Reconciling cost-benefit analysis with the principle that safety matters more than money

New York University Law Review

Volumn 76, Issue 1, 2001, Pages 114-189

  Geistfeld, Mark ^a  

^a NEW YORK UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 23044528464     PISSN: 00287881     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (199)

1. William Glaberson, Looking for Attention with a Billion-Dollar Message, N.Y. Times, July 18, 1999, § 4 (Week in Review), at 3. The award was subsequently reduced to $1.2 billion.
2. Milo Geyelin, Lasting Impact: How an Internal Memo Written 26 Years Ago Is Costing GM Dearly, Wall St. J., Sept. 29, 1999, at Al. The jury was infuriated by a 1973 internal GM memo stating that postcollision fires cost GM $2.40 per car based on forty-one million cars on the road at that time, rear-end collision fatality rates (about 500 per year), and an estimated value of human life of $200,000 based on government statistics.
3. Milo Geyelin, Wall St. J., 1999, Id. Al
4. GM and the Law, Economist, July 17, 1999, at 16, 16. The jury did not consider a complete cost-benefit analysis (CBA).
5. Micheal M. Weinstein, Analysis: Cost of Life Issue in GM Suit Lost on Price of a Part, J. Rec. (Okla. City), Aug. 13, 1999, 1999 WL 9848045. Nevertheless, the jury's indignation at an internal company memo acknowledging and monetizing the risk poses a problem for cost-benefit analyses of safety measures, as such analyses must identify and monetize risk.
6. Grimshaw v. Ford Motor Co., 174 Cal. Rptr. 348 (Ct. App. 1981).
7. Gary T. Schwartz, The Myth of the Ford Pinto Case, 43 Rutgers L. Rev. 1013, 1014 (1991) (observing that Ford Pinto case "shows how disturbed the public can be by corporate decisions that balance life and safety against monetary cost").
8. Poll Shows Four of Five Americans Support Environment, Even Over Economy, 23 Env't Rep. (BNA) 1155 (Aug. 7, 1992).
9. E.g., Charles Fried, An Anatomy of Values: Problems of Personal and Social Choice 192 (1970);
10. Alan Gewirth, Reason and Morality 54-57 (1978) (arguing for lexical ranking of goods into three tiers in which physical integrity is ranked as first-tier "basic" good, and wealth and convenience are ranked as third-tier "additive" goods);
11. Judith Jarvis Thomson, The Realm of Rights 176-248 (1990) (arguing that individuals have right that others not cause them harm, which permissibly can be infringed by others only if doing so would produce sufficiently large and appropriately distributed increment of good);
12. Ernest J. Weinrib, The Idea of Private Law 147-52 (1995) (arguing that corrective justice based on Kantian right renders cost considerations irrelevant for "real" risks);
13. Gregory C. Keating, Reasonableness and Rationality in Negligence Theory, 48 Stan. L. Rev. 311, 349-60 (1996) (arguing that Rawlsian social contract theory places higher value on security interests than ordinary economic interests);
14. David McCarthy, Rights, Explanation, and Risks, 107 Ethics 205, 212-15 (1997) (arguing that individuals have right that others not impose risks upon them, which permissibly can be infringed only if resultant good "sufficiently outweighs" burden imposed on bearer of right and is appropriately distributed).
15. W. Page Keeton et al., Prosser and Keeton on the Law of Torts 132 (5th ed. 1984). There is some question whether this weighting of interests applies to accidental harms. See infra note 98. The analysis here shows why the principle should apply to accidental harms. See infra Parts III, VII.B.
16. Case T-13/99 R, Pfizer Animal Health SA/NV v. E.U. Council, 3 C.M.L.R. 79, 82 (1999).
17. 29 U.S.C. §§ 651-678 (1994).
18. Am. Textile Mfrs. Inst. v. Donovan, 452 U.S. 490, 509 (1981).
19. The standards for criteria pollutants under the Clean Air Act cannot rely on cost considerations. See 42 U.S.C. § 7409 (1994). Instead, the standards must ensure "an adequate margin of safety," id. § 7409(b)(1) (ambient air quality standards), or an "ample margin of safety," id. § 7412(d)(4) (hazardous air pollutants). But cf. Am. Trucking Ass'ns v. EPA, 175 F.3d 1027, 1033 (D.C. Cir.) (finding Clean Air Act's provisions for ambient air quality standards, as interpreted by EPA, to be unconstitutional delegation of legislative power), reh'g denied, 195 F.3d 4 (D.C. Cir. 1999), rev'd sub nom. Whitman v. Am. Trucking Ass'ns, Nos. 99-1257, 99-1426, 2001 WL 182549 (U.S. Feb. 27, 2001). The Act also contains the pervasive requirement that companies adopt the "best available control technology" rather than technology that passes a cost-benefit test or is cost-effective. See 42 U.S.C. §§ 7411, 7475. Similarly, water pollution standards for toxic pollutants must provide an "ample margin of safety." 33 U.S.C. § 1317 (a)(4) (1994). Another example is provided by the Endangered Species Act, which forbids tradeoffs between preservation and cost except in rare circumstances. 16 U.S.C. § 1533 (1994).
20. Dennis J. Paustenbach, Retrospective on U.S. Health Risk Assessment: How Others Can Benefit, 6 Risk 283, 287 & n.17 (1995).
21. A leading advocate of the EPA's methodology acknowledges a sevenfold overestimation. Adam M. Finkel, Is Risk Assessment Really Too Conservative? Revising the Revisionists, 14 Colum. J. Envtl. L. 427, 431 (1989);
22. Adam M. Finkel, A Second Opinion on an Environmental Misdiagnosis: The Risky Prescriptions of Breaking the Vicious Circle, 3 N.Y.U. Envtl. L.J. 295, 349 (1995).
23. Others argue that the methodology overstates the risk by a factor of more than 100 or even infinity in some cases. E.g., Philip H. Abelson, Exaggerated Risks of Chemicals, 48 J. Clinical Epidemiology 173, 175 (1995).
