Regulatory compliance preclusion of tort liability: Limiting the dual-track system

Georgetown Law Journal

Volumn 88, Issue 7, 2000, Pages 2167-2186

  Stewart, Richard B ^a  

^a NEW YORK UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

FOOD AND DRUG ADMINISTRATION; HUMAN; LEGAL ASPECT; LEGAL LIABILITY; MEDICAL TECHNOLOGY; NOTE; RISK MANAGEMENT; SAFETY; UNITED STATES;

CONSUMER PRODUCT SAFETY; HUMANS; LIABILITY, LEGAL; RISK MANAGEMENT; TECHNOLOGY, MEDICAL; UNITED STATES; UNITED STATES FOOD AND DRUG ADMINISTRATION;

EID: 0034350301     PISSN: 00168092     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (98)

1. AMERICAN LAW INSTITUTE, REPORTER'S STUDY, ENTERPRISE RESPONSIBILITY FOR PERSONAL INJURY (1991) [hereinafter ALI STUDY & 2 ALI STUDY]. The author was the Chief Reporter of the Enterprise Responsibility for Personal Injury project until 1989. Professor Paul Weiler thereupon became Chief Reporter and saw the project to a successful completion. Professor Rabin was an Associate Reporter for the project.
2. Rabin states in his conclusion that "a cautious approach to adopting the regulatory compliance defense is arguably well-taken." Robert L. Rabin, Reassessing Regulatory Compliance, 88 GEO. L.J. 2049, 2084 (2000). His discussion focuses on a regulatory compliance defense, but I think it is fair to conclude that his arguments against a regulatory compliance defense would, in general, apply also to legislation providing that regulatory compliance preempts tort liability.
3. See Rabin, supra note 2, at 2083-84 (discussing Ramirez v. Plough, 863 P.2d 167 (Cal. 1993)).
4. The ALI Study elaborated this requirement as follows: Under a system of regulatory screening, a risk or category of risks is placed under regulatory control after accurate up-to-date data on such risks are provided to the responsible agency by the enterprise or otherwise obtained by the agency; the data and risks are evaluated by the agency in accordance with authoritative criteria; clearance is granted upon a reasoned determination that the risk is acceptable; and there is an ongoing system of agency monitoring and review in place to deal with new information or changed circumstances. Under a system of regulatory standards the criteria are essentially the same, except that the standards adopted must be intended to limit and must limit, directly or indirectly, the amount of such risks that may be generated. 2 ALI STUDY, supra note 1, at 96 (footnotes omitted).
5. With respect to this requirement the ALI Study further stated: Some regulatory statutes, such as the Federal Food, Drug, and Cosmetic Act, impose such reporting and disclosure requirements. But whether or not such requirements have been imposed by statute, the regulatory compliance defense would not be available as a matter of tort law if the defendant failed to report such information. Id. at 97.
6. 2 ALI STUDY, supra note 1, at 95 (footnotes omitted). The excerpt discusses a "regulatory compliance defense"; this term was intended to encompass statutory preemption of liability as well as judicially recognized preclusion. Id.
7. See generally RICHARD B. STEWART, THE ROLES OF LIABILITY AND REGULATION IN CONTROLLING ENTERPRISE RISKS (1987) (describing dual-track regime);
8. Paul Dueffert, Note, The Role of Regulatory Compliance in Tort Actions, 26 HARV. J. ON LEGIS. 175 (1989) (same).
9. See 2 ALI STUDY, supra note 1, at 87-89.
10. Id.
11. See discussion infra note 86 and accompanying text.
12. See 2 ALI STUDY, supra note 1, at 87-88.
13. See id. at 89.
14. See ALI STUDY, supra note 1, at 384.
15. See 2 ALI STUDY, supra note 1, at 89.
16. Id. Such overdeterrence, for example, has led to the nonmanufacture of Thalidomide, despite its potentially unique anti-inflammatory and immunosuppressant properties. See Sylvia A. Law, Tort Liability and the Availability of Contraceptive Drugs and Devices in the United States, 23 N.Y.U. REV. L. & SOC. CHANGE 339, 361 (1997).
17. See 2 ALI STUDY, supra note 1, at 85-86.
18. For an analysis of the advantages and disadvantages of regulatory compliance preclusion in the context of prescription drugs, see generally Margaret Gilhooley, Innovative Drugs, Products Liability, Regulatory Compliance, and Patient Choice, 23 SETON HALL L. REV. 1481 (1994).
