Overcoming the Identification Burden in DES Litigation: The Market Share Liability Theory


The English case of Winterbottom v. Wright is generally viewed as marking the beginning of product liability law. When Winterbottom was decided in 1842, the industrial sector of society was in its infancy and the courts, in order to encourage industrial development, sought to restrict the tort liability of industrial concerns. In the 140 years since Winterbottom, as the industrial sector has become more powerful, the courts have expanded the liability of manufacturers and generally made it easier for a plaintiff to recover for a product-related injury. This expansion has occurred in response to drastic changes in the market place since pre-industrial times. The direct relationship between manufacturer and consumer has disappeared as more sophisticated methods of manufacture and distribution lengthen the chain of commerce. Also, products have become more complex, reducing the ability of the consumer to evaluate his purchases intelligently. These changes have forced a shift in the judiciary’s view of which parties should be afforded greater protection as a matter of public policy.

Overcoming the Identification Burden in DES Litigation: The Market Share Liability Theorys, Marquette Law Review, Volume 65, Issue 4, Article 4, Summer 1982.

Despite the historical trend toward liberalization of the requirements for recovery in product liability actions, plaintiffs must still establish three elements:

  1. that the product that injured them was defective;
  2. that the defect caused the plaintiff’s injury;
  3. and that the defect was attributable to the party to be held responsible.

The last element – called the identification requirement – requires the plaintiff to specify the particular manufacturer of the product that injured him. Unless the plaintiff can do so, an essential element of his case will be lacking and recovery will be precluded. Traditionally, the rule has been strictly applied even where two or more manufacturers produce an identical product: unless the plaintiff can exclude all but the specific manufacturer of the particular product that injured him, recovery will not lie, even if all manufacturers were clearly negligent

Recently the courts have been faced with a series of cases wherein the wisdom of the identification requirement, when applied to a certain factual context, has been seriously questioned. Most of these cases have been brought by the daughters of women who ingested the drug diethylstilbestrol (DES), prescribed to prevent miscarriage. These daughters allege that their in utero exposure to DES has caused them to develop a variety of serious genital tract disorders. The plaintiffs in these cases have faced numerous roadblocks to recovery, the most serious being their inability to identify the specific manufacturer of the DES their mothers ingested. In earlier DES cases where the identification requirement was strictly applied, plaintiffs were barred from recovery. More recently, however, two courts have shown a willingness to apply a market share theory of liability under which plaintiffs can circumvent the identification requirement in DES cases, thus increasing their chances of recovery.

This comment will assess the unique nature of the DES cases, and analyze and critique the various theories that have been proposed in an effort to assist plaintiffs in overcoming the identification requirement. Particular emphasis will be placed on the market share liability theory and how it can best be adapted to achieve an equitable result in DES litigation.


Diethylstilbestrol (DES) is a synthetic estrogen. In 1947, the Food and Drug Administration (FDA) authorized the marketing of DES for use as a miscarriage preventative, but only on an experimental basis, with the requirement that the drug contain a warning label regarding its experimental status. Between 1947 and 1971, hundreds of pharmaceutical companies marketed DES, and it was prescribed for millions of pregnant women. In 1971, statistically significant relationships between the use of DES and the subsequent development of genital tract cancer in the users’ daughters were reported in the medical literature. In that same year, the FDA banned the marketing and promotion of the drug for use as a miscarriage preventative is on the basis of its apparent danger and ineffectiveness.

Estimates of the number of “DES daughters” – women whose mothers took DES during pregnancy – range from five hundred thousand to three million. In lawsuits against DES manufacturers, plaintiffs have alleged that DES causes a variety of gynecological disturbances including structural deformities, adenosis and clear-cell adenocarcinoma, the form of cancer linked to DES in earlier studies. Estimates of the actual number of suits filed range up to one thousand, and many of the major drug companies have been brought in as defendants.

