Hymowitz v. Eli Lilly & Co. , 73 N.Y.2d 487 ( 1989 )

OPINION OF THE COURT

Chief Judge Wachtler.

Plaintiffs in these appeals allege that they were injured by the drug diethylstilbestrol (DES) ingested by their mothers during pregnancy. They seek relief against defendant DES manufacturers. While not class actions, these cases are representative of nearly 500 similar actions pending in the courts in this State; the rules articulated by the court here, therefore, must do justice and be administratively feasible in the context of this mass litigation. With this in mind, we now resolve the issue twice expressly left open by this court, and adopt a market share theory, using a national market, for determining liability and apportioning damages in DES cases in which identification of the manufacturer of the drug that injured the plaintiff is impossible (see, Kaufman v Lilly & Co., 65 NY2d 449, 456; Bichler v Lilly & Co., 55 NY2d 571, 580). We also hold that the Legislature’s revival for one year of actions for injuries caused by DES that were previously barred by the Statute of Limitations (see, L 1986, ch 682, § 4) is constitutional under the State and Federal Constitutions.

I.

The history of the development of DES and its marketing in this country has been repeatedly chronicled (see, e.g., Bichler v Lilly & Co., supra; Martin v Abbott Labs., 102 Wash 2d 581, 689 P2d 368; Sindell v Abbott Labs., 26 Cal 3d 588, 607 P2d 924, cert denied 449 US 912; Sheiner, DES and a Proposed Theory of Enterprise Liability, 46 Fordham L Rev 963). Briefly, DES is a synthetic substance that mimics the effect of estrogen, the naturally formed female hormone. It was invented in 1937 by British researchers, but never patented.

In 1941, the Food and Drug Administration (FDA) approved the new drug applications (NDA) of 12 manufacturers to market DES for the treatment of various maladies, not directly involving pregnancy. In 1947, the FDA began approving the NDAs of manufacturers to market DES for the purpose of preventing human miscarriages; by 1951, the FDA had concluded that DES was generally safe for pregnancy use, and stopped requiring the filing of NDAs when new manufacturers sought to produce the drug for this purpose. In 1971, however, the FDA banned the use of DES as a miscarriage preventative, when studies established the harmful latent effects of DES upon the offspring of mothers who took the drug. Specifically, tests indicated that DES caused vaginal adenocarcinoma, a form of cancer, and adenosis, a precancerous vaginal or cervical growth.

Although strong evidence links prenatal DES exposure to later development of serious medical problems, plaintiffs seeking relief in court for their injuries faced two formidable and fundamental barriers to recovery in this State; not only is identification of the manufacturer of the DES ingested in a particular case generally impossible, but, due to the latent nature of DES injuries, many claims were barred by the Statute of Limitations before the injury was discovered.

The identification problem has many causes. All DES was of identical chemical composition. Druggists usually filled prescriptions from whatever was on hand. Approximately 300 manufacturers produced the drug, with companies entering and leaving the market continuously during the 24 years that DES was sold for pregnancy use. The long latency period of a DES injury compounds the identification problem; memories fade, records are lost or destroyed, and witnesses die. Thus the pregnant women who took DES generally never knew who produced the drug they took, and there was no reason to attempt to discover this fact until many years after ingestion, at which time the information is not available.

We recognized this predicament in Bichler v Lilly & Co. (supra, at 579), where the court stated that in DES cases it is a "practical impossibility for most victims [to] pinpoint * * * the manufacturer directly responsible for their particular injury”. We allowed plaintiff’s recovery in that case, however, notwithstanding the failure of the plaintiff to identify the manufacturer of the injurious DES, on the limited basis that "the evidence was legally sufficient to support the jury verdict for the plaintiff” on the law as charged to the jury, and unobjected to by the defendant (see, Kaufman v Lilly & Co., 65 NY2d 449, 456, supra). The question, therefore, of whether nonidentification of the manufacturer precludes plaintiffs from recovering for DES caused injuries, remained unresolved after Bichler v Lilly & Co. (supra).