19. See Michael D. Green, Safety as an Element of Pharmaceutical Quality: The Respective Roles of Regulation and Tort Law, 42 ST. LOUIS U. L.J. 163, 164 (1998) [hereinafter Green, Safety).
20. For a discussion of the Bendectin litigation, see generally MICHAEL D. GREEN, BENDECTIN AND BIRTH DEFECTS: THE CHALLENGES OF MASS Toxic SUBSTANCES LITIGATION (1996);
21. JOSEPH SANDERS, BENDECTIN ON TRIAL: A STUDY OF MASS TORT LITIGATION (1998). Bendectin was first marketed in 1956 before the 1962 Amendments to the Food, Drug, and Cosmetic Act (FDCA) came into effect. Nevertheless, the Food and Drug Administration (FDA) conducted a hearing in 1980, concluding that the available data failed to demonstrate a link between birth defects and Bendectin. As Professor Green notes, the FDA "essentially exonerated" Bendectin.
22. Michael D. Green, Expert Witnesses and Sufficiency of Evidence in Toxic Substances Litigation: The Legacy of Agent Orange and Bendectin Litigation, 86 NW. U. L. REV. 643, 679-80 (1992).
23. See Oxendine v. Merrell Dow Pharrn., Inc., 506 A.2d 1100, 1100 (D.C. 1986).
24. See James T. Rosenberg, Lessons from Litigation over Silicone Breast Implants: A Call for Activism by Scientists, 276 SCIENCE 1524, 1524 (1997).
25. See James T. Rosenberg, SCIENCE id. 276 (1997). 1524
26. See MARCIA ANGELL, SCIENCE ON TRIAL 10-11, 90-110 (1996) (describing studies);
27. see also David A. Kessler, The Basis of the FDA's Decision on Breast Implants, 326 NEW ENG. J. MED. 1713, 1713-15 (1992) (explaining withdrawal of FDA approval); Rosenberg, supra note 20 (discussing expert testimony in breast implant litigation). Silicone breast implants were grandfathered under the 1976 Medical Device Amendment to the FDCA, having been manufactured, seemingly without problems, since 1960. See ANGELL, supra, at 51.
28. See Rosenberg, supra note 20.
29. See ANGELL, supra note 22.
30. The FDA regulates both the development and manufacture of vaccines, requiring data on employee qualifications and workplace conditions as well as samples from each lot produced. See Daniel A. Cantor, Comment, Striking a Balance Between Product Availability and Product Safety: Lessons from the Vaccine Act, 44 AM. U. L. REV. 1853, 1879 (1995).
31. W. Kip Viscusi, The Social Costs of Punitive Damages Against Corporations in Environmental and Safety Torts, 87 GEO. L.J. 285, 326 (1998).
32. See Green, Safety, supra note 18, at 164-66 (describing Vaccine Act). Even within the administrative procedure set up by Congress, recovery for vaccine injuries is predicated on a showing of causation. In many cases, general causation is firmly established, and the markers for specific causation are well established. However, some claims reach beyond the epidemiological evidence; here, causation issues can be quite difficult. See Derry Ridgway, No-Fault Vaccine Insurance: Lessons from the National Vaccine Injury Compensation Program, 24 J. HEALTH POL. POL'Y & L. 59, 68 (1999). Under the Vaccine Act, the Department of Health and Human Services decides which adverse conditions are entitled to a presumption that they were caused by a particular vaccine; these presumptions are based on epidemiological data from scientific research.
33. See Marlene K. Tandy, Vaccine Compensation, 6 FED. CIRCUIT B.J. 413, 414 (1996).
34. See Kellen F. Cloney, AIDS Vaccine Manufacturers v. Tort Regime: The Need for Alternatives, 49 WASH. & LEE L. REV. 559, 566-69 (1992).
35. Rabin, supra note 2, at 2063.
36. Id.
37. Risk-benefit tradeoffs compare the benefits of measures to reduce risk against their costs. Risk-risk tradeoffs are a subset of risk-benefit tradeoffs; they compare the benefits of measures to reduce given risks against the costs that they might impose by increasing other forms of risk. Benefit-benefit tradeoffs compare the extent of risk reduction benefits achieved by alternative investments of societal resources in different risk reduction measures.