24. In the terminology of statistical decision theory, the issue is one of determining how to compare scientific false positives (i.e., the chance that the estimation procedure identifies a substance as being hazardous when it is not) with scientific false negatives (i.e., the chance that the substance is deemed harmless when it actually is hazardous). A decision procedure designed to overestimate risk reduces the chance of false negatives while increasing the chance of false positives. The justification for tipping the balance in this way is that false negatives cost lives, whereas false positives cost money. Talbot Page, A Generic View of Toxic Chemicals and Similar Risks, 7 Ecology L.Q. 207, 219-20 (1978).
25. E.g., Frank B. Cross, Paradoxical Perils of the Precautionary Principle, 53 Wash. & Lee L. Rev. 851, 851-54 (1996).
26. Framework Convention on Climate Change, May 9, 1992, art. 3(3), 31 I.L.M. 849, 854 (1992).
27. See Comm'n of the Eur. Cmtys., Communication from the Commission on the Precautionary Principle 11, 27 (Feb. 2, 2000) [hereinafter Communication on the Precautionary Principle] (discussing definition and use of precautionary principle in European and international agreements, including Biosafety Protocol), at http://europa.eu.int/comm/dgs/health_consumer/library/pub/pub07_en.pdf. The Protocol was signed in Nairobi by sixty-five countries on May 25, 2000; the European Community and its members were signatories.
28. Biological Diversity: Biosafety Protocol Attracts 65 Signatures, Eur. Env't, June 1, 2000, 2000 WL 20766626;
29. see also Marcela Valente, Environment/Trade: No Simple Answer in Debate on Biotechnology, Inter Press Serv., Feb. 17, 2000, 2000 WL 4089890 (discussing Biosafety Protocol).
30. E.g., Valuing Health for Policy: An Economic Approach (George Tolley et al. eds., 1994).
31. E.g., Alan M. Garber, Advances in Cost-Effectiveness Analysis of Health Interventions 6 (Nat'l Bureau of Econ. Research, Working Paper No. 7198, 1999);
32. Alice M. Rivlin, Rationalism and Redemocratization: Time for a Truce, in Worst Things First? The Debate Over Risk-Based National Environmental Priorities 25 (Adam M. Finkel & Dominic Golding eds., 1994) (describing how early efforts to compare disease control programs at U.S. Department of Health, Education, and Welfare "evoked a storm of protest, especially from the medical professionals, who objected to trying to quantify benefits" of lives saved).
33. See generally E. J. Mishan, Cost-Benefit Analysis (3d ed. 1982).
34. Numerous examples of federal legislation requiring CBA are provided in Am. Textile Mfrs. Inst. v. Donovan, 452 U.S. 490, 510-11 & 510 n.30 (1981). The most significant recent example is the Unfunded Mandates Reform Act, which requires CBA of all federal regulations that involve significant expenditures by state, local, or tribal governments. 2 U.S.C. §§ 1501-1571 (Supp. IV 1998). Other prominent recent examples include the Safe Drinking Water Act Amendments, 42 U.S.C. § 300g-1(b)(3)(C) (Supp IV 1998); and the Toxic Substances Control Act, 15 U.S.C. § 2605(c)(1) (1994).
35. President Reagan's order concerning CBA is Exec. Order No. 12,291, 3 C.F.R. 127 (1981), reprinted in 5 U.S.C. § 601 (1988). President Bush did not repeal this order. President Clinton subsequently repealed the order and replaced it with another requiring CBA. See Exec. Order No. 12,866, 3 C.F.R. 638 (1993), reprinted in 5 U.S.C. § 601 (1994). The differences between the two orders are described in Richard H. Pildes & Cass R. Sunstein, Reinventing the Regulatory State, 62 U. Chi. L. Rev. 1, 6-7 (1995).
36. Ten states currently require CBA of all proposed agency rules, and seven states require CBA of selected rules. Robert W. Hahn, State and Federal Regulatory Reform: A Comparative Analysis 3 (AEI-Brookings Joint Ctr. for Regulatory Studies, Working Paper 98-3, 1998), http://www.aei.brook.edu/publications/working/working_98_03.pdf.
37. Communication on the Precautionary Principle, supra note 17.
38. Cross, supra note 15, at 859 (showing how safety principle, as embodied in precautionary principle of environmental law, "can be attacked as an uncertain decision rule").
39. For a discussion of the issues that arise when risks are traded off against one another, see John D. Graham & Jonathan Baert Wiener, Confronting Risk Tradeoffs, in Risk Versus Risk: Tradeoffs in Protecting Health and the Environment 1 (John D. Graham & Jonathan Baert Wiener eds., 1995).
40. This claim is central to any reconciliation of CBA and the safety principle, because welfare economics is based on the "non-paternalistic assumption that a household has 'sovereignty' over how its welfare is to be measured." Robin Boadway & Neil Bruce, Welfare Economics 39 (1984). The claim does not appear to be morally problematic, as there seems to be widespread agreement among philosophers that well-informed consent is sufficient to justify risk exposures. E.g., McCarthy, supra note 6, at 215 ("Consent is one of the most important notions within the realm of rights, and it is clearly relevant to the permissibility of many risk-imposing activities . . . ."). To be sure, there is philosophical disagreement concerning the appropriateness of relying on individual preferences to resolve moral issues, but that debate is not relevant to the question of whether the safety principle should apply to well-informed, voluntary risk exposures. Those who reject the use of individual preferences for resolving moral issues typically do so on the ground that individual preferences are not sufficiently well informed or otherwise concern purely private matters, making them an unreliable indicator of an individual's views on public matters.
41. See Daniel A. Farber, Eco-Pragmatism: Making Sensible Environmental Decisions in an Uncertain World 51-56 (1999) (summarizing debate). By definition, an individual's well-informed choice to face risk-threatening injury to the self is a private matter not plagued by an informational problem. (If others also would be injured by the risk exposure, for them the risk is involuntary, and so the safety principle would become relevant in that respect.) Hence the concerns that might justify rejection of individual preferences for resolving moral issues are not relevant to the question of whether the safety principle should apply to well-informed voluntary risk exposures.