The plaintiffs in these cases have been able to present a strong case that the drug manufacturers were negligent in the marketing and testing of DES. Justice Mosk in Sindell v. Abbott Laboratories stated that:

During the period defendants marketed DES, they knew or should have known that it was a carcinogenic substance, that there was a grave danger after varying periods of latency it would cause cancerous and precancerous growths in the daughters of the mothers who took it, and that it was ineffective to prevent miscarriage. Nevertheless, defendants continued to advertise and market the drug as a miscarriage preventative. They failed to test DES for efficacy and safety; the tests performed by others, upon which they relied, indicated that it was not safe or effective. In violation of the authorization of the Food and Drug Administration, defendants marketed DES on an unlimited basis rather than as an experimental drug, and they failed to warn of its potential danger.

Plaintiffs have faced a raft of difficulties in the DES suits, however. These include problems in meeting the requirements for class action certification, the possible inability to establish a cause of action for fetal injury prior to viability, and the running of the statute of limitations in jurisdictions where the cause of action is deemed to accrue at the time of actual injury, rather than at the time the injury is discovered. Clearly, however, the plaintiffs’ most serious problem has been their inability to identify the specific manufacturer of the DES that caused them harm. Several factors have given rise to this problem. First, the sheer number of DES manufacturers (between approximately one hundred and three hundred firms) makes identification difficult. Second, the genital tract cancer associated with DES (adenocarcinoma) has a ten to twenty year latency period. With the passage of time, prescription or purchase records which might identify the source of the DES to which the plaintiff was exposed – whether those records were possessed by the plaintiff’s mother, the mother’s doctor, or the mother’s pharmacist – are likely to have disappeared. Third, even if prescription records were still available, it could not be said with certainty that the drug brand specified on the prescription blank was the actual brand sold to the plaintiff’s mother. DES is a fungible drug produced by many firms, and it was once a common practice of pharmacists to fill prescriptions for DES with whatever brand they had on hand, whether or not that brand was the one specified on the prescription blank.

For the above reasons, plaintiffs argue that strict adherence to the identification requirement in DES cases is unfair. Plaintiffs have proposed three alternate theories of liability – a “concert of action” theory, an “enterprise liability” theory, and an “alternative liability” theory – under which they might recover notwithstanding the difficulties posed by the identification requirement. These three theories, and the courts’ responses to them, are discussed below.


Section 876 of the Restatement (Second) of Torts recognizes three factual patterns in which a concerted action theory might be applied to hold multiple defendants jointly and severally liable. In the first, all defendants act tortiously toward the plaintiff according to a common plan or design. Thus, for example, if A, B, C and D went together to E’s house with the common intent to commit a robbery, and A broke down E’s door, B tied E up, C beat him and D converted his property, A, B, C and D would be jointly and severally liable for the total damages caused by their own and each other’s torts. In the second fact pattern, one or more defendants give substantial assistance or encouragement to another defendant, knowing that the latter’s behavior is tortious. Thus, for example, where A urges B to throw a rock at C, and B does so, injuring C, both A and B are jointly and severally liable for C’s damages, even though A’s behavior, separately considered, was not tortious. In the third fact pattern, two or more defendants participate in a joint activity, and any one defendant is jointly and severally liable for the results of the united effort if his act, considered separately, is tortious, irrespective of his knowledge that his act or the acts of the others is tortious. For example, each of a number of trespassers who are jointly excavating a ditch is liable for the entire harm caused thereby, although none believes that he is trespassing. Prosser summarizes the concerted action doctrine as follows:

Those who, in pursuance of a common plan or design to commit a tortious act, actively take part in it, or further it by cooperation or request, or who lend aid or encouragement to the wrongdoer, or ratify and adopt his acts done for their benefit, are equally liable with him. Express agreement is not necessary, and all that is required is that there be a tacit understanding ….

The gravamen of the plaintiffs’ charge of concert in the DES cases is that the defendants relied upon each other’s inadequate tests and took advantage of each other’s promotional and marketing techniques. For example, plaintiffs note that in 1941, twelve drug companies submitted a joint clinical file pursuant to their request for FDA approval of DES.