38. See Rabin, supra note 2 at 2075-76; 2 ALI STUDY, supra note 1, at 108.
39. See Richard H. Pildes & Cass R. Sunstein, Reinventing the Regulatory State, 62 U. CHI. L. REV. 1, 4-6 (1995).
40. Kip Viscusi points, by way of example, to a chemical spill case in which the jury penalized the defendant $210 million notwithstanding the defendant's compliance with all EPA regulations and the fact that no land was taken out of agricultural use as a result of the spill. See Viscusi, supra note 26, at 318 (discussing Wilhite v. Rockwell Int'l Corp., 93-CI-00158 (Ky. Cir. Ct. June 24, 1996)).
41. Cf. JERRY L. MASHAW, BUREAUCRATIC JUSTICE: MANAGING SOCIAL SECURITY DISABILITY CLAIMS (1983) (arguing, in specific context of social security disability determinations, that a centralized administrative bureaucracy is better able than a process of case-by-case judicial determinations to make decisions that will promote program objectives in effective and consistent manner because it can gather and analyze information and make policy decisions on a systemic and coordinated basis).
42. See John D. Graham & Jonathan B. Wiener, Risk Versus Risk: Tradeoffs in Protecting Health and the Environment (1995);
43. Cass R. Sunstein, Health-Health Tradeoffs, 63 U. CHI. L. REV. 1533, 1536-40 (1996).
44. See Corrosion Proof Fittings v. EPA, 947 F.2d 1201, 1224-26 (5th Cir. 1991). Some commentators argue that almost all regulatory decisions indirectly present risk-risk tradeoffs. They argue that even if a regulatory requirement does not directly result in increased environmental health and safety risks, excessive regulatory stringency will deter the introduction of new technologies that are safer than old ones. They also argue that "wealthier is healthier," relying on aggregate data which indicate higher levels of health in more prosperous societies and among wealthier sectors of the population. Accordingly, they argue, excessively stringent regulation will reduce society's health by reducing its wealth. See THE LIABILITY MAZE: THE IMPACT OF LIABILITY LAW ON SAFETY AND INNOVATION (Peter W. Huber & Robert E. Litan eds., 1991);
45. Peter W. Huber, Safety and the Second Best: The Hazards of Public Risk Management in the Courts, 85 COLUM. L. REV. 277, 324-25 (1985).
46. See Wesley A. Magat et al., Consumer Processing of Hazard Warning Information, 1 J. RISK & UNCERTAINTY 201 (1988).
47. See W. KIP VISCUSI, PRODUCT-RISK LABELING: A FEDERAL RESPONSIBILITY 1-7 (1993).
48. See STEPHEN G. BREYER, BREAKING THE VICIOUS CYCLE: TOWARD EFFECTIVE RISK REGULATION 39-51 (1993);
49. COMPARING ENVIRONMENTAL RISKS: TOOLS FOR SETTING GOVERNMENT PRIORITIES (J. Clarence Davies ed., 1996);
50. WORST THINGS FIRST? THE DEBATE OVER RISK-BASED NATIONAL ENVIRONMENTAL POLICIES 33-87 (Adam M. Finkel & Dominic Golding eds., 1994).
51. See STEPHEN G. BREYER ET AL., ADMINISTRATIVE LAW AND REGULATORY POLICY 30-31 (4th ed. 1999).
52. But see Lisa Heinzerling, Regulatory Costs of Mythic Proportions, 107 YALE L.J. 1981, 1998-2017 (1998) (criticizing some of the assumptions used in generating high cost-per-life-saved figures).
53. If it were assumed that compliance resource inputs were fixed, this optimization process could be analyzed in risk-risk terms, because a decision to devote more resources to one regulatory program in order to reduce the risks that it addresses would result in a corresponding increase in the risks addressed by other programs.
54. See Rabin, supra note 2, at 2065-67.
55. 509 U.S. 579 (1993) (outlining new approach for admission of expert testimony under which trial judge considers relevance and reliability of proffered testimony, essentially acting as a "gatekeeper").
56. See Rabin, supra note 2, at 2064-66.
57. Id. at 2065 & n.74.
58. See ANGELL, supra note 22, at 109-32 (1996). Marcia Angell describes how "[s]ome distinctly second-rate scientists testify in trial after trial. . . . [I]n essence they become well-practiced, professional witnesses, whose major talent is convincing juries, not evaluating evidence." Id. at 117.
59. See Rabin, supra note 2, at 2065-67.
60. For a thorough description of Judge Pointer's experiment, see Olivia Judson, Slide-Rule Justice, NAT'L J., Oct. 9, 1999, at 2882.
