologies and activities represent sound, progressive risk investments. The tort system's risk preferences are further scrambled by the fuel on which it feeds. As we have seen, the system understandably favors as its targets risks that Be scientific, parascientific, or plausibly pseudoscientific communities profess to know something about. Pseudoscience does not, of course, provide much rational ordering. And the great paradox when the tort system seeks to rely on real science a paradox that the system utterly fails to grasp is that the better the scientific information we have about a nsk, the less serious the risk is likely to be. One reason is that new technologies, which are (on average) safer Han the old ones they displace, come under much closer scrutiny in the administrative regulatory process. Another is that the more we know, the easier it is to protect against a risk in the first place. Thus, the tort system, driven primarily by the volume of risk information available, has a definite tendency to come down hardest where the deterrence it offers is least needed.
200 PETER W. HUBER Beyond this, the tort system selects its targets according to criteria that make them attractive to the average, contingency-fee lawyer. It therefore prefers risks that are lurid, that stem from especially new and (to the lay jury) unfamiliar technologies, or that engage socially and politically divisive issues and arouse strong public passions.6 Criteria such as these provide poor guidance in selecting objectively between good risks and bad ones. As a result, the new tort system brazenly fails to discriminate among the good, the bad, and the ugly in the public-risk world. The biggest losers in the tort game are those who do not play at all. The traditional judicial assumption and the only assumption on which the judicial-regulatory machinery can operateis that in generously compen- sating the injured consumer the courts also protect the safety interests of other members of the public as well. This may well be true when the court deals win focused, comparatively grave private risks that can readily be identified as antisocial, such as drunk driving or hazardous conditions on land. But with diffuse, low-level, public risks, the balance between risk created and risk averted is much more subtle. Anyone can declare with confidence that drunk driving contributes unfavorably to the state of our risk environment. But it is much less easy to conclude the same of a somewhat risky vaccine, pesticide, or power plant that serves millions and adds both a small (per capita) quantum of risk and a quantum of risk reduction (perhaps smaller, perhaps larger) to our ambient environment. With public risks Were often is, in fact, a sharp division between affirming, We security interests of the public as a whole and tending to Bose who have been injured by the hoard in question. For many somewhat risky but in the aggregate nsk-reducing products and services Cat may be deterred by tort law activity, He interests of future consumers are directly opposed to Pose of unlucky prior consigners who have already been injured by the hazard in ques- tion. lathe injured obviously wish to be compensated, and a compassionate, generous society should surely respond. But compensation at Be expense of Be creator of Be public risk will be against Be interests of future consumers whenever Be good that Be risk attends removes more risk from Be environment than it adds. In such cases Be public security is best served by absolving Be creator of Be public risk from responsibility for Be private injury, notwith- standing Be fact Cat Be private injury is real and its cause is known. lnefficzent Compensation The new tort law does not serve as an effective tool for compensating public-nsk victims, either. One might think that the loss for the large, cor- porate defendants must be a gain for small, individual plaintiffs, and this is indeed a belief that the Tnal Lawyers of America have clllt~vated with the greatest and most delicate care. But litigation is by no means a zero-sum
THE BHOPALlZATION OF AMERICAN TORT LAW 201 game at least not until it is understood that lawyers themselves are always among, the players and invariably among the winners. Study after study has revealed that the tort law is highly capricious, in- efficient, unfair, and most of all, terribly costly. For every dollar that finally ends up in the pocket of an injured plaintiff, perhaps three to five are diverted to lawyerslawyers for the plaintiff and defendant, judges, law clerks, and miscellaneous other camp followers. The tort system is, quite simply, an insurance scheme with astronomically expensive agents and middlemen. If a private insurance company pocketed 80 cents on every dollar collected it would surely be prosecuted for fraud. But the tort system provides insurance at about that price. On top of this, many injured plaintiffs do not recover at all from Tort Law