The Sindell court went on to state that there is no allegation here that each defendant knew the other defendant’s conduct was tortious toward the plaintiff, and that they assisted and encouraged one another to inadequately test DES and to provide inadequate warnings…. There was no concert of action among defendants within the meaning of that doctrine.

Historically, the concerted action theory developed to discourage tortious group behavior by expanding the scope of liability for a tortiously caused injury: it was not created specifically to aid plaintiffs in overcoming the identification requirement. The concerted action theory typically is applied in situations in which, unlike in the DES cases, a particular defendant is already identified as the factual cause of the plaintiff’s harm, and the plaintiff merely wishes to extend liability to those acting in league with that defendant. For example, the theory has been applied in illegal drag race cases to hold all participants jointly and severally liable even though only one participant actually struck the plaintiff. In cases such as these where the concert of action theory has been applied, identification of the defendants has rarely been an issue.

For the above reasons, plaintiffs have rarely been successful’in suing on a concert of action theory in DES litigation. Outside the DES area, the concerted action theory has never been successfully applied to avoid the identification requirement in product liability cases.


A second theory under which plaintiffs have attempted to avoid the identification requirement in DES cases has been termed “enterprise liability.” Numerous different theories have appeared under this general rubric. This comment adopts the theory of enterprise liability first suggested in Hall v. E.L DuPont de Nemours & Co. That case involved twelve separate accidents in which thirteen children were injured by dynamite blasting caps. Evidence identifying the manufacturer of the caps was destroyed in the explosions. The defendants were the six major domestic manufacturers of blasting caps and their trade association, which comprised virtually the entire blasting cap industry in the United States. Several Canadian manufacturers, however, could have supplied the caps. The plaintiffs alleged that the failure of the blasting cap industry as a whole to place a warning on individual blasting caps created an unreasonable risk of harm which resulted in the plaintiffs’ injuries.

Despite the plaintiffs’ inability to identify the specific manufacturer of the blasting caps that caused their injuries, the Hall court held that the defendants were not entitled to a dismissal for plaintiffs’ failure to state a claim. The court noted that there was evidence that the individual defendants had adhered to industry-wide safety standards and had in effect delegated duties of safety investigation and design, such as labeling, to a trade association. The court reasoned that this evidence could support a conclusion that all the defendants jointly controlled the risk, and if such control were shown, the issue of who caused the injury would become secondary to the fact that the enterprise or industry as a whole engaged in joint, hazardous conduct. Under these circumstances, if plaintiffs could show by a preponderance of the evidence that the caps were manufactured by one of the joined defendants, the burden of proof as to causation would shift to the defendants, allowing a specific manufacturer to exculpate itself only if it could show that it did not manufacture the caps that injured the plaintiffs..

The theory of enterprise liability suggested in Hall is predicated largely upon the existence of industry-wide standards or practices adhered to by a group of defendants. When the standards can be shown to be deficient, identification of a specific defendant decreases in importance, since the standards themselves can be conceptualized as the primary causative agent. Each industry member contributes to the plantiff’s injury by adhering to, and therefore perpetuating, the standard, which results in the manufacture of the defective product.

The enterprise liability theory is distinguishable from the concerted action theory in that, under the latter theory, defendants cannot exculpate themselves by showing that they did not manufacture the injury-producing product. Indeed, the very purpose of the concerted action theory is to extend liability beyond manufacturers to those who render substantial encouragement or assistance to them. Also, under the enterprise liability theory, proof of an express agreement or “tacit understanding” is not required, as it is in the concert cases: the joint, parallel activities of the industry members in adhering to the safety standard substitutes for this requirement. Finally, unlike the concerted action theory, the enterprise liability theory was specifically formulated to aid plaintiffs in overcoming the difficulties posed by the identification requirement. This purpose becomes clear from an examination of the procedural history of the Hall case. When the plaintiffs in Hall amended their complaint to name a particular manufacturer alleged to have caused their injuries, the court held that “the amended complaint .. .does not preserve the joint [enterprise] liability aspects of the case.