42. As an analytical matter, risk-averse individuals have a declining marginal utility of wealth, a behavioral trait that plausibly describes most people. Moreover, only risk-averse individuals would be willing to pay insurance premiums, which are based on the expected cost of the insured-against event plus administrative charges. The fact that individuals commonly purchase insurance therefore supports the claim that most individuals are risk averse.
43. As Gregory Keating argues, "[accidental injury threatens grave disruption to the pursuit of a conception of the good, and accidental death prematurely ends the pursuit of such a conception. . . . Increased monetary expenditures on precaution, by contrast, burden persons' capacity to realize their conceptions of the good in more modest and incremental ways." Keating, supra note 6, at 354-55.
44. Cass R. Sunstein, Paradoxes of the Regulatory State, 57 U. Chi. L. Rev. 407, 413-14 (1990); see also Weyerhaeuser Co. v. Costle, 590 F.2d 1011, 1043 (D.C. Cir. 1978) (stating that best practicable technology standard under Clean Water Act is based on notion that public has right to clean environment, limited only by extent to which cleanup is impractical or unachievable).
45. Ronald Dworkin, Taking Rights Seriously 184-205 (1978).
46. For an explication of three related uses of the term "incommensurability," see Matthew Adler, Introduction to Symposium, Law and Incommensurability, 146 U. Pa. L. Rev. 1169, 1170-81 (1998).
47. See supra notes 18-19 and accompanying text.
48. "[I]f the entire U.S. gross domestic product were devoted to avoiding fatal accidents, we would have only $55 million to spend per life at risk. That expenditure would leave us no available resources to combat cancer and AIDS or to provide for other health-related needs, such as food and housing." W. Kip Viscusi, The Dangers of Unbounded Commitments to Regulate Risks, in Risks, Costs, and Lives Saved: Getting Better Results from Regulation 135, 135 (Robert W. Hahn ed., 1996).
49. This lack of coordination explains the irrational outcomes created by the federal legislation that does not allow agencies to set regulatory standards on the basis of cost considerations. One of these statutes involves workplace standards for toxic materials and harmful physical agents. See supra note 9 and accompanying text. A regulation promulgated under this statute concerning potentially carcinogenic benzene emissions during waste operations costs $19 million per year of life saved, whereas it would cost about $17,000 per year of life saved if all women over age fifty were to receive regular mammograms. See Tammy O. Tengs & John D. Graham, The Opportunity Costs of Haphazard Social Investments in Life-Saving, in Risks, Costs and Lives Saved, supra note 34, at 167, 167. A more coordinated regulatory approach would enable us to save about twice as many lives for the same amount of money currently expended on life-saving regulatory interventions. See id. at 172-74.
50. The claim that money and lives are truly noncomparable is controversial for those who believe that the two are incommensurable. See Ruth Chang, Introduction to Incommensurability, Incomparability, and Practical Reason 1, 13-24 (Ruth Chang ed., 1997).
51. See Elizabeth Anderson, Value in Ethics and Economics 59-64 (1993).
52. See generally Cass R. Sunstein, Incommensurability and Valuation in Law, 92 Mich. L. Rev. 779 (1994).
53. The rudiments of such a process are described in Pildes & Sunstein, supra note 22, at 89-95.
54. Even if the safety principle should account for the care taken by potential victims, the way in which the safety principle alters CBA is unlikely significantly to reduce the incentive of potential victims to take care. As shown in infra Part IV, the safety principle modifies CBA whenever potential victims cannot be compensated fully for facing the risk. The modification does not compensate potential victims fully for the risk, giving them an incentive to avoid the injury.
55. CBA does not yield consistent answers to the question of whether one project involving fatal risks is better than another. Charles Blackorby & David Donaldson, Can Risk-Benefit Analysis Provide Consistent Policy Evaluations of Projects Involving Loss of Life?, 96 Econ. J. 758 (1986). The problem of consistency does not imply that CBA ought to be rejected by legal decisionmakers, however, since CBA still might be the most defensible decision procedure for courts or agencies.
56. Matthew D. Adler & Eric A. Posner, Rethinking Cost-Benefit Analysis, 109 Yale L.J. 165 (1999). Another problem with CBA pertains to the difficulty of getting good estimates for the price of risks in nonmarket settings. The problem is not that the risk measures are methodologically inappropriate. Rather, the problem is one of figuring out how to get reasonably accurate estimates by relying on surveys or other methods.
57. E.g., Peter A. Diamond & Jerry A. Hausman, Contingent Valuation: Is Some Number Better than No Number?, 8 J. Econ. Persp. 45 (1994). Finally, some scholars question whether it is appropriate for CBA to define benefits and costs in terms of individual preferences. See, e.g., Adler & Posner, supra;
58. Allan Gibbard, Risk and Value, in Values at Risk 94 (Douglas MacLean ed., 1986). The argument here does not question whether CBA inappropriately relies on preferences, because it seems implausible that the safety principle routinely would reject the safety choices made by well-informed individuals. See supra note 27 and accompanying text.
59. A point worthy of biblical recognition. See Job 2:4 (New Jerusalem) ("Someone will give away all he has to save his life.").
60. For a more formal statement of the properties of the willing to pay (WTP) measure, see infra app. § 2.
61. See infra app. § 1.
62. See W. Michael Hanemann, Willingness to Pay and Willingness to Accept: How Much Can They Differ?, 81 Am. Econ. Rev. 635 (1991) (showing that willing to accept (WTA) and WTP increasingly will diverge as there are fewer substitutes for good being risked). A related empirical reason for the difference is that individuals are significantly more averse with respect to losses relative to a reference point than they are favorably disposed to gains, and that individuals tend to value goods they possess more highly than those they do not possess.
63. Daniel Kahneman, Jack L. Knetsch & Richard H. Thaler, Experimental Tests of the Endowment Effect and the Coase Theorem, 98 J. Pol. Econ. 1325 (1990).