Insurance, Inc., while a quite unacceptable number of uninjured plaintiffs recover in large amount. The new tort system has a high decree of randomness to it. This appeals to gamblers, of course, as well as to "house" employees who take their cut in any event. Insurance, however, is supposed to take the gambling out of life. Kindling the Flames Finally, the new tort law does not offer one useful social function that was certainly provided under the old tort law: resolving fresh, focused, bipolar disputes. Some method must exist for resolving civil controversies, and as a sociological matter it is probably best that ordinary two-person quarrels be resolved by neutral nonexperts and lay junes. Not because these traditional decision makers are more likely than specialists to render verdicts of Solo- monic insight and wisdom, but because Hey are most likely to be perceived by the public at large as accessible, moderate, and experienced with life's more ordinary vicissitudes. As a great judge once pointed out, it is often more important that things be settled Man that they be settled right. The new tort law does not settle, it unsettles. When 3 million plaintiffs sue 60 defendants for something Hat happened 30 years ago or that may happen 30 years from now the courts do not deal with a "dispute." They are engaged, instead, in large-scale social engineering. As such, their activ- ities should be tolerated only if they offer efficient deterrence or fair com- pensation. As we have seen, the new tort law offers neither. The Waiving on the Wall Courts, to be sure, have been reluctant to acknowledge their own limi- tations when it comes to prudent management of broad-ranging "public" risks. Recent judicial trends suggest that the courts generally believe Hat they are perfectly competent to resolve any public-risk question, no matter
202 PETER W. HUBER how large or complex. Congress and state legislatures, however, appear to be reaching precisely the opposite conclusion with increasing frequency. The list of activities legislatively insulated from the tender mercies of the plain- tiffs' bar grows steadily longer. For years state legislatures have seen fit to place liability limits on employee tort recoveries from employers.7 The United States, aloe:, with 130 other governments, has determined that international civil aviation requires liability limits;8 likewise, nuclear power has been thought to require liability-limiting legislation..9 Congress Wanted tort immunity to pharmaceutical companies in order to assure their participation in the swine flu vaccination program; currently, both Congress and the Reagan administration are looking at a proposal to limit manufacturer liability for all types of vaccination. Other activities that have been granted partial or complete tort immunity include cleaning up a hazardous waste dump,~° cleaning up accidental discharges of oil and hazardous substances, participating in "unusually hazardous or nuclear" activities on behalf of the Department of Defense, and partici- pating in the Space Shuttle program. ]3 Additional statutes that have codified liability limits or immunities include the Outer Continental Shelf Lands Act Amendments of 1978,~4 the Deepwater Port Act, and the Trans-Alaska Pipeline Act.'6 Two recent federal initiatives have advanced comprehensive proposals to curtail and standardize product-liability recoveries through pro- defendant changes in rules of evidence and standards of conducted And in 1984, Congress passed legislation retroactively batting lawsuits against pri- vate contractors who participated in the early atomic weapons testing pro- gram. '8 Finally, Congress has recently considered a bill to provide a federal insurance system for "orphan" drugs. i9 The most striking feature of the long and steadily growing list is that it addresses activities that, in a more rational world, would entail medals of honor, not tort suits. Steady employment in almost any industry is much safer than living in the poverty that attends unemployment; indeed We "healthy worker" effect is such a strong one that epidemiologists attempting, to eval- uate occupational heals risks must expressly make allowance for it in their statistical studies of worker health. Nuclear power is demonstrably much safer than its main alternative coal-fired power generation. At the time it was enacted, the swine flu vaccination program represented a prudent mea- sure to protect the public health, and vaccination generally has been one of the greatest risk reducers of the century. Consumer products reduce aggregate risk in numerous but modest ways. And it is magnificently ironic to discover that ton immunities are needed for those who clean up hazardous waste dumps or oil spills, or who manufacture certain types of valuable drugs. Lawyers tend to react to liability-limiting initiatives with disgruntlement, recognizing perhaps that liability limits cut off lawyers' livelihoods. It is often suggested, especially in connection with nuclear power, that no industry