The second barrier to recovery, involving the Statute of Limitations, arose from the long-standing rule in this State that the limitations period accrued upon exposure in actions alleging personal injury caused by toxic substances (Fleishman v Lilly & Co., 62 NY2d 888, cert denied 469 US 1192). In Fleishman v Lilly & Co. (supra) it became clear that this exposure rule led to many DES cases being barred by the Statute of Limitations before the discovery of injury; we held, however, that any change in the accrual date from exposure to discovery was more properly the prerogative of the Legislature (id., at 890; see, id., at 891 [Cooke, Ch. J., dissenting]). Two years after Fleishman v Lilly & Co. the Legislature addressed the Statute of Limitations problem, and instituted a discovery rule for "the latent effects of exposure to any substance” (L 1986, ch 682, § 2). The Legislature also, for one year, revived causes of action for exposure to DES that had been time barred (L 1986, ch 682, § 4).

It is estimated that eventually 800 DES cases will be brought under the revival portion of this recent statute. Moreover, as indicated in Bichler v Lilly & Co. (supra), and as apparent from the record now before the court, in the vast majority of these cases identification of the manufacturer of the DES that injured the plaintiff will be impossible. The Legislature, however, while reviving these time-barred actions, did not resolve the identification problem.

The present appeals are before the court in the context of summary judgment motions. In all of the appeals defendants moved for summary judgment dismissing the complaints because plaintiffs could not identify the manufacturer of the drug that allegedly injured them. In three of the appeals defendants also moved on Statute of Limitations grounds, arguing that the revival of the actions was unconstitutional under the State and Federal Constitutions, and that the complaints, therefore, are time barred and should be dismissed. The trial court denied all of these motions. On the Statute of Limitations issue, the trial court also granted plaintiffs’ cross motions, dismissing defendants’ affirmative defenses that the actions were time barred. The Appellate Division affirmed in all respects and certified to this court the questions of whether the orders of the trial court were properly made. We answer these questions in the affirmative.

II.

In a products liability action, identification of the exact defendant whose product injured the plaintiff is, of course, generally required (see, e.g., Morrissey v Conservative Gas Corp., 285 App Div 825, affd 1 NY2d 741; Prosser and Keeton, Torts § 103, at 713 [5th ed]). In DES cases in which such identification is possible, actions may proceed under established principles of products liability (Bichler v Lilly & Co., supra, at 579). The record now before us, however, presents the question of whether a DES plaintiff may recover against a DES manufacturer when identification of the producer of the specific drug that caused the injury is impossible.

A.

As we noted in Bichler v Lilly & Co. (supra, at 580, n 5), the accepted tort doctrines of alternative liability and concerted action are available in some personal injury cases to permit recovery where the precise identification of a wrongdoer is impossible. However, we agree with the near unanimous views of the high State courts that have considered the matter that these doctrines in their unaltered common-law forms do not permit recovery in DES cases (see, e.g., Sindell v Abbott Labs., supra; Collins v Lilly & Co., 116 Wis 2d 166, 342 NW2d 37; Martin v Abbott Labs., supra; but see, Abel v Lilly & Co., 418 Mich 311, 343 NW2d 164 [held that there was a question of fact presented as to alternative liability and concerted action]).

The paradigm of alternative liability is found in the case of Summers v Tice (33 Cal 2d 80, 199 P2d 1). In Summers (supra), plaintiff and the two defendants were hunting, and defendants carried identical shotguns and ammunition. During the hunt, defendants shot simultaneously at the same bird, and plaintiff was struck by bird shot from one of the defendants’ guns. The court held that where two defendants breach a duty to the plaintiff, but there is uncertainty regarding which one caused the injury, "the burden is upon each such actor to prove that he has not caused the harm” (Restatement [Second] of Torts § 433B [3]; Bichler v Lilly & Co., supra, at 580, n 5; cf., Ravo v Rogatnick, 70 NY2d 305 [successive tort-feasors may be held jointly and severally liable for an indivisible injury to the plaintiff]). The central rationale for shifting the burden of proof in such a situation is that without this device both defendants will be silent, and plaintiff will not recover; with alternative liability, however, defendants will be forced to speak, and reveal the culpable party, or else be held jointly and severally liable themselves. Consequently, use of the alternative liability doctrine generally requires that the defendants have better access to information than does the plaintiff, and that all possible tort-feasors be before the court (see, Summers v Tice, supra, at 86; Restatement [Second] of Torts § 433B, comment h). It is also recognized that alternative liability rests on the notion that where there is a small number of possible wrongdoers, all of whom breached a duty to the plaintiff, the likelihood that any one of them injured the plaintiff is relatively high, so that forcing them to exonerate themselves, or be held liable, is not unfair (see, Sindell v Abbott Labs., supra, at 603).