The existence of industry-wide standards for drug manufacturing would seem to render the DES cases ripe for the application of the enterprise liability theory. For several reasons, however, this theory has generally not been successfully applied in the DES litigation to date . First, the promulgation of standards for drug manufacturing is largely uncontrolled by the drug manufacturers themselves. The Federal Food and Drug Administration regulates many phases in the testing, manufacturing and marketing of drugs, including the contents of warning labels. While adherence to FDA standards cannot insulate a manufacturer from liability, to the extent that compliance with these standards is compelled by the federal government it would be unfair to impose liability on a drug manufacturer simply for following the standards, when the plaintiffs are unable to identify that manufacturer’s product as the cause of their injuries.

A second factor militating against enterprise liability in DES litigation is the large number of defendants involved (between one hundred and three hundred). The Hall case involved only six manufacturers, representing the entire blasting cap industry. The court cautioned against applying the enterprise liability theory in cases involving large numbers of defendants, stating that what would be fair and feasible with regard to an industry of five or ten producers might be manifestly unreasonable if applied to a decentralized industry composed of thousands of small producers. As the number of manufacturers increases, joint control and awareness of the risks at issue, and joint capacity to reduce those risks, becomes more difficult to imply as a predicate to liability.

Finally, it should be emphasized that the enterprise liability theory, as explicated in Hall, does not allow a plaintiff to avoid the identification requirement in toto. The plaintiff is still required to show that at least one of the joined defendants manufactured the injury-producing product, which most plaintiffs in the DES cases simply cannot do.


The most promising theory advanced by DES plaintiffs in an effort to avoid the identification requirement has been termed “alternative liability.” This theory is based on the celebrated California case of Summers v. Tice. In Summers, two hunters negligently and simultaneously shot in the plaintiff’s direction, injuring his eye. Neither the plaintiff nor the defendants could determine which hunter had injured the plaintiff, and only one could have done so. The California Supreme Court held that when two negligent tortfeasors independently harm a plaintiff, and the plaintiff, through no personal fault, cannot prove which tortfeasor caused the harm, the burden of proof as to causation shifts to the defendants, who will be held jointly and severally liable unless able to prove that one defendant did not cause the plaintiff’s injury. This holding represented a policy determination that an innocent plaintiff should not go remediless simply because the plaintiff is faultlessly unable to determine which of two negligent tortfeasors caused the injury. The holding was also based on the presumption that under the circumstances in Summers, the defendants are often in a “far better position” to offer evidence on the causation issue. Section 433B(3) of the Restatement (Second) of Torts has codified the Summers rule.

As a means of avoiding the identification requirement in DES litigation, the alternative liability theory, as explicated in Summers, is ostensibly superior to the concert of action and enterprise liability theories. Unlike the concerted action theory, alternative liability was specifically formulated to aid the plaintiff in overcoming the identification requirement where its imposition would be unfair. Also, to prove concerted action, the plaintiff must show an express agreement or “tacit understanding” among the defendants: no such requirement exists under the alternative liability theory, which can impose joint and several liability on defendants who act independently. In addition, unlike the enterprise liability theory, alternative liability requires no showing that the defendants adhered to an inadequate industry-wide safety standard for which the industry itself bore primary responsibility.

The facts in Summers are superficially similar to the facts in the DES cases. As one commentator has noted:

Many of the elements of the Summers fact pattern are present in the DES cases. Defendants’ manufacturer of dangerous pills for the unwary public can be compared to the hunters shooting in the direction of their companion. In each situation, all defendants are tortfeasors owing a duty of care to the injured plaintiff. In both the DES cases and Summers, the tortious nature of each of the defendants’ conduct was identical and created the same type of risk. Neither the plaintiff in Summers hit by a bullet nor the DES daughter who developed cancer is at fault for being unable to identify the one who caused his injury.