THE BHOPALIZATION OF AMERICAN TORT LAW 203 marketing an "acceptably" safe product or service should require the pro- tection of a liability limit; the very existence of such a limit indicates that the hazards of the activity are unacceptably high. The argument is laughable. Nuclear power, civil aviation, commercial drugs, and the like are creatures of science and engineering, and their risks are therefore both predictable and inherently self-limiting. Yet undertaking a venture that is financially sound considering the physical risk involved is often a pure gamble in light of what may happen in the courts. Immunities and liability limits are the public-risk equivalents of Good Samaritan laws; that they are required at all is a ringing indictment of the level of foolishness to which our contemporary tort system aspires. Something is dangerously wrong, when a pharmaceutical company requires legislative protection before it will dare to manufacture a risk- . , . reducing vaccine. THE AGENCIES AND THE COURTS The fact remains that the universe of public risk, and compensating victims for their injures, cannot be left simply to the free market. Public hazards inherently require public control, and despite our unwillingness to socialize many natural risks, our society has been unwilling to leave the victims of many man-made hazards uncompensated. If the courts are not qualified to manage public risks, who is? The answer is painfully obvious to almost everyone outside the legal community administrative agencies, not the courts. Institutional Competence The spheres of competence for the courts and the agencies mirror the division between private and public risks. The courts perform adequately in risk regulation when they deal with private risksfocused, high-probability, bilateral hazards that have ripened (or are about to npen) into concrete injuries. This class of risks is amenable to rational control through the retail, retrospective regulation that courts have traditionally supplied. The judicia- ry's competence in dealing with such risks cannot be questioned. The private dispute involving the realized risks of car accidents, slips-and-falls, hazardous conditions on land, and the like can find no better governmental forum for resolution than a court. No doubt the court fulfills some regulatory role in adjudicating tort Suits of this nature, but the regulation is of a bilateral controversy for which the antisocial nature of the challenged conduct can be ascertained with relative ease and certainty. In this private-risk settin;,, of course, the regulatory agency can be of little help. Constitutional difficulties would attend relegating this type of dispute to an administrative forum, and private-risk controversies are so common and varied that as a practical matter they inevitably will have to be resolved
204 PETER W. HUBER in trial-like proceedings presided over by generalist judges. The administra- tive agency is not needed here; the judicial arena already provides more or less the right forum for resolving social confrontations of this character. Precisely the opposite holds true for public-nsk choices. The assessment of these diffuse, low-probability, multilateral, and temporally remote hazards requires close attention to both sides of the public-risk picture risks incurred and risks averted, persons harmed and persons helped by the activity in question. The task, in other words, requires a "public'' point of view on the problem.20 This is a perspective that public agencies can supply, and that courts plainly cannot. The reasons are not difficult to discern. Beneficiaries of risk-reducing products and services do not litigate, and contingency-fee lawyers rarely seek them out as clients. The courts are thus not likely even to hear from those whose interests are a critical component of public-risk assessment. To be sure, the vaccine manufacturer or the operator of the nuclear power plant will attempt to serve as a surrogate spokesperson. But industry serves very poorly in this representative capacity, because it has every incentive to claim risk-reducing and other social benefits from its activities whether or not they exist. In contrast, regulatory agencies, such as FDA, EPA, and NRC, can ground their assessment of public risks on a comparison of the hazards of the alter- native drug, power plant, or pesticide already on the market Even in the agencies, comparative-risk regulation is not all that it should be. But there are, at least, some promising signs here. Progressive movement toward more comparat~ve-risk regulation remains possible, perhaps even inevitable in the agencies, because their focus is a relentlessly public one. "Public law" adjudication in the courts, so eloquently described by Abram Chayes2~ (in general terms) and by Joel Yellin, David Rosenberg, and others22 (with specific application to risk regulation), simply