64. See, e.g., Robert Cameron Mitchell & Richard T. Carson, Using Surveys to Value Public Goods: The Contingent Valuation Method 30 (1989) ("The choice between the WTP or WTA. formulation is a question of property rights: does the agent have the right to sell the good in question or, if he wants to enjoy it, does he have to buy it?").
65. See generally Richard A. Posner, The Economics of Justice (1983).
66. The analytical problems created by Posner's attempt to answer all legal questions by reference to CBA, in the guise of wealth maximization, are described in Lewis A. Kornhauser, Wealth Maximization, in 3 The New Palgrave Dictionary of Economics and the Law 679 (Peter Newman ed., 1998).
67. Posner now acknowledges this problem. Richard A. Posner, Wealth Maximization and Tort Law: A Philosophical Inquiry, in Philosophical Foundations of Tort Law 99, 99-100 (David G. Owen ed., 1995) (acknowledging that wealth maximization is incomplete criterion for assessing legal policy because wealth depends on assignment of property rights). Note that the indeterminacy disappears for perfectly reciprocal, nonconsensual risks. See infra Part IV.B.2.
68. For risks threatening injuries fungible with money, a risk-averse individual is willing to pay more than the expected value of the risk (probability of loss times value of loss) in order to avoid facing it. Risk aversion cannot be defined in this way for injuries not fungible with money, because the individual's valuation of the loss (the WTA or WTP measure) includes all costs posed by the risk, including the cost of risk aversion. See infra app. §§ 1, 2.
69. See supra Part I.
70. See, e.g., Allan M. Feldman, Welfare Economics, in 4 The New Palgrave: A Dictionary of Economics 889, 889-90 (John Eatwell ed., 1987) (tracing origins of welfare economics to nineteenth century debates concerning question of whether laissez-faire is "better" or "superior" to regulated economy).
71. See, e.g., Amartya Sen, The Possibility of Social Choice, 89 Am. Econ. Rev. 349, 351-52 (1999) (tracing origins of traditional welfare economics to influence of utilitarianism of Jeremy Bentham). The ensuing discussion of the new welfare economics draws on this source and on Mishan, supra note 20, at 301-14.
72. See, e.g., I.M.D. Little, A Critique of Welfare Economics (2d ed. 1957).
73. E.g., Richard W. Tresch, Public Finance: A Normative Theory 39 (1981);
74. Hal R. Varian, Microeconomic Analysis 405 (3d ed. 1992).
75. See generally Madeline Morris, The Structure of Entitlements, 78 Cornell L. Rev. 822 (1993) (identifying components of entitlements giving force to individual interests).
76. See supra notes 45-47 and accompanying text.
77. E.g., Thomson, supra note 6, at 205-48 (arguing that individuals have right that others not cause them harm); McCarthy, supra note 6, at 212-15 (arguing that individuals have right that others not impose risks upon them). This specification of the entitlement is commonly employed in economic analyses. Jennifer H. Arlen, Reconsidering Efficient Tort Rules for Personal Injury: The Case of Single Activity Accidents, 32 Wm. & Mary L. Rev. 41, 43 n.9 (1990).
78. The individual right to bodily security underlies the intentional tort of battery and is the same type of interest protected by the rules of negligence and strict liability. Restatement (Second) of Torts § 1 cmt. d (1965) ("[T]he interest in bodily security is protected against not only intentional invasion but against negligent invasion or invasion by the mischances inseparable from an abnormally dangerous activity."); id. ch. 2 introductory note (stating that interest in "freedom from bodily harm" is "given the greatest protection" by various intentional torts and also by tort rules concerning negligence and strict liability); id. § 281 cmt. b (stating that one element of negligence is "that the interest which is invaded must be one which is protected, not only against acts intended to invade it, but also against unintentional invasions"); see also infra Part VII (showing how entitlement held by potential victims conforms to important tort doctrines).
79. The risk reduction available from any given precaution typically depends on whether other precautions also are being taken. Strictly speaking, then, for each precaution there is an associated risk of injury conditional on other precautions being taken.
80. United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir. 1947) (Hand, J.).
81. Mark Geistfeld, Placing a Price on Pain and Suffering: A Method for Helping Juries Determine Tort Damages for Nonmonetary Injuries, 83 Cal. L. Rev. 773, 804-05 (1995).
82. Andrew J. McClurg, It's a Wonderful Life: The Case for Hedonic Damages in Wrongful Death Cases, 66 Notre Dame L. Rev. 57, 66-67 (1990).
83. Id. at 62-64; see also Restatement (Second) of Torts § 925 cmt. b (1977) (noting that, in most states, damages in wrongful death actions "are determined by the present worth of the contributions and aid that the deceased probably would have made to the survivors had he lived"). Note that the available damages in wrongful death actions are directly connected to the bequest motive of potential victims. Therefore, the distributive properties of cost-benefit tort rules can be analyzed in terms of someone who has no bequest motive and receives no compensation upon death.
84. Under the cost-benefit negligence standard and the rule of strict liability, potential injurers give equal weight to their own interests and the security interests of potential victims, causing them to forego any safety precaution costing more than the benefit of risk reduction. (Such a precaution would not be required by the cost-benefit negligence standard, and strictly liable potential injurers would not take the precaution because it would be cheaper to face the prospect of liability rather than to pay for the precaution.) A negligence standard based on the safety principle would require greater precautions on the part of potential injurers, and there are persuasive reasons for concluding that potential injurers would take these additional precautions, thereby reducing risk below the levels that attain under the cost-benefit negligence standard and strict liability. Insofar as potential injurers seek to conform their behavior to the requirements of the law, they will adhere to a more exacting negligence standard, even if in some cases it would be cheaper for them to forego a required precaution and risk liability. These individuals would not be similarly motivated by a rule of strict liability, which does not impose any standard of care on potential injurers. Instead, potential injurers in a regime of strict liability compare their precaution costs with their expected liability costs, leading them to take the same precautions as would be required by a perfectly enforced cost-benefit negligence standard. Even if potential injurers do not care about conforming their behavior to the legal standard of reasonable care and instead make safety decisions entirely on a cost-benefit calculus, they will adhere to a higher standard of care when courts faced by evidentiary problems impose liability on defendants whose negligence was not a but-for cause of the injury, which is likely to be a general phenomenon. See Stephen Marks, Discontinuities, Causation, and Grady's Uncertainty Theorem, 23 J. Legal Stud. 287 (1994); see also Zuchowicz v. United States, 140 F.3d 381, 390 (2d Cir. 1998) (holding that causation can be established if "(a) a negligent act was deemed wrongful because that act increased the chances that a particular type of accident would occur, and (b) a mishap of that very sort did happen"). Cases in which liability is imposed improperly on defendants due to the absence of cause-in-fact are much rarer under strict liability.