Arguably, then, the Summers rule should operate in the DES cases to shift the burden of identification to the defendant manufacturers and to impose upon them joint and several liability if they fail to discharge that burden. As with the other theories plaintiffs have presented, however, numerous factors militate against the application of the alternative liability theory to DES litigation, particularly as that theory was developed in Summers.

In civil cases, a plaintiff must generally prove liability by a “preponderance of the evidence, interpreted as a mathematical probability of greater than fifty percent. Where the alternative liability theory is applied to a situation involving only two defendants, as in Summers, the probability of causation for each defendant is still only fifty percent, which is less than the civil standard. This is tolerated under the theory in light of the policy determination that, under certain circumstances, the requirement that a plaintiff show at least a fifty one percent likelihood that a particular defendant harmed him will be relaxed when the plaintiff can establish a one hundred percent probability (i.e., a certainty) that at least one of a group of tortfeasors harmed him.

The fairness of this policy is contingent upon two factors. First, the group of tortfeasors to whom the identification burden, and potential joint and several liability, is shifted must be small enough so that the probability that any one tortfeasor in fact injured the plaintiff does not become inequitably low. In Summers, only two tortfeasors were involved, with a fifty percent possibility of causation. Consider an analogous situation, however, involving ten tortfeasors. There, the probability that any one defendant injured the plaintiff diminishes to ten percent, considerably less than the civil standard. The precise point at which such probabilities diminish to where the inference of causation, and the imposition of joint and several liability, becomes inequitable is of course a debatable issue. In the DES litigation, however, the most conservative estimates of the number of tortious DES manufacturers hover around one hundred, which would establish a mere one percent probability of causation for each individual defendant. Clearly, this level of probability provides neither a rational nor an equitable basis upon which to infer causation or impose joint and several liability. In this circumstance, a defendant who had only a one percent probability of being the causal tortfeasor could potentially be held liable for the plaintiff’s entire damage assessment – a patently unfair application of the alternative liability theory.

The fairness of the policy underlying alternative liability is also contingent upon complete joinder of all tortious defendants. Complete joinder, which would ensure that the negligence of at least one of the tortfeasors was causal, constitutes the quid pro quo for relaxing the civil standard of proof of causation as applied to each single defendant. It also guarantees that the actual causal party will not escape liability.

In the DES cases, however, complete joinder is a practical impossibility; hence the application of alternative liability is again contraindicated. Even if all DES manufacturers could be joined, the presumption, implied by the alternate liability theory, that all defendants would possess an equal probability of being the causal party is dubious at best. Some manufacturers held a greater share of the DES market than others, and thus were more likely to have supplied the drug to a particular plaintiff. Under the alternative liability theory, however, market share would have no bearing upon the assessment of probabilities related to the identification issue. Neither would it bear upon the issue of damage assessment: in those states that provide for equal contribution among negligent defendants, each DES manufacturer would be equally liable for the plaintiff’s damages despite its share of the relevant market.

Finally, it should be noted that in the Summers case, from which the alternative liability theory was derived, the events took place simultaneously in a single location with the defendants in each other’s presence. This allowed them a reasonable opportunity to identify the causal party, which was doubtless a factor in the court’s decision to shift the identification burden. By contrast, the defendants’ conduct in the DES cases occurred over the entire United States over a period of years. This fact, in conjunction with the lack of documentary evidence of identification, virtually eliminates any way for DES defendants to exculpate themselves by inculpating other defendants. Given the basic wisdom of the policy underlying alternative liability, the defendants’ evidentiary problems should not prevent the shift of the identification burden to them, especially where innocent plaintiffs are faced with essentially the same problems. However, the Summers practice of both shifting the evidentiary burden and imposing joint and several liability raises the potential, in the DES cases, for damage apportionments grossly disproportionate to the defendant’s probability of causation. Primarily for this reason, the alternative liability theory, as explicated in Summers, has not been successfully applied in the DES litigation to date.