61. Id. at 2882 (noting that total cost of panel was at least $1.3 million). Moreover, Judge Pointer presided over multidistrict proceedings, not a class action. Thus, his findings on causation will be reported back to each of the individual litigations and each jury will independently consider the report of his panel. In other words, the findings of the panel are in no way dispositive in individual breast implant cases; some judges may use the panel report to exclude plaintiff expert testimony, and, potentially, grant summary judgment, but juries may end up deciding some cases.
62. At the request of Congress, the Institute of Medicine (an arm of the National Academy of Science) surveyed the scientific studies on breast implants and found "no definitive evidence linking breast implants to cancer, immunological diseases, neurological problems, or other systemic diseases." Nancy Nelson, Institute of Medicine Finds No Link Between Breast Implants and Disease, 91 J. NAT'L CANCER INST. 1191, 1191 (1999). The panel spent two years reviewing over three thousand publications and hearing testimony.
63. See Nancy Nelson, J. NAT'L CANCER INST. id. 1191 (1999)
64. See Rabin, supra note 2, at 2067.
65. See BREYER ET AL., supra note 41, at 203-413.
66. Rabin, supra note 2, at 2068.
67. See id. at 2065-67.
68. See 28 U.S.C. § 1407(a) (1994) ("When civil actions involving one or more common questions of fact are pending in different districts, such actions may be transferred to any district for coordinated or consolidated pretrial proceedings.").
69. See Rabin, supra note 2, at 2066.
70. Professor Rabin's paper assumes, for purposes of his analysis, that agencies are not captured or moribund, although he notes that such problems are commonly invoked as arguments against regulatory compliance preclusion. See id. at 2068. As noted below, the requirement in the ALI Study's compliance preclusion proposal of affirmative public disclosure by regulated firms of information in their possession relating to the risks of their products and processes, provides a "sunshine" antidote to these problems. Another possible response is to follow the model of many federal environmental regulatory statutes and provide more generally for citizens' suits against both against regulators and regulated, including fee shifting provisions.
71. 21 U.S.C. §§ 301-397 (1994).
72. See, e.g., 21 C.F.R. § 314.50(d) (1999) (FDA application regulations for new drug technical reporting requirements); id. § 314.80 (FDA regulations for post-marketing reporting requirements for adverse drug reactions).
73. See 21 U.S.C. § 331 (1994) (defining prohibited acts under the federal Food, Drug, and Cosmetic Act); id. § 333 (defining penalties under the federal Food, Drug, and Cosmetic Act).
74. Unlike tort actions or even individual product approvals, however, rulemaking initiatives produce "a wealth of information" from other agencies and independent scientific research efforts. See Lars Noah, Rewarding Regulatory Compliance: The Pursuit of Symmetry in Products Liability, 88 GEO. L.J. 2147, 2161 & nn.57-61 (2000).
75. As noted by the ALI Study, its disclosure condition could present trade secret issues with respect to some portions of the information required to be disclosed to agencies, particularly information relating to the "means for controlling product risks." 2 ALI STUDY, supra note 1, at 97 n.30. Thus, some of the information required to be disclosed to the agency might not be available to the public.
76. The ALI Study proposal also helps to address the problem of regulatory sloth or capture by providing regulated firms the benefits of preclusion, requiring that the risk in question have been directly addressed by a regulatory standard or order. This leaves the problem, however, presented by cases where the risk has been addressed directly but the regulatory requirement is insufficiently stringent. Judicial review of the agency's decision through litigation initiated by environmental, consumer, or workers' groups often provides a potential remedy for problems of regulatory inadequacy when they are serious and have widespread implications. Even after opportunity for judicial review, however, inadequate regulations may well remain. Tort liability offers a further remedy for this problem. For reasons discussed earlier in the text, however, I believe that in this case the cure is worse than the disease. Further, Lars Noah argues that evidence for the agency capture thesis is anecdotal and inconclusive at best. See Noah, supra note 62, at 2154-55 & n.29.
77. Professor Rabin discusses a version of regulatory compliance preclusion in which preclusion can only be avoided if the plaintiff demonstrates fraud on the part of the defendant in concealing information. See Rabin, supra note 2, at 2069. Requiring proof of fraud is, of course, a far narrower exception to preclusion than the broad requirements of affirmative disclosure in the ALI proposal.