does not and cannot yield progressive choices among public risks. The public law model of ju- dicial action is a ;~,rand design for having lawyers decide everything every- where. But in risk matters, at least, experience and common sense teach that the only beneficiaries are likely to be lawyers themselves. Deferring to the Experts What, then, are the courts to do with individual or class actions that seek redress from risk or injury caused by an intrauterine device (IUD), vaccine, herbicide, moming-sickness drug, nuclear waste reprocessing facility, or the like? My general prescription is less, not more. The courts should defer to the experts. Not experts summoned by the parties for the edification of the judiciary. The scientific community is large and heterogeneous, and a Ph.D. can be found to swear to almost any "expert'' proposition, no matter how
THE BHOPAWZATlON OF AMERICA TORT LAW 205 false or foolish. The expert public-risk choices that should be respected are those made by a risk-re,ulatory agency concerned with the public risk in question. This is not the current law. Judges, at present, generally feel quite free, when addressing, risk problems, to ignore prior administrative determinations regarding which public risks represent progressive choices. The Restatement of Torts flatly declares that "compliance with a legislative enactment or an administrative regulation does not prevent a finding, of negligence."23 Many nsk-related administrative statutes, such as the Consumer Products Safety Act, expressly announce that compliance with safety rules is not to serve as a shield from tort liability.24 When the statute is silent or even the least bit ambi ,uous on the question of tort-remedy preemption, the courts will always presume that no preemption was intended.95 The Karen Silkwood decision, for example, announced that the operator of a nuclear facility can be assessed punitive damages despite complete compliance with applicable NRC regu- lations. And in Chevron v. Ferebee, a recent case from the D.C. Circuit Court of Appeals, a chemical company was held liable for '`mislabeling" a can of paraquat, even though the label had been approved by the EPA and even though the manufacturer could not lawfully have used anything else- only EPA had the legal power to change the label. In short, the settled judicial refusal to defer to agency choices among public risks has spawned a never-ending cycle of truly perverse risk- regulatory decisions.26 The FDA's experts may conclude after a careful ex- amination of the substitutes that the Sabin polio vaccine is a better bet than the Salk vaccine, or the NRC may conclude that a nuclear power plant is safer than the available alternatives. But a mass-exposure lawyer and an injured or merely frightened client can effectively overturn the expert judg- ments by finding what is almost too easy to find: a judge who is of the view that the courts have something useful to offer in these matters, and a sym- pathetic and well-meanin~ jury. The agency's blessing of the vaccine, or pesticide, or power plant is usually no more than a prelude to a second tier of regulation in the courts. The resulting you-say-yes-but-I-say-no pattern of public-risk decision mak- in~ is routinely explained by the courts and often accepted by the agen- cies~n the ground that administrative risk regulation is intended to set only a safety "floor" a threshold of performance, a minimum definition of `' acceptable" safety. Agencies will point out that they rely mostly on data furnished by the regulatees themselves and therefore cannot and should not be viewed as the final arbiters of how much safety is enough. It is, of course, politically comfortable for agencies to preserve the tort system as a safety valve. When bodies do fall the agency is grateful to see at least part of the victims' hostility directed at the re~,ulatees. This rationalization for judicial nondeference may make some sense when
206 PETER W. HUBER the administrative regulatory regime is casual or sporadic, as with consumer products. But it is wholly unpersuasive for comprehensively regulated in- dustries. Vaccines, pesticides, aircraft, electric power plants, and the like all entail potentially enormous mass-exposure hazards. Precisely because they can create public risk of this nature, these products and services are also subject to the most searching and complete state and federal safety regulation. When that regulation reaches a favorable conclusion about a potentially hazardous technology, that conclusion must inevitably be based at some level on a conclusion Hat the technology represents some measure of progress or, at worst, no measure of regression in the risk market in question.27 Once that determination has been made by an expert licensing agency, the courts should respect it. Requiringor at least strongly encouraging- the courts to respect the comparative-nsk choices made by competent, expert agencies