85. Instead, the cause-in-fact requirement more plausibly enables some strictly liable injurers to escape liability due to the plaintiff's inability to establish this requirement, thereby reducing the incentive of potential injurers to take precautions satisfying the cost-benefit test. For formal demonstration of these incentives under different rules, see Robert Cooler & Thomas Ulen, Law and Economics 270-75 (2d ed. 1997);
86. Marcel Kahan, Causation and Incentives to Take Care Under the Negligence Rule, 18 J. Legal Stud. 427 (1989). Notice that a negligence standard will not necessarily reduce risk relative to the level that would attain under strict liability if negligent defendants escape liability due to the plaintiff's inability to prove the absence of reasonable care. These situations provide a risk-reducing role for strict liability that is discussed in infra Part VII.C.
87. E.g., William M. Landes & Richard A. Posner, The Economic Structure of Tort Law 186-89, 314-15 (1987). The proposal by Landes and Posner was subsequently developed in McClurg, supra note 61.
88. Very few, if any, physical risks threaten only death. Instead, risks threaten a range of different physical injuries in addition to the possibility of death. The probability of injury P therefore can be decomposed into two parts. Let P1 denote the probability that the potential victim will suffer a nonfatal injury, and let P2 denote the probability of fatal injury. The overall risk is P = P1 + P2 and has an associated WTA measure. The potential victim could use the WTA risk proceeds to purchase a guarantee to a tort award from the potential injurer in the event of a nonfatal injury. By the same reasoning used in the text accompanying supra note 60, this guarantee implies a tort award of D = WTA/P1. Because the tort award is based on the WTA measure for the total risk, rather than the WTA measure for the risk of nonfatal injury P1, it includes compensation for both the nonfatal and fatal injuries.
89. See supra notes 42-49 and accompanying text.
90. See infra app. § 3.
91. See, e.g., Morris, supra note 54, at 878-79 (using nonnegligent injuries as example of "uncompensated" entitlement because entitlement holder, potential victim, does not have right to monetary compensation).
92. See supra note 63 and accompanying text. This same logic suggests that strict liability might be justified on deterrence grounds whenever it reduces risk below the level attainable by a negligence standard. See infra Part VII.C.
93. The uncompensated risk measure includes components of economic loss, and the additional weight given to such safety interests is not mandated by the safety principle. See supra Part I (arguing that safety principle does not govern tradeoffs between economic interests). However, the potential victim still will face nonconsensual fatal risks under this approach, making her less well off than in a world in which she is compensated fully for facing nonconsensual risks. Inclusion of economic interests therefore does not seem to benefit potential victims unfairly.
94. If there are only two potential victims, for example, the bargaining range is between 2WTA and 4WTA, in which case conventional CBA chooses the rule least favorable to potential victims (based on 2WTA). In the example used in text, by contrast, the midpoint of the bargaining range is 11WTA, and conventional CBA relies on 10WTA. Thus, as the number of potential victims increases, conventional CBA becomes increasingly favorable to potential victims and relies on a WTA measure that approaches the midpoint of the bargaining range.
95. Cf. supra notes 30-32 and accompanying text (discussing safety principle as embodying absolute right to safety comparable to individual's constitutional rights).
96. For a formal demonstration of this point, see Geistfeld, supra note 60, at 851-52. For analyses of how bilateral risks affect the welfare properties of different tort rules, see Jennifer H. Arlen, Liability for Physical Injury When Injurers as Well as Victims Suffer Losses, 8 J.L. Econ. & Org. 411 (1992); Arlen, supra note 56.
97. When potential victims have different preferences for safety, the cost-benefit rule will be more advantageous to some potential victims and less advantageous to others. In these contexts, however, the distributive issue involves different types of potential victims and does not involve any distributive inequities between potential injurers and victims. The safety principle has no obvious relevance for resolving distributive issues between potential victims. Moreover, it is virtually impossible to devise legal rules to effectuate desired transfers between classes of buyers/potential victims. See Richard Craswell, Passing On the Cost of Legal Rules: Efficiency and Distribution in Buyer-Seller Relationships, 43 Stan. L. Rev. 361 (1991).
98. Contractual settings in which the risk threatens the seller (like an employee) might not require modification of CBA. The contractual relationship creates the opportunity for transfer of the WTA proceeds to the potential victim. To the extent the transfer occurs, there is no need to modify CBA.
99. See supra Part II.B.
100. Joseph E. Stiglitz, Pareto Efficient and Optimal Taxation and the New New Welfare Economics, in 2 Handbook of Public Economics 991, 992 (Alan J. Auerbach & Martin Feldstein eds., 1987).
101. E.g., Tresch, supra note 53, at 13-14 (1981). Some have interpreted the efficiency-equity tradeoff to reflect the idea that economic concerns may be traded off against concerns of justice, but that conception of the tradeoff is not defensible. Ronald Dworkin, Is Wealth a Value?, 9 J. Legal Stud. 191, 201-05 (1980). Instead, distributional concerns must serve as constraints for efficiency analysis.