78. As Professor Rabin points out, however, a regulatory compliance preclusion provision with exceptions, whether for fraud or failure to meet a broader nondisclosure requirement, does involve additional administrative costs in litigating the applicability of the exception in particular cases. See id. at 2076.
79. See id. at 2075.
80. See id.
81. See id. at 2073.
82. See 2 ALI STUDY, supra note 1, at 441-83; Rabin, supra note 2, at 2074. See generally Robert L. Rabin, Some Thoughts on the Efficacy of a Mass Toxics Administrative Compensation Scheme, 52 MD. L. REV. 951 (1993).
83. Creating a separate compensation system for each drug product would arguably achieve distributional justice in a more finely tuned fashion but would involve excessive administrative costs. Nonetheless, several commentators have proposed such a scheme to increase availability of contraceptive devices. See, e.g., Janet Benshoof, Protecting Consumers, Prodding Companies, and Preventing Conception: Toward a Model Act for No Fault Liability for Contraceptives, 23 N.Y.U. REV. L. & Soc. CHANGE 403, 403-10 (1997).
84. Professor Rabin suggests that the risk-benefit test applied by regulators is premised on the availability of compensation in tort, and addresses the contention that, in the absence of compensation mechanisms other than tort liability, adoption of regulatory compliance preclusion could lead regulators to adopt overly stringent requirements in order to prevent the occurrence of uncompensated injuries. See Rabin, supra note 2, at 2076. The ALI study also noted this possibility. See 2 ALI STUDY, supra note 1, at 108-09. This issue requires further study and analysis. Some regulatory systems, such as OSHA, impose stringent regulations even though no-fault compensation is readily available. Regulation of certain environmental risks, such as radon exposure, is weak, even though compensation is not available. As these examples suggest, there are many other variables involved that will complicate the determination of whether or not there is any systematic correlation between regulatory stringency and the availability of compensation.
85. See Rabin, supra note 2, at 2076.
86. See RICHARD EPSTEIN, TORTS §§ 13.3, 16.10 (1999). A few courts have imposed strict liability for failure to warn of asbestos hazards. See Beshada v. Johns-Manville Prods. Corp., 447 A.2d 539 (N.J. 1982).
87. Rabin, supra note 2, at 2073.
88. See Soule v. General Motors Corp., 882 P.2d 298, 304-11 (Cal. 1994) (describing consumer expectations test for liability due to design defect).
89. Professor Rabin states that under a consumer expectations test for product design defect, "it is quite possible that 'true' strict liability would be applicable." Rabin, supra note 2, at 2072. He suggests that a "warranty-based theory of design defect liability" would have the same effect. Id.
90. See Soule, 882 P.2d at 307 (noting that consumer expectations test is one of "two alternative ways to prove a design defect").
91. This determination is, of course, directly contrary to the prior regulatory determination that the product not only may but should be marketed because it is of net benefit to society; the state's tort system has rejected the regulator's approval of the product.
92. See 2 ALI STUDY, supra note 1, at 45 & n.18.
93. Rabin, supra note 2, at 2072.
94. Soule, 882 P.2d 298 (Cal. 1994).
95. Id. at 308.
96. The discussion in the text assumes that, under Soule, the consumer expectations test of defect and the risk-benefit test of defect are exclusive alternatives that will each be applied to distinct and readily identifiable categories of products. In California and other jurisdictions that recognize both tests, however, the line between cases appropriate for application of the consumer expectations test and those in which the risk-benefit analysis should prevail may be far from unambiguous and there may also be overlaps in application. The Soule court suggested as much in stating that "ordinary consumer expectations are not irrelevant simply because expert testimony is required to prove that the product failed as marketed. . . ." Id. at 308 & n.6. Uncertainty regarding the application of the two tests could undermine many of the benefits of regulatory compliance preclusion as applied to cases decided under the risk-benefit test, because a manufacturer could not know ex ante which test would apply in an action against its product.
97. See Rabin, supra note 2, at 2071 & n.99.
98. See 2 ALI STUDY, supra note 1, at 101. The Study stated: Because of its relatively stringent requirements, [the study's proposal for regulatory compliance preclusion] would probably not have a very extensive application. It would apply primarily to closely regulated products such as pharmaceuticals, medical devices, pesticides, and aircraft; to certain features of automobiles and other general consumer and industrial products; to some forms of pollution and toxic waste cleanup; and to aspects of the services provided by doctors, architects, engineers, and other professions. Id.