If the plaintiffs in the DES cases were confined to the above theories of liability, recovery would doubtless be precluded. Nonetheless, sound policy reasons exist for not denying recovery in DES litigation. Foremost among these is that advanced in Summers: as between innocent plaintiffs and negligent tortfeasors, the latter should bear the cost of injury. This policy factor is particularly applicable when, as in the DES cases, the plaintiffs are not at fault for their inability to obtain recovery under traditional tort doctrines.

Various economic criteria also support shifting the tort losses to the DES defendants. The DES manufacturers are best able to absorb these losses by distributing them among the public as a cost of doing business. The losses would be appropriately allocated, via increased prices, to that segment of the public that benefits most by the drug industry’s activities, that is, drug consumers. The drug industry is apparently in good financial health, and individual manufacturers are able to insure against drug-related losses.

The drug manufacturers are also in the best position to discover, warn against and prevent defects in their products. Drug consumers are virtually helpless to protect themselves from drug-related injuries and must rely almost wholly upon the manufacturers’ representations of safety. Shifting the losses to the DES defendants would also provide incentive for safer manufacturing practices.

The contradiction in the DES cases between the above policy factors supporting recovery, and the inadequacy of traditional tort doctrines in providing recovery, forces a “crisis” in the tort recovery system. The response of courts could be either to stick to traditional doctrines and deny recovery or to fashion new remedies that are more equitable given changed social, technological and economic conditions. In taking the latter course, Justice Traynor, in Escola v. Coca-Cola Bottling Co., urged the adoption of strict liability over traditional standards of negligence in product liability actions. The changed relationships between manufacturer and consumer which gave rise to his opinion are even more evident in the modern era. Referring specifically to the DES litigation, one commentator has noted that:

Technological advances and current market conditions now allow an entire industry to manufacture a complex fungible product; modern scientific research can link contact with this product to harmful effects after a significant lapse of time. Since these advances now make identification of the injury-producing product inaccessible to the consumer, themanufacturer’s obligation to the consumer can only be met by some new form of liability.

The California Supreme Court, in Sindell v. Abbott Laboratories, fashioned just such a “new form of liability,” termed market-share liability, as a means of avoiding the identification requirement in DES litigation. Under the market share liability theory, if a plaintiff who is unable to identify the manufacturer of the DES which injured her is able to join as defendants manufacturers that together account for a “substantial” market share of the DES her mother might have taken, the burden of proof as to which manufacturer actually supplied the injury-producing DES would be shifted to the defendants. A particular defendant could exculpate itself by showing that it could not have produced the DES to which the plaintiff was exposed. Each defendant that failed to do this would be held liable for a proportion of the plaintiff’s judgment corresponding to its percentage share of the market.

The Advantages of the Market Share Approach

Two practical advantages of the market share theory to DES plaintiffs, in addition to the elimination of the identification burden, are evident. First, the plaintiff need prove neither concerted action nor adherence to a defective industry-wide safety standard. Second, complete joinder, a necessity under the alternative liability theory, is not required under the market share approach.

Numerous theoretical advantages also accrue under the market share theory. This approach provides a much more accurate measure of the probability that a particular defendant caused the harm than the Summers alternative liability theory provides. Under the Summers theory, causation probability is treated purely as a mathematical function of the number of defendants joined; thus, where two defendants are present, the causation probability is fifty percent; with four defendants, twenty-five percent. The problem with this approach is that there is simply no rational connection between the number of defendants joined and the probability that any one defendant caused the plaintiff’s injury. There is clearly a rational connection, however, between market share and causation probability, at least in the context of the DES cases. Obviously, the more prevalent a particular manufacturer’s DES was in the marketplace, the more likely that a particular plaintiff ingested that manufacturer’s drug.