would inject a first, small measure of rationality into a judicial regulatory system that currently runs quite wild. The Nuclear Regulatory Commission's licensing of a nuclear power plant or reprocessing facility must be viewed for what it isconsiderably more, in other words, than a routine and irrelevant pleasantry, to be forgotten as soon as the first tort plaintiff enters He courthouse. Regulatory agencies are equipped to make He risk comparisons on which all progressive transformation of the risk environment must be based. It has always been true Hat ignorance of the law is no excuse. At present, knowledge of the law is no excuse either. It should be. Compensating victims The problem remains: How do we attend to the needs of the tragic vic- timsthe unfortunate few who are injured by the whooping cough vaccine, or He FDA-approved IUD, or He EPA-approved toxic waste dump? There are some risks that are real, Hat have tragic consequences, and that must be socialized if only because He courts will not allow otherwise. It may seem perverse to insist on a generous social program to deal win the hazards of die whooping cough vaccine when we have only the stingiest social programs to deal win He hazards of whooping cough itself. But the judicial system has left no over choice. But once again, there are models outside the judicial process to be followed, and once again He lead In adopting these models must come from legislatures and executive branch officials. Two models of administrative compensation systems are already in place to be emulated. The first is worker's compensation: it addresses public risks falling toward He "inf~nity-zero" end of the spectrum, in which broad exposure to the hazards is certain but in which the likelihood of consequences to any exposed individual is small. One may start by acknowledging some of the problems Cat worker?s compensation statutes have encountered. Compensation sched-
THE BHOPALIZATION OF AMERICAN TORT LAW 207 ules, for example, have not kept pace with inflation. There have been abuses under the worker's compensation statutes, most especially in the areas of causation. Faced with the "deep pocket" of an insurance fund, arbitrators and courts have been inclined to adopt very expansive definitions of what constitutes a "job-related" injury. Finally, worker's compensation schemes have been successfully bypassed, increasingly often, by lawyers and plaintiffs eager to play once again at the judicial gaming tables. Novel theories have been invented, for example, to bootstrap occupational injures into "product- liability" cases. Yet despite all the criticisms that have been raised against worker's com- pensation systems, they have worked tolerably well. It is inconceivable that any jurisdiction in this country will choose to return to He alternative of open-ended tort litigation Nat preceded the adoption of worker's compen- sation statutes at the turn of this century. Administrative compensation sys- tems are correctable, and can be corrected across the board, without recourse to expensive legal middlemen, if and when problems develop. Thus the problems relating to "causation" and the definition of "on-thejob" injures encountered by the present system are not beyond resolution, most especially within an administrative forum that can learn from expenence. And the spillover from worker's compensation to product-liability liiigai~on and other areas can be curtailed, most especially if similar administrative compensation schemes are developed around the boundaries of worker's compensation statutes. ~ Be development of administrative compensation schemes for vaccines, drugs, and He like, worker's compensation systems can still serve as a valuable model. A second compensation model is the one embodied in the much-maligned Price-Anderson Act. This compensation system addresses hazards at He 'izero-inf~nity" end of the public-nsk spectrum, in which exposure is very unlikely, but if it occurs it is likely to be in conjunction with a very large and far-reaching accident. Other activities presenting risks of this character (many of Hem considerably larger Han He risks from nuclear power) include the operation of a chemical plant, gas pipeline, liquid natural gas (LNG) tanker, hydroelectric dam, and so on. If we are to use technologies that present risks of this character at all, we should do so under a regime of definite and predictable disaster insurance. The insurance must establish a fund to be tapped in case of accident, delineate the rights of victims, identify a party win primary financial responsibility, provide for prompt compensation, and establish fixed limits on liability wi~- out which none of the other components will be accepted and the enterprises themselves will not be possible. Such schemes need not eliminate the ad- judicatory function of He courts altogether, but Hey must, at a minimum, establish }ules to streamline He adjudicatory fimction and reduce the incen- tives for tactical maneuvering and delay. The areas in which streamlined administrative compensation schemes are