In DES cases, however, there is a great number of possible wrongdoers, who entered and left the market at different times, and some of whom no longer exist. Additionally, in DES cases many years elapse between the ingestion of the drug and injury. Consequently, DES defendants are not in any better position than are plaintiffs to identify the manufacturer of the DES ingested in any given case, nor is there any real prospect of having all the possible producers before the court. Finally, while it may be fair to employ alternative liability in cases involving only a small number of potential wrongdoers, that fairness disappears with the decreasing probability that any one of the defendants actually caused the injury. This is particularly true when applied to DES where the chance that a particular producer caused the injury is often very remote (Sindell v Abbott Labs., supra, at 603; Collins v Lilly & Co., supra, at 184). Alternative liability, therefore, provides DES plaintiffs no relief.

Nor does the theory of concerted action, in its pure form, supply a basis for recovery. This doctrine, seen in drag racing cases, provides for joint and several liability on the part of all defendants having an understanding, express or tacit, to participate in "a common plan or design to commit a tortious act” (Prosser and Keeton, Torts §46, at 323 [5th ed]; see, Bichler v Lilly & Co., supra, at 580-581; De Carvalho v Brunner, 223 NY 284). As we noted in Bichler v Lilly & Co., and as the present record reflects, drug companies were engaged in extensive parallel conduct in developing and marketing DES (see, id., at 585). There is nothing in the record, however, beyond this similar conduct to show any agreement, tacit or otherwise, to market DES for pregnancy use without taking proper steps to ensure the drug’s safety. Parallel activity, without more, is insufficient to establish the agreement element necessary to maintain a concerted action claim (Sindell v Abbott Labs., supra, at 605; Collins v Lilly & Co., supra, at 185; Martin v Abbott Labs., supra, at 599). Thus this theory also fails in supporting an action by DES plaintiffs.

In short, extant common-law doctrines, unmodified, provide no relief for the DES plaintiff unable to identify the manufacturer of the drug that injured her. This is not a novel conclusion; in the last decade a number of courts in other jurisdictions also have concluded that present theories do not support a cause of action in DES cases. Some courts, upon reaching this conclusion, have declined to find any judicial remedy for the DES plaintiffs who cannot identify the particular manufacturer of the DES ingested by their mothers (see, Zafft v Lilly & Co., 676 SW2d 241 [Mo] [en banc]; Mulcahy v Lilly & Co., 386 NW2d 67 [Iowa] [stating that any change in the law to allow for recovery in nonidentification DES cases should come from the Legislature]). Other courts, however, have found that some modification of existing doctrine is appropriate to allow for relief for those injured by DES of unknown manufacture (e.g., Sindell v Abbott Labs., supra; Collins v Lilly & Co., supra; Martin v Abbott Labs., supra).

We conclude that the present circumstances call for recognition of a realistic avenue of relief for plaintiffs injured by DES. These appeals present many of the same considerations that have prompted this court in the past to modify the rules of personal injury liability, in order "to achieve the ends of justice in a more modern context” (see, People v Hobson, 39 NY2d 479, 489; Codling v Paglia, 32 NY2d 330, 341), and we perceive that here judicial action is again required to overcome the " 'inordinately difficult problems of proof ” caused by contemporary products and marketing techniques (see, Bichler v Lilly & Co., supra, at 579-580 [quoting Caprara v Chrysler Corp., 52 NY2d 114, 123]).

Indeed, it would be inconsistent with the reasonable expectations of a modern society to say to these plaintiffs that because of the insidious nature of an injury that long remains dormant, and because so many manufacturers, each behind a curtain, contributed to the devastation, the cost of injury should be borne by the innocent and not the wrongdoers. This is particularly so where the Legislature consciously created these expectations by reviving hundreds of DES cases. Consequently, the ever-evolving dictates of justice and fairness, which are the heart of our common-law system, require formation of a remedy for injuries caused by DES (see, Woods v Lancet, 303 NY 349, 355; see, also, Kaye, The Human Dimension in Appellate Judging: A Brief Reflection on a Timeless Concern, 73 Cornell L Rev 1004).