102. The term is misleading because it assumes that equitable advances necessarily come at the expense of efficiency. The general problem that makes lump-sum transfers impossible also may make it impossible to achieve allocatively efficient outcomes, creating the possibility that regulations can yield outcomes that are more efficient and equitable than unregulated outcomes. See Louis Putterman, John E. Roemer & Joaquim Silvestre, Does Egalitarianism Have a Future?, 36 J. Econ. Literature 861, 862-65 (1998);
103. see also Isabel H. Correia, On the Efficiency and Equity Trade-Off, 44 J. Monetary Econ. 581 (1999) (using general-equilibrium analysis to show that some non-lump-sum redistributive measures can enhance allocative efficiency).
104. See supra note 53 and accompanying text.
105. See infra notes 96-97 and accompanying text.
106. Any other funding scheme would need to figure out the total resources that should be distributed to potential victims in each period, a daunting task. Moreover, a funding scheme not based on the WTA obligations of individual potential injurers will create further inequities. Some taxpayers will pay more than their WTA obligations, and some of these individuals will receive tax transfers less than their WTA risk proceeds. For such individuals, inequities in funding will exacerbate inequities in distribution.
107. Let P1 denote the probability that the potential victim will suffer a nonfatal injury, and let P2 denote the probability of fatal injury. The overall risk is P = P1 + P2. For reasons provided earlier, the plaintiff is fully compensated for the nonfatal injury if she receives tort damages equal to D = WTA/P1. See supra note 65. For these same reasons, to satisfy fully the obligation pertaining to fatal risks, the defendant must pay a fine to the state equal to D = WTA/P2.
108. For cases that settle, this funding scheme would cause a substantial increase in administrative costs if an additional judicial proceeding were required to calculate and collect the fine. Daniel L. Rubinfeld, On Determining the Optimal Magnitude and Length of Liability in Torts, 13 J. Legal Stud. 551, 553-57 (1984). Moreover, agency problems in the plaintiff-lawyer relationship render the theoretical attributes of these schemes indeterminate, making it "more difficult to use them as an effective policy tool."
109. Marcel Kahan & Bruce Tuckman, Special Levies on Punitive Damages: Decoupling, Agency Problems, and Litigation Expenditures, 15 Int'l Rev. L. & Econ. 175, 176 (1995).
110. See Mark Geistfeld, Should Enterprise Liability Replace the Rule of Strict Liability for Abnormally Dangerous Activities?, 45 UCLA L. Rev. 611, 627-33, 639-46 (1998) (providing reasons why strict liability can be expensive to administer and empirical data suggesting that shift from negligence to strict liability would increase total administrative costs);
111. see also John Fabian Witt, Note, The Transformation of Work and the Law of Workplace Accidents, 1842-1910, 107 Yale L.J. 1467, 1485 & n.96 (1988) (providing evidence that liability insurance rates for employers increased after adoption of workers' compensation);
112. John Fabian Witt, The Accidental Republic: Amputee Workingmen, Destitute Widows, and the Remaking of American Law, 1868-1922, at 350-52 (2000) (unpublished Ph.D. dissertation, Yale University) (providing further evidence that adoption of workers' compensation increased employer liability premiums).
113. The two forms of justice are described in Stephen R. Perry, Tort Law, in A Companion to Philosophy of Law and Legal Theory 57, 71-78 (Dennis Patterson ed., 1996).
114. Modified CBA therefore represents an important example of how an allocatively inefficient legal rule may be able to effectuate normatively desirable distributions in a more cost-effective manner than would the tax system. Cf. Chris William Sanchirico, Taxes Versus Legal Rules as Instruments for Equity: A More Equitable View, 29 J. Legal Stud. 797 (2000) (providing analytical reasons why allocatively inefficient legal rules may be more effective transfer mechanisms for purposes of distributive justice than tax system).
115. But cf. Louis Kaplow & Steven Shavell, Should Legal Rules Favor the Poor? Clarifying the Role of Legal Rules and the Income Tax in Redistributing Income, 29 J. Legal Stud. 821 (2000) (arguing that tax system is presumptively superior to allocatively inefficient legal rules for redistributing income from rich to poor).
116. In the presence of ideal redistributive institutions, the cost-benefit test can assume the marginal social value of money is equal for all individuals, as the ideal redistribution will equalize the marginal social value of money across individuals. When the required distributions cannot be made by some other institution, the marginal social value of money will not be equal for all individuals, so CBA must weight individual costs and benefits in terms of the marginal social value of money for the individual. See R. Layard & A.A. Walters, Income Distribution: Allowing for Income Distribution, in Cost-Benefit Analysis 179, 84-85 (Richard Layard & Stephen Glaister eds., 2d ed. 1994). This type of weighting is relied upon by modified CBA. Without a transfer of real money between potential injurers and victims, the marginal social value of money is higher for potential victims, who have not received their due, than for potential injurers, who have paid less than they should. The more exacting safety requirements imposed by modified CBA reduce the marginal social value of money for potential injurers, who are forced to spend more money on safety, while decreasing the marginal social value of money for potential victims, who are able to spend their WTA risk proceeds in a beneficial manner.
117. See Exec. Order No. 12,866, § 1(a), 3 C.F.R. 638, 639 (1993), reprinted in 5 U.S.C. § 601 (1994).
118. See supra Part II.B.
119. See supra Part IV.B.2.
120. The reason is that the Pareto principle "takes no interest whatever in distributional issues, which cannot be addressed without considering conflicts of interest and of preferences." Sen, supra note 51, at 352. Because modified CBA conforms to conventional CBA in the absence of distributional issues, it complies with the Pareto principle. Cf. Louis Kaplow & Steven Shavell, The Conflict Between Notions of Fairness and the Pareto Principle, 1 Am. L. & Econ. Rev. 63 (1999) (using contexts in which individuals are symmetrically situated to show how legal rules justified by nonwelfarist notions of fairness can violate Pareto principle).
121. Different forms of entitlements have different efficiency and distributive effects. E.g., Morris, supra note 54, at 847-49. Distributive considerations serve as a constraint on economic analyses such as CBA. See supra note 78 and accompanying text. As the distributive desirability of any given entitlement form cannot be determined solely with economic analysis, even the most ardent proponents of the economic analysis of law must acknowledge that any given entitlement form ultimately requires normative justification.