The market share approach also provides a more rational means of apportioning damages than the joint and several liability imposed by the other theories. Since each defendant is liable only for that proportion of the plaintiff’s damages corresponding to its percentage share of the DES market, each defendant’s damage assessment is more closely related to the probability that it caused the plaintiff’s harm. Curiously, under the market share approach, a particular defendant’s aggregate liability in the total number of DES cases is arguably the same as it would be if identification in all cases could be made. As one commentator explains:

If X Manufacturer sold one-fifth of all the DES prescribed for pregnancy and identification could be made in all cases, X would be the sole defendant in approximately one-fifth of all cases and liable for all the damages in those cases. Under [market share] liability, X would be joined in all cases in which identification could not be made, but liable for only one-fifth of the total damages in these cases. X would pay the same amount either way. Although the correlation is not, in practice, perfect, it is close enough so that defendants’ objections on the ground of fairness lose their value.

Delimiting the defendant’s liability via market share also provides a quid pro quo for holding the defendant liable where its probability of causation is less than the civil standard of proof, here defined as a mathematical probability of greater than fifty percent. Thus, for example, it becomes tolerable to hold a defendant liable where its probability of causation (that is, market share) is only ten percent – much lower than the civil standard – since its damage assessment will be correspondingly limited.

Problems with the Market Share Approach

Market share liability represents a novel and creative solution to the plaintiff’s identification problems in DES litigation. The theory, however, is subject to numerous criticisms, on both operational and public policy grounds.

Although under the market share approach joint and several liability is eliminated, the plaintiff is still able to recover one hundred percent of the damages. Since the plaintiff need join only defendants which together represent a substantial, rather than a complete, share of the DES market, each defendant may be held liable for a portion of the plaintiff’s damages that substantially exceeds the defendant’s market share. For example, assume that the plaintiff joined two defendants, each with thirty-five percent of the DES market. Assume further that their combined market share, seventy percent, is deemed “substantial” by the court, which results in the shift of the identification burden to the defendants. Since the plaintiff is entitled to full recovery, if the defendants cannot satisfy the identification burden, each will be liable for fifty percent of the plaintiff’s damages, rather than the thirty-five percent that represents their respective market share.

The discrepancy between the market share and damage apportionment percentages will vary according to the actual combined market shares present in individual cases: the discrepancy decreases as the aggregate share approaches one hundred percent and disappears at that point. Of course, the combined market shares present in individual cases will depend greatly upon how individual courts define a “substantial” share of the market since, presumably, a plaintiff will be motivated to join defendants only up to the point where a substantial market share is represented in her case. The Sindell court did not define the point at which combined market share percentages become “substantial,” a question with which trial courts will have to grapple in future DES litigation.

Difficulties also arise under the market share theory in the definition of the relevant DES market. Market share data may simply be unavailable, particularly for the time periods when plaintiffs’ mothers first ingested DES. Also, DES was used for various purposes other than to prevent miscarriage, and it may be difficult to determine what percentage of DES was sold for use by pregnant women. In addition, a troublesome question arises concerning the geographic scope of the relevant DES market: should it be restricted to a particular state, city or even pharmacy, where the plaintiff’s mother purchased, or might have purchased, DES? While recognizing these problems, the Sindell court did not indicate how they might be resolved or which party would have the burden of defining the relevant market.

Ironically, the market share approach may operate to treat plaintiffs who cannot identify the causal manufacturer more favorably than those who can. In the ordinary tort case where a single causal defendant is identified, the plaintiff’s damages will be recoverable solely from that defendant, who may or may not be solvent. This would presumably be the case where a particular DES defendant was identified as the causal party: the plaintiff’s damage recovery would be restricted to that defendant alone, and the plaintiff assumes the risk that the defendant can respond financially. Where the plaintiff cannot identify the causal manufacturer, however, the market share theory literally imposes industry-wide liability, substantially reducing the risk that the plaintiff will be without a solvent defendant. The theory therefore operates to “punish” those plaintiffs who are able to satisfy traditional tort doctrines by identifying the causal manufacturer. The theory also operates to expose defendants to double liability, once to plaintiffs who can identify them as the causal party, and once again to plaintiffs who cannot.