208 PETER W. HUBER most suitable share some common features. First, there are areas in which there is a pressing need to provide compensation to the victim (as distin- guished from his lawyer) quickly and without a great deal of legal maneu- vering. Tragic accidents attributable to vaccines, occupational hazards, nuclear power plants, and the like can be shockingly disruptive and difficult for the injured individual and his or her family. In this context, the ponderous compensation machinery provided by the conventional tort system is unac- ceptable: it moves too slowly, its outcomes are too unpredictable, and it diverts far too much to lawyers. Almost all nonlawyers who have watched the sordid legal maneuvering over the tragedy at Bhopal, India, have been forced to wonder whether the system that the chemical industry now must use for dealing with the financial aspects of such accidents is as it should be. Most disinterested observers find it clear that compensation in this area could and should have been paid quickly and distributed fairly without the endless legal wrangling already witnessed. Second, industries that attract legislation often involve novel technologies and rare or unfamiliar hazards. The reason is simple: it is precisely in these areas that insurance is most difficult to obtain, not because the hazards are necessarily large, but because there is no actuanal record of real-world ac- cident experienceor of legal-world tort historyon which the insurance industry can rely. This is most clearly true with the emerging, so-called toxic torts, which usually involve low-level, broadly distributed, long-latency pe- riod nsks. It is these two factors- the need to expedite compensation to victims and the need to provide important industries with some measure of protection from the caprices of the tort system that have applied steady pressure to legislatures across the country. The worker's compensation statutes and the Price-Anderson Act are not anomalies; in an increasing number of areas they are serving, as vital models. Perhaps the major risk in setting up new administrative compensation systems is that they may become the quintessential "deep pocket" targets for nuisance actions and fraudulent claims. One vice in the present tort system is also a virtue the system is so tembly ponderous, slow, and unpredictable that it filters out some nonmeritonous claims through sheer inertia. But if so-called Orangemail is becoming a real problem even in the tort system, the problem might be magnified under a more streamlined administrative compensation system. Worker's compensation has, for example, been sought by the wife of an executive who suffered a heart attack while engaging in sexual relations with his secretary, and insurance funds set up under He Price-Anderson Act have been occasional targets for nuisance suits. The only protection against this problem Is the old-fashioned one: a claim- ant for funds must be required to show causation, if not beyond a reasonable doubt, then at least with some serious degree of scientific credibility. En
THE BHOPALIZATION OF ~ERICk TORT LAW 209 addition, disaster compensation schemes can be insulated from nuisance actions by provisions such as the "Extraordinary Nuclear Occurrence" threshold codified in the Price-Anderson Act: no payouts are pennitted until some appropriate, high-level official has found that the alleged disaster was a disaster. There has been much criticism from plaintiffs' lawyers of the fact that the accident at Three Mile Island was not "extraordinary" enough for Price-Anderson purposes. But then, there has been much criticism in the scientific community of the fact that Bendectin and Agent Orange were thought to justify enormous tort settlements, notwithstanding underlying "evidence" of causation that ranged from the tenuous to the entirely fanciful. PUBLIC RISKS AND POLITICAL LEGITIMACY The expansion of tort law into public-risk areas provides a poor vehicle for compensating deserving victims of diffuse hazards, even while it provides some compensation to capriciously chosen, nonvictim members of the public, and a great deal of compensation to systematically self-selected members of the legal profession. The expansion serves as a very poor indeed often counterproductive regulatory system: tort law provides no vehicle for sys- tematically selecting worse from better performers in the public-risk market. But perhaps most troubling of all, the new tort law undermines the process of democratic self-govemment. The Constitution itself embodies a vision of the courts as the retrospective "retailers" of justice and the political branches of government as the pro- spective "wholesalers." Legislatures and agencies are broadly disqualified from doing the types of things courts traditionally do: singling out individuals