122. In a regime of strict liability with fully compensatory damages, potential injurers take the cost-benefit amount of care. See supra note 63.
123. See supra Part I.
124. Recall that the WTA measure is based on the premise that potential victims should not have to pay to protect themselves from nonconsensual risks. It includes the cost of risk aversion, and the amount by which it exceeds the WTP measure is largely attributable to the fact that money is a poor substitute for health or life. See supra Part II.A.
125. In evaluating this problem, whatever welfare loss an individual faces in her role as potential victim should be considered in conjunction with the welfare gain she receives in her role as potential injurer. By permitting some nonconsensual risks, modified CBA benefits all individuals in their roles as potential injurers. By mitigating the welfare loss of potential victims, modified CBA enhances the likelihood that everyone, on balance, benefits from this form of risk regulation.
126. The weighting of interests sanctioned by the safety principle underlies the various defenses to intentional torts involving property. See, e.g., Restatement (Second) of Torts § 77 (1965). Because intentional torts are the only area of tort law in which the safety principle has been recognized widely, some tort scholars have argued that the principle is not relevant for accidental injuries. See David G. Owen, The Moral Foundations of Products Liability Law: Toward First Principles, 68 Notre Dame L. Rev. 427, 469-70 (1993). An obvious problem with this argument is that defenses to intentional torts turn on the question of whether the conduct was reasonable. For example, absent a threat to personal safety, it is not reasonable to cause serious harm to another in defense of property, because safety interests are more important than property interests. Restatement (Second) of Torts § 77 cmt. i; Keeton et al., supra note 7, § 21, at 132-33. The question of reasonableness, which addresses the mediation of normatively acceptable, competing interests, is central to negligence law. Hence, tort law's invocation of the safety principle to determine the issue of reasonableness for intentional torts is likely to be relevant in the negligence context. In particular, by elevating safety interests over property interests in the intentional-torts context, the tort system sanctions the use of the WTA measure for accidental harms, as this risk measure corresponds to an entitlement that gives such priority to the respective interests. See supra Part III.A. The WTA measure, with its distributive implications, forms the basis of the safety principle as implemented by modified CBA.
127. Restatement (Second) of Torts § 901 cmt. a (1979); see also Keeton et al., supra note 7, § 2, at 7 (stating that "primary purpose [of tort law] is to compensate for the damage suffered").
128. Restatement (Second) of Torts § 901 cmt. a; see also Keeton et al., supra note 7, § 2, at 9 (noting that "[t]he idea of punishment, or of discouraging other offenses, usually does not enter into tort law").
129. See Perry, supra note 86, at 66-68.
130. See Mark Geistfeld, Economics, Moral Philosophy, and the Positive Analysis of Tort Law, in Philosophy and U.S. Tort Law (Gerald Postema ed., forthcoming 2001). A more forceful critique would focus on the inability of CBA to determine initial entitlements, which implies that CBA is incapable of determining every aspect of tort law. Cf. id. (arguing that any moral theory of tort law can be translated into social welfare function).
131. See supra notes 46-47 and accompanying text.
132. E.g., Jules L. Coleman, Risks and Wrongs 374-85 (1992); Weinrib, supra note 6, at 48 (concluding that "structure [of tort law] is precisely what economic analysis ignores"). Some moral philosophers find this to be the most persuasive critique of the economic interpretation of tort law. Perry, supra note 86, at 66.
133. Restatement (Second) of Torts §§ 892-892D (1979).
134. See Restatement (Second) of Torts § 283 cmt. e (1965) (stating that interests advanced by injurer's conduct do not depend upon actor's subjective evaluation, but rather on "the value which the law attaches to [them]"); id. § 291 (stating that party "is not excused because he is peculiarly inconsiderate of others . . . nor is he negligent if his moral or social conscience is so sensitive that he regards as improper conduct which a reasonable man would regard as proper").
135. For example, CBA can be based on measures of individual well-being that need not correspond to individual preferences. See generally Adler & Posner, supra note 41.
136. 3 Fowler V. Harper et al., The Law of Torts § 12.1 (2d ed. 1986).
137. See supra notes 59-60 and accompanying text. I qualify this claim because the case adopting this standard involved contributory negligence. See United States v. Carroll Towing Co., 159 F.2d 169 (2d Cir. 1947). The amount of care someone exercises for her own protection is not governed by the safety principle. See supra text accompanying note 40.
138. Weinrib, supra note 6, at 147-52; Stephen G. Gilles, The Invisible Hand Formula, 80 Va. L. Rev. 1015, 1026 n.28 (1994); Keating, supra note 6, at 349-60;
139. cf. B.S. Markesinis & S.F. Deakin, Tort Law 146 (3d ed. 1994) ("It is fair to say that the 'Hand formula', loosely conceived, is an approach followed by the English courts in appropriate cases.").
140. E.g., Richard W. Wright, The Standards of Care in Negligence Law, in Philosophical Foundations of Tort Law, supra note 47, at 249, 260-61.
141. See, e.g., Landes & Posner, supra note 64, at 86-88 (arguing that Hand formula minimizes accident costs and that "something like [it] has long been used to decide negligence cases").
142. See Weinrib, supra note 6, at 147-52; Keating, supra note 6, at 349-60; Wright, supra note 111, at 274-75.
143. Gilles, supra note 110, at 1017 & n.6.
144. Id. Judge Richard Posner's preference for the Hand formula is beginning to influence courts, however. See U.S. Fid. & Guar. Co. v. Plovidba, 683 F.2d 1022 (7th Cir. 1982) (Posner, J.) (adopting Hand formula for admiralty actions).