We stress, however, that the DES situation is a singular case, with manufacturers acting in a parallel manner to produce an identical, genetically marketed product, which causes injury many years later, and which has evoked a legislative response reviving previously barred actions. Given this unusual scenario, it is more appropriate that the loss be borne by those that produced the drug for use during pregnancy, rather than by those who were injured by the use, even where the precise manufacturer of the drug cannot be identified in a particular action. We turn then to the question of how to fairly and equitably apportion the loss occasioned by DES, in a case where the exact manufacturer of the drug that caused the injury is unknown.

B.

The past decade of DES litigation has produced a number of alternative approaches to resolve this question. Thus, in a sense, we are now in an enviable position; the efforts of other courts provided examples for contending with this difficult issue, and enough time has passed so that the actual administration and real effects of these solutions now can be observed. With these useful guides in hand, a path may be struck for our own conclusion.

First, this court’s opinion in Bichler v Lilly & Co. (supra) must be considered. There the jury was instructed on a modified version of concerted action, which, in effect, substituted the fact of conscious parallel activity by manufacturers for the usual common-law requirement that there be proof of an actual agreement between actors to jointly act tortiously (id., at 584). The defendant in Bichler did not object to this instruction, and the modified concerted action theory became the law applicable to that particular case (id., at 583-584).

Now given the opportunity to assess the merits of this theory, we decline to adopt it as the law of this State. Parallel behavior, the major justification for visiting liability caused by the product of one manufacturer upon the head of another under this analysis, is a common occurrence in industry generally. We believe, therefore, that inferring agreement from the fact of parallel activity alone improperly expands the concept of concerted action beyond a rational or fair limit; among other things, it potentially renders small manufacturers, in the case of DES and in countless other industries, jointly liable for all damages stemming from the defective products of an entire industry (accord, Sindell v Abbott Labs., supra, at 605).

A narrower basis for liability, tailored more closely to the varying culpableness of individual DES producers, is the market share concept. First judicially articulated by the California Supreme Court in Sindell v Abbott Labs, (supra), variations upon this theme have been adopted by other courts (see, Collins v Lilly & Co., supra; Martin v Abbott Labs., supra). In Sindell v Abbott Labs (supra), the court synthesized the market share concept by modifying the Summers v Tice (supra) alternative liability rationale in two ways. It first loosened the requirement that all possible wrongdoers be before the court, arid instead made a "substantial share” sufficient. The court then held that each defendant who could not prove that it did not actually injure plaintiff would be liable according to that manufacturer’s market share. The court’s central justification for adopting this approach was its belief that limiting a defendant’s liability to its market share will result, over the run of cases, in liability on the part of a defendant roughly equal to the injuries the defendant actually caused (id., at 612).

In the recent case of Brown v Superior Ct. (44 Cal 3d 1049, 751 P2d 470), the California Supreme Court resolved some apparent ambiguity in Sindell v Abbott Labs., and held that a manufacturer’s liability is several only, and, in cases in which all manufacturers in the market áre not joined for any reason, liability will still be limited to market share, resulting in a less than 100% recovery for a plaintiff. Finally, it is noteworthy that determining market shares under Sindell v Abbott Labs, proved difficult and engendered years of litigation. After attempts at using smaller geographical units, it was eventually determined that the national market provided the most feasible and fair solution, and this national market information was compiled (see, In re Complex DES Litig., No. 830/109, Cal Super Ct).

Four years after Sindell v Abbott Labs., the Wisconsin Supreme Court followed with Collins v Lilly & Co. (116 Wis 2d 166, 342 NW2d 37, supra). Deciding the identification issue without the benefit of the extensive California litigation over market shares, the Wisconsin court held that it was prevented from following Sindell due to "the practical difficulty of defining and proving market share” (id., at 189, 342 NW2d, at 48). Instead of focusing on tying liability closely to the odds of actual causation, as the Sindell court attempted, the Collins court took a broader perspective, and held that each defendant is liable in proportion to the amount of risk it created that the plaintiff would be injured by DES. Under the Collins structure, the "risk” each defendant is liable for is a question of fact in each case, with market shares being relevant to this determination (id., at 191, 200). Defendants are allowed, however, to exculpate themselves by showing that their product could not have caused the injury to the particular plaintiff (id., at 198).