The dissent in the Sindell case felt that the market share theory represented an unwarranted and unprecedented extension of liability. As Justice Richardson stated, under the market share theory,

recovery is permitted from a handful of defendants each of whom individually may account for a comparatively small share of the relevant market, so long as the aggregate business of those who have been sued is deemed “substantial.” In other words, a particular defendant may be held proportionately liable even though mathematically it is much more likely than not that it played no role whatever in causing plaintiff’s injuries …. The majority rejects over 100 years of tort law which required that before tort liability was imposed a “matching” of defendant’s conduct and plaintiff’s injury was absolutely essential.

The dissent additionally claimed that

market share” liability… represents a new high water mark in tort law. The ramifications seem almost limitless, a fact which prompted one recent commentator, in criticizing a substantially identical theory, to conclude that “elimination of the burden of proof as to identification [of the manufacturer whose drug injured plaintiff would impose a liability which would exceed absolute liability“.

Courts have expressed concern that changes in the tort law might overly enlarge a manufacturer’s duties and extend liability too far. The Defense Research Institute has suggested that product liability law can continue to function effectively only if present theories of recovery are not extended further. The Institute notes that unmitigated extensions of liability for defective products may have an effect opposite to that intended: instead of improving consumer protection and making damage reparations more certain, recovery will be diminished because of lack of money. Extending the liability of drug manufacturers may have the additional undesirable effect of “chilling” research and development efforts channeled toward the development of new or experimental drugs. The recent enactments of product liability statutes restricting the scope of a manufacturer’s liability 16 may be a response to the fear that product liability has been, or will be, extended too far.


The criticisms leveled at the market share theory have merit and deserve attention. They are insufficient, however, to impeach the basic soundness of this approach, at least as applied to the DES line of cases.

Uncertainties in the determination of market share, and discrepancies between market share and liability, are indeed inevitable under this theory; however, lack of complete precision in the determination of liability is a problem that is ubiquitous in the law of torts and hardly restricted to the market share theory. For example, juries cannot be expected to precisely correlate fault and liability in applying the doctrine of comparative negligence or partial indemnity. As the court stated in Summers, where a perfect division of liability among tortfeasors cannot be made, the trier of fact may make it the best it can.

Fears that the market share theory will generally lead to counterproductive extensions of product liability ignore the unique factual circumstances out of which the theory arose and to which it was narrowly adapted. In the DES cases, all the defendants were shown to have been negligent; market share data was used solely to apportion liability, not to prove negligence. Further, the plaintiffs’ injuries were uniquely traceable to a single product, rendering market share a reasonably good estimate of the harm done by individual manufacturers. More importantly, plaintiffs were injured by a product which masked its harmful effects for many years, making identification of the causal tortfeasor impossible. Courts can and should use great caution in extending the application of market share liability to cases where these distinguishing factual circumstances are absent.

Minor modifications in the market share theory could go far toward ameliorating the remaining inequities that exist in its application. One possible modification would be to limit a defendant’s liability to its exact percentage share of the market, which would eliminate the discrepancies between market share and damage apportionment percentages that result under the “pure” form of the theory. The defendant’s damage liability would thus be maximally correlated with its probability of causation. Of course, plaintiffs would no longer be guaranteed recovery of one hundred percent of their damages, since recovery would be restricted to that percentage of plaintiff’s damages that corresponded to the aggregate market share of all defendants joined. Clearly, however, this would motivate plaintiffs to join as many defendants as possible, which would help assure that the plaintiff’s damages would be more equitably distributed among all tortious defendants. Another possible modification in the market share theory would be to broaden its applicability to plaintiffs who can identify the causal DES manufacturer. …

Randy S. Parlee, Summer 1982 .

Click to download the full study.

More DES DiEthylStilbestrol Resources

Have your say! Share your views