145. The particular weighing scheme embodied in modified CBA does not accord with the disproportionate standard as articulated by some tort scholars. Some have argued that this standard renders cost considerations irrelevant when the risk is substantial. See Weinrib, supra note 6, at 149; Wright, supra note 111, at 261-63. A fundamental problem with this argument is that it fails to explain why the rule of strict liability for "abnormally dangerous" activities does not depend solely on the level of risk, nor does it explain why English courts have not applied strict liability to such activities. See Geistfeld, supra note 85, at 623 n.43. Another formulation of the disproportionate standard holds that proportionately greater weight should be given to security interests as the risk increases. See Keating, supra note 6, at 349-60. This standard might be consistent with modified CBA. Consider a risk, defined in terms of the WTA measure, that receives twice as much weight under modified CBA as ordinary economic interests. The WTA measure increases with risk at an increasing rate. See infra app. § 1. Consequently, if the WTA measure is $100 for a risk of 1 in 10,000, it could be $250 for a risk of 2 in 10,000. For the smaller risk, potential injurers would have to spend up to $200 on precaution costs (twice the WTA measure of $100), whereas they would have to spend up to $500 once the risk doubles. The required safety expenditures therefore increase at a higher rate than do increases in risk. Under modified CBA, then, economic interests face proportionately higher burdens as the risk increases, even though there is no change in the relative weight given to security interests and ordinary economic interests.
146. A survey of 100 judges found that most applied the negligence standard in a manner consistent with CBA for cases involving property damage, whereas all judges applied the negligence standard in a manner that favored safety over money for an otherwise identical case (in cost-benefit terms) involving serious personal injury. W. Kip Viscusi, How Do Judges Think About Risk?, 1 Am. L. & Econ. Rev. 26, 40-46 (1999).
147. Studies have found that lay individuals find the following contextual features to be relevant to the valuation of risk: (1) [T]he catastrophic nature of the risk; (2) whether the risk is uncontrollable; (3) whether the risk involves irretrievable or permanent losses; (4) the social conditions under which a particular risk is generated and managed, a point that connects to issues of consent, voluntariness, and democratic control; (5) how equitably distributed the danger is or how concentrated on identifiable, innocent, or traditionally disadvantaged victims, which ties to both notions of community and moral ideals; (6) how well understood the risk process in question is, a point that bears on the psychological disturbance produced by different risks; (7) whether the risk would be faced by future generations; and (8) how familiar the risk is. Pildes & Sunstein, supra note 22, at 57; see also W. Kip Viscusi, Corporate Risk Analysis: A Reckless Act?, 52 Stan. L. Rev. 547, 552 (2000) (describing study which found that mock jurors tend to reject corporate decisions involving risks to others that are based on CBA).
148. See Restatement (Second) of Torts §§ 519-520 (1977).
149. "The choice between negligence and strict liability is fundamental; yet after centuries of debate, no clear choice has been made." Richard A. Epstein, Torts § 4.1 (1999).
150. See supra note 63 and accompanying text.
151. See Geistfeld, supra note 85, at 646-58 (arguing that rule of strict liability for abnormally dangerous activities is most plausibly understood in terms of its ability to reduce risk below level attainable by negligence standard for contexts in which plaintiffs have difficulty proving lack of reasonable care); Mark Geistfeld, Tort Law and Criminal Behavior (Guns), 43 Ariz. L. Rev. (forthcoming 2001) (showing how deterrence rationale for strict liability is required if rule is to be consistent with negligence for third-party criminal behavior). Notice that this interpretation of strict liability also suggests strict liability may be appropriate if it leads to the same risk level as negligence, and potential victims would be compensated inadequately under a negligence rule. A good example of such a case involves reasonable behavior that creates a high degree of nonreciprocal risk and only implicates the interests of the two parties to the lawsuit. See id. Together, these roles for strict liability persuasively explain the rule of strict liability for abnormally dangerous activities. See id.
152. See supra note 63.
153. The risk-utility test determines design and warning defects. Restatement (Third) of Torts: Products Liability § 2 cmts. d, i (1997).
154. E.g., id. cmt. f ("[A]n alternative design is reasonable if its marginal benefits exceed its marginal costs.").
155. See supra Part IV.B.3. Additional consideration must be given to bystanders - those who do not benefit from the product use - as these individuals may monetize the product risk in terms of the WTA measure, depending on the degree of reciprocity. As long as the vast majority of product-caused injuries are suffered by users, the risk-utility test ordinarily will approximate closely the appropriately modified cost-benefit test.
156. See Mark Geistfeld, Products Liability, in 3 Encyclopedia of Law and Economics 347 (Boudewijn Bouckaert & Gerrit De Geest eds., 2000).
157. See Restatement (Third) of Torts: Products Liability § 2 cmt. d (1997) ("Assessment of a product design in most instances requires a comparison between an alternative design and the product design that caused the injury, undertaken from the viewpoint of a reasonable person.");
158. Mark Geistfeld, Inadequate Product Warnings and Causation, 30 U. Mich. J.L. Reform 309, 322-29 (1997) (showing how legal requirements for adequate product warnings depend on informational needs of average consumer).
159. See supra notes 2-4 and accompanying text.
160. For extensive documentation of this point, see Viscusi, supra note 118, at 552-57 (describing mock juror study); id. at 568-78 (noting examples from case law).
161. One study of 100 judges found that over two-thirds of them felt that
162. See, e.g., Viscusi, supra note 118, at 589 (proposing that "responsibility for deterring corporate misbehavior" be moved from courts to regulatory agencies).
163. See supra notes 116-18 and accompanying text.
164. The most promising solution is to rely on the consumer-expectations test framed in a manner that defines expectations in terms of consumer preferences for safety investments satisfying the cost-benefit test. Any other formulation of consumer expectations involves an inherent ambiguity - expectations of risk or safety? - that creates inconsistent results. See Geistfeld, supra note 127, at 367-68.
165. A point that applies to other tort issues. See Geistfeld, supra note 128, at 329-35 (showing poor guidance provided by current jury instructions for product warnings and providing set of more useful jury instructions); Geistfeld, supra note 60, at 841-43 (showing poor guidance provided by current jury instructions for pain and suffering damages and providing more useful jury instruction).
166. See, e.g., John D. Graham, Laura C. Green & Marc J. Roberts, In Search of Safety: Chemicals and Cancer Risk (1988) (discussing methods regulators