The Washington Supreme Court, writing soon after Collins v Lilly & Co., took yet another approach (see, Martin v Abbott Labs., 102 Wash 2d 581, 689 P2d 368, supra). The Martin court first rejected the Sindell market share theory due to the belief (which later proved to be erroneous in Brown v Superior Ct. [supra]) that California’s approach distorted liability by inflating market shares to ensure plaintiffs of full recovery (id., at 601). The Martin court instead adopted what it termed "market share alternative liability,” justified, it concluded, because "[e]ach defendant contributed to the risk of injury to the public, and, consequently, the risk of injury to individual plaintiffs” (id., at 604, 689 P2d, at 382).

Under the Washington scheme, defendants are first allowed to exculpate themselves by proving by the preponderance of the evidence that they were not the manufacturer of the DES that injured plaintiff. Unexculpated defendants are presumed to have equal market shares, totaling 100%. Each defendant then has the opportunity to rebut this presumption by showing that its actual market share was less than presumed. If any defendants succeed in rebutting this presumption, the liability shares of the remaining defendants who could not prove their actual market share are inflated, so that the plaintiff received a 100% recovery (id., at 605-606).¹ The market shares of defendants is a question of fact in each case, and the relevant market can be a particular pharmacy, or county, or State, or even the country, depending upon the circumstances the case presents (George v Parke-Davis, 107 Wash 2d 584, 733 P2d 507).

Turning to the structure to be adopted in New York, we heed both the lessons learned through experience in other jurisdictions and the realities of the mass litigation of DES claims in this State. Balancing these considerations, we are led to the conclusion that a market share theory, based upon a national market, provides the best solution. As California discovered, the reliable determination of any market smaller than the national one likely is not practicable. Moreover, even if it were possible, of the hundreds of cases in the New York courts, without a doubt there are many in which the DES that allegedly caused injury was ingested in another State. Among the thorny issues this could present, perhaps the most daunting is the spectre that the particular case could require the establishment of a separate market share matrix. We feel that this is an unfair, and perhaps impossible burden to routinely place upon the litigants in individual cases.

Nor do we believe that the Wisconsin approach of assessing the “risk” each defendant caused a particular plaintiff, to be litigated anew as a question of fact in each case, is the best solution for this State. Applied on a limited scale this theory may be feasible, and certainly is the most refined approach by allowing a more thorough consideration of how each defendant’s actions threatened the plaintiff. We are wary, however, of setting loose, for application in the hundreds of cases pending in this State, a theory which requires the fact finder’s individualized and open-ended assessment of the relative liabilities of scores of defendants in every case. Instead, it is our perception that the injustices arising from delayed recoveries and inconsistent results which this theory may produce in this State outweigh arguments calling for its adoption.

Consequently, for essentially practical reasons, we adopt a market share theory using a national market. We are aware that the adoption of a national market will likely result in a disproportion between the liability of individual manufacturers and the actual injuries each manufacturer caused in this State. Thus our market share theory cannot be founded upon the belief that, over the run of cases, liability will approximate causation in this State (see, Sindell v Abbott Labs., supra, at 612). Nor does the use of a national market provide a reasonable link between liability and the risk created by a defendant to a particular plaintiff (see, Collins v Lilly & Co., supra; Martin v Abbott Labs., supra). Instead, we choose to apportion liability so as to correspond to the over-all culpability of each defendant, measured by the amount of risk of injury each defendant created to the public-at-large. Use of a national market is a fair method, we believe, of apportioning defendants’ liabilities according to their total culpability in marketing DES for use during pregnancy. Under the circumstances, this is an equitable way to provide plaintiffs with the relief they deserve, while also rationally distributing the responsibility for plaintiffs’ injuries among defendants.