for special benefit or burden (barred by the Bill of Attainder Clause) and administering retrospective punishment (batted by the Ex Post Facto Clause). Until recently, activities of the courts were confined by mirror-image re- straints found in the constitutional requirement that the courts resolve only "cases" or "controversies" the courts had to operate retail, and they had to operate more or less retrospectively. These divisions of authority make a good bit of sense: courts do well in administering retail, corrective justice; only the political branches, on the other hand, are qualified to make wholesale public policy. The new tort law, and the entire vision of '~public law" adjudication that animates it, is politically dangerous. It removes the most controversial social issues from the arena in which controversy can be brokered and tensions diffused through the very process of political participation. The legal system is adversarial; the answers it provides are often grudgingly accepted, but rarely leave participants satisfied that the system is reachable and open for change another day. Judges are by deliberate constitutional design iso- lated, remote, not accountable to the electorate. These are valuable assets
210 PETER W. HUBER for decision makers engaged in protecting civil rights and minonties, re- solving private disputes, and umpinn=, the democratic process. But they are grave liabilities for decision makers who would address and determine Me broadest terms of Me social contract Our choices of "acceptable" public risks and the decisions we make to compensate- or not to compensate for the ha:Tns that may attend risks of this character ultimately must be made in the political arenas. Choices of this character that are made anywhere else will not, in the long term, prove acceptable to a people who have grown accustomed to governing themselves. NOTES 1. See Weinberg, Science and Its Limits: The Regulators' Dilemma, paper presented at the National Academy of Engineenag Symposium on Hazards: Technology and Fairness, Wash- ington, D.C., June 3-4, 1985. 2. 28 Ch. 688 (1885). 3. Scientists' Institute for Public Information, Inc. v. Atomic Energy Commission. 481 F.2d 1079 (D.C. Cir. 1973). 4. U.S. Nuclear Regulatory Commission, Reactor Safety Study, Report WASH-1400, NUREG 75/014 (1975). See Kitch, A Shot in the Dark: Product Liability and Vaccination, Regulation, pp. 11-18 (May-June 1985). See also Committee on Public-P:ivate Sector Relations in Vaccine Inno- vation, Institute of Medicine, National Academy of Sciences, Vaccine Supply and Innovation (Washington, D.C.: National Academy Press, 1985). 6. I discuss these factors at greater length in Huber, Safety arid the Second Best: The Hazards of Public Risk Management in the Courts, 85 Colum. L. Rev. 277, 317-320 (1985). Predictably, however, trial lawyers have begun to find their way around these limits by bringing suits against product suppliers and manufacturers rather than against the employers themselves. See, e.g., Kajala v. Johns-Manville Prods. Corp., 523 F.2d 155 (8th Cir. 1975). Courts that believe this is not liability `'against'' the employer are engaged in self-delusion. Employee awards against manufacturers will be channeled back to employers quickly enough, by means of rnanufacntrer demands to employers for indemnity, insurance, or a risk premium on the products that they sell. See 7 U.S. Deptt of Commerce, Interagency Task Force Report on Product Liability 75-77 (1977). At least one state, Connecticut, has responded by requinug worker's compensation awards to be set off against verdicts or settlements against product manufacturers. 1979 Conn. Pub. Act 79483, § 1 1(~). 8. The original agreement is contained in the Warsaw Convention, 131 L.N.T.S. 11 (1929). 9. See 42 U.S.C. § 2210 (1982). 10. See Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency, Back- ground Report for the Inde:T~nificaiion Report to Congress at A-27 (1983). 1 1. See 33 U.S.C. § 1321 (1982). 12. See 50 U.S.C. § 1431 (1982); Executive Order 10,789' 23 Fed. Reg. 8897 (1958); see also Stencel Aero Eng'g Corp. v. United States, 431 U.S. 666 (1977) (immunity for suppliers of materials to government where government itself enjoys immunity). 13. See Background Report, supra note 10, at A-26. 14. Pub. L. No. 95-372, 92 Stat. 629 (see § 304(b)(1), liability limit) (codified as amended in scattered sections of 43 U.S.C.). 15. 33 U.S.C. §§ 1501-1524 (1982) (see § 18(d), liability limit). 16. § 204, 43 U.S.C. § 1653(a)(2) (1982)