To be sure, a defendant cannot be held liable if it did not participate in the marketing of DES for pregnancy use; if a DES producer satisfies its burden of proof of showing that it was not a member of the market of DES sold for pregnancy use, disallowing exculpation would be unfair and unjust. Nevertheless, because liability here is based on the over-all risk produced, and not causation in a single case, there should be no exculpation of a defendant who, although a member of the market producing DES for pregnancy use, appears not to have caused a particular plaintiff’s injury. It is merely a windfall for a producer to escape liability solely because it manufactured a more identifiable pill, or sold only to certain drugstores. These fortuities in no way diminish the culpability of a defendant for marketing the product, which is the basis of liability here.²

Finally, we hold that the liability of DES producers is several only, and should not be inflated when all participants in the market are not before the court in a particular case. We understand that, as a practical matter, this will prevent some plaintiffs from recovering 100% of their damages. However, we eschewed exculpation to prevent the fortuitous avoidance of liability, and thus, equitably, we decline to unleash the same forces to increase a defendant’s liability beyond its fair share of responsibility.³

III.

The constitutionality of the revival statute remains to be considered (see, L 1986, ch 682, § 4). This section revives, for the period of one year, actions for damages caused by the latent effects of DES, tungsten-carbide, asbestos, chlordane, and polyvinylchloride. Defendants argue that the revival of barred DES claims was unconstitutional as a denial of both due process and equal protection, under the State and Federal Constitutions (see, NY Const, art I, §§ 6, 11; US Const, 14th Amend, § 1). We are concerned here only with the constitutionality of the statute as it pertains to DES; there are no producers of the other substances, or plaintiffs alleging injury therefrom, before the court on these appeals.

The Federal Due Process Clause provides very little barrier to a State Legislature’s revival of time-barred actions (see, Chase Sec. Corp. v Donaldson, 325 US 304). In Chase, the United States Supreme Court upheld the revival of a time-barred action, stating that Statutes of Limitation "represent a public policy about the privilege to litigate * * * the history of pleas of limitation shows them to be good only by legislative grace and to be subject to a relatively large degree of legislative control” (id., at 314). Under State law, the level of review seems somewhat more stringent. In Gallewski v Hentz & Co. (301 NY 164, 174) the court stated that "the Legislature may constitutionally revive a personal cause of action where the circumstances are exceptional and are such as to satisfy the court that serious injustice would result to plaintiffs not guilty of any fault if the intention of the Legislature were not effectuated.” It appears, however, that we have applied a less strict test in other cases, and have been satisfied if there was an apparent injustice which "calls for [a] remedy,” and which is "reasonable” and not "arbitrary.” (Robinson v Robins Dry Dock & Repair Co., 238 NY 271, 279-280.)

We need not light upon a precise test here, however, because the Legislature’s revival of DES claims meets the highest standard. For at least 25 years this court maintained an exposure rule for toxic substances, because it was felt that change in this area was the responsibility of the Legislature (see, e.g., Schwartz v Heyden Newport Chem. Corp., 12 NY2d 212, 220; Fleishman v Lilly & Co., supra, at 890). Indeed, in Fleishman v Lilly & Co. (supra) the Legislature’s attention was drawn specifically to DES by the majority, which stated that any change in the exposure rule was the Legislature’s role.

The Legislature has now revived DES actions that were time barred under the exposure rule, while also instituting a discovery rule for future application (L 1986, ch 682, § 4; CPLR 214-c). The latent nature of DES injuries is well known, and it is clear that in the past the exposure rule prevented the bringing of timely actions for recovery. Thus we believe that exceptional circumstances are presented, that an injustice has been rectified, and that the requirements of Gallewski v Hentz & Co. (supra) have been met.

Defendants argue further that, even if the statute is generally valid, it may be unconstitutionally applied in cases in which the plaintiff could have sued originally, but did not. It does seem that some plaintiffs may have known of their injuries a day, or a week, a month, or perhaps longer, before the original limitations period ran. Some may have known of their exposure, but did not develop injuries during the limitations period. Others may have known of some effect upon them of DES exposure, which became cancerous only after any action would have been time barred. Under these circumstances, the Legislature properly determined that it would be more fair for all plaintiffs to uniformly now have one year to bring their actions, rather than for the courts to begin drawing arbitrary lines transecting this area’s shades of gray.

Defendants also argue that the revival statute violates equal protection, because the Legislature designated only five substances for revival, including DES, while instituting a prospective only discovery rule for other substances. Defendants claim that this categorization is without sufficient basis, and that it is the result of a "political compromise.” But most, if not all legislation is the product of some compromise, so that this objection surely is no basis for finding the revival statute unconstitutional. Instead, here we must proceed on the presumption that the law is constitutional, and will hold otherwise only if it is established that the distinction drawn has no reasonable basis (Trump v Chu, 65 NY2d 20; Montgomery v Daniels, 38 NY2d 41, 61). Moreover, because defendants allege no impairment of a fundamental right, the Legislature has substantial leeway in making classifications in this area of " 'economics and social welfare’ ” (see, Montgomery v Daniels, supra, at 61 [quoting Dandridge v Williams, 397 US 471, 485]).

As it pertains to DES, surely the revival statute has a rational basis, and the Legislature acted within its broad range of discretion in enacting the law. The number of DES-caused injuries was relatively well known by the Legislature, which allowed for the ramifications of revival of DES claims, such as the effect on insurance interests, and the other costs, to be reasonably predicted (Record of Proceedings, New York Assembly, June 24, 1986 [statement of Assemblyman M. H. Miller]). Furthermore, it was also well known, particularly after Fleishman v Lilly & Co. (supra), that DES victims were prejudiced under current law. This, we believe, is enough of a basis for the Legislature to revive DES claims now, and wait as to other substances until it is felt that these substances present a problem suitable for resolution. The Legislature does not violate equal protection by providing a rational piecemeal remedy for what may be a larger problem (Williamson v Lee Opt. Co., 348 US 483).

Accordingly, in each case the order of the Appellate Division should be affirmed, with costs, and the certified question answered in the affirmative.

1 . The actual operation of this theory proved more mathematically complex when the court was presented with the question of what to do about unavailable defendants. Recognizing that the possibility of abuse existed when defendants implead unavailable defendants, who would then be assumed to have had an equal share of the market, the court placed the burden upon appearing defendants to prove the market share of the absent ones (George v Parke-Davis, 107 Wash 2d 584, 733 P2d 507). If this can be proved, the plaintiff simply cannot recover the amount attributable to the absent defendant, and thus recovery in the case is less than 100%. If the market share of the absent defendant cannot be shown, the remaining defendants who cannot prove their market shares have their shares inflated to provide plaintiff with full recovery. Finally, if all appearing defendants can prove their market shares, their shares are never inflated, regardless of whether the market share of a nonappearing defendant can be proved or not; thus, in this situation, the plaintiff again will not recover her full damages (id.).

2 . Various defendants argue here that although they produced DES, it was not sold for pregnancy use. If a defendant was not a member of the national market of DES marketed for pregnancy, it is not culpable, and should not be liable. Consequently, if a particular defendant sold DES in a form unsuitable for use during pregnancy, or if a defendant establishes that its product was not marketed for pregnancy use, there should be no liability. From the record before the court here, however, the facts are not developed well enough to establish that any defendants were not in the national market of DES sold for pregnancy use. Thus summary judgment cannot at this time be granted on this issue as to any defendants.

3 . The dissenter misapprehends the basis for liability here. We have not by the backdoor adopted a theory of concerted action. We avoided extending this theory, because its concomitant requirement of joint and several liability expands the burden on small manufacturers beyond a rational or fair limit. This result is reached by the dissent, not by the majority, so that criticism on this front is misplaced.

We are confronted here with an unprecedented identification problem, and have provided a solution that rationally apportions liability. We have heeded the practical lessons learned by other jurisdictions, resulting in our adoption of a national market theory with full knowledge that it concedes the lack of a logical link between liability and causation in a single case. The dissent ignores these lessons, and, endeavoring to articulate a theory it perceives to be closer to traditional law, sets out a construct in which liability is based upon chance, not upon the fair assessment of the acts of defendants. Under the dissent’s theory, a manufacturer with a large market share may avoid liability in many cases just because it manufactured a memorably shaped pill. Conversely, a small manufacturer can be held jointly liable for the full amount of every DES injury in this State simply because the shape of its product was not remarkable, even though the odds, realistically, are exceedingly long that the small manufacturer caused the injury in any one particular case.

Therefore, although the dissent’s theory based upon a "shifting the burden of proof’ and joint and several liability is facially reminiscent of prior law, in the case of DES it is nothing more than advocating that bare fortuity be the test for liability. When faced with the novel identification problem posed by DES cases, it is preferable to adopt a new theory that apportions fault rationally, rather than to contort extant doctrines beyond the point at which they provide a sound premise for determining liability.