THE BHOPAI~ZAT10N OF AMERICAN TORT LAW 211 17. For a summary of recent attempts at federal legislation see Kircher, Federal Product Legislation and Toxic Torts: The Defense Perspective, 28 Vill. L. Rev. 1116 (1983); Phillips, The Proposed Federal Product Liability Statute from the Toxic Tort Plaintiff's Perspective, 28 Vill. L. Rev. 1156 (1983); Schwartz & Means, The Need for Federal Product Liability and Toxic Tort Legislation: A Current Assessment, 28 Vill. L. Rev. 1088 (1983); Twerski. National Product Liability Legislation: In Search For the Best of all Possible Worlds, 18 Idaho L. Rev. 411 (1982). After extensive study, the Commerce Department proposed a Union Product Liability Act (UPLA) as a model for state legislation; parallel federal legislation has also been submitted to Congress. See S. 2631, 97th Cong., 2d Sess. (1982). The bill was reported on favorably by the Senate Commerce Committee in October 1982, but was not passed by the Senate. It was reintroduced in the 98th Congress. See S. 44, 98th Cong., 1st Sess. (1983). The springboard for federal action was a study conducted by an interagency task force. The report criticized contemporary tort law for failing to develop a clear standard of responsibility, imposing liability when the product has been misused or altered, failing to apply standards of comparative fault, failing to provide for contribution or indemnity among tortfeasors, imposing liability for design defects beyond the technological state of the art, imposing liability for unknowable nsks, and imposing ling,enng liability beyond the expected useful life of a product. Numerous state legislatures have followed up on the federal initiative. See Ghiardi, Products Liability Where is the Borderline Now? 13 Forum 206, 212 (1977). 18. See Department of Defense Authorization Act, 1985, § 1631 P.L. 98-525, 98 Stat. 2492, 2646 (1985); see Taylor, New Act Restricts Atomic Test Suits, N.Y. Times, Nov. 4, 1984, § 1 (News), at 26, colt 1. All lawsuits must be brought, instead, against the federal government under the Federal Tort Claims Act. 28 U.S.C. § 2674 (1982). We new legislation eliminates jury trials, punitive damages, and other advantages of state law adjudication that had prompted more than 100 lawsuits by veterans and others against the contractors. 19. See H.R. 7089, 96th Cong., 2d Sess. (1980). 20. It requires, in addition, technical resources that the courts do not possess. See, e.g., WiLkey, Activism by the Branch of Last Resort: Of the Seizure of Abandoned Swords and Purses 10- 12 (1984); Muntzing, The Courts and Energy Policy in the Judiciary in a Democratic Society 140 (1-heberge, ea., 1977). 21. Chayes, The Role of the Judge in Public Law Litigation, 89 Harv. L. Rev. 1281 (1976). 22. See Yellin, High Technology and the Courts: Nuclear Power and the Need for Institutional Reform, 94 Harv. L. Rev. 489 (1981); Rosenberg, The Causal Connection in Mass Exposure Cases: A "Public Law" Vision of the Tort System, 97 Harv. L. Rev. 851 (1984). Other contributions to the debate include Perrow, Living with High-Risk Technologies (1984); Ba- zelon, Science and Uncermuty: A Junst's View, 5 Harv. Envt'l. L. Rev. 209 (1981); Bogen, Public Policy and Technological Risk, 1980 Idea 37; Cohen, Innovation and Atomic Energy: Nuclear Power Regulation, 196~Present, 43 Law & Contemp. Probs. 67 (1979); Kraus, Environmental Carcinogenesis: Regulation on the Frontiers of Science, 7 Envt'l. L. 83 (1976); Merrill, Risk-Benefit Decisionmaking by the Food and Drug Ac~rr inistration, 45 Geo. Wash. L. Rev. 994 (197?); Pierce, Encouraging Safety: The Limits of Tort Law and Government Regulation, 33 Vand. L. Rev. 1281 (1980). 23. Restatement (Second) of Torts § 288C (1965) . 24. Section 25(a) of the Consumer Products Safety Act provides: "Compliance with consumer product safety nobles or other rules or orders under this chapter shall not relieve any person from liability at common law or under State statutory law to any otherperson." 15 U.S.C. § 2074(a) (1982). 25. The Supreme Court, for example, recently let stand a $4.75-million damage award against a manufacturer of a birth control pill in a suit brought by a woman whose kidneys had failed after several years of using He drug. The manufacturer challenged only the $2.75 million in punitive damages that were awarded notwithstanding its complete compliance with regulatory
212 PETER W. HUBER requirements imposed by the FDA. Ortho Pharmaceutical Corp. v. Wooderson, 235 Kan. 387. 681 P.2d 1038, cert. denied, 105 S. Ct. 365 (1984). 26. The mo~ning-sickness drug Bendectin, for example, continues to enjoy FDA approval; there is no scientifically credible evidence that Bendectin does anything except what it is supposed to: relieve the misery of morning sickness that can threaten the health of both mother and child. But Bendectin became such a popular target for tort lawsuits that its manufacturer withdrew the drug (the only such drug on the American market) and submitted to a $120- million class-action settlement. This prompted even the New York Times, not lcnown for its conservative views on nsk-regulatory matters, to editorialize: ''With Bendectin . . . the law has made a devastation and called it a settlement." N.Y. Times, July 30, 1984, at A20, colt 2. 27. For further discussion of my view of risk ''markets," see Huber. The Market for Risk Regulation, p. 33 (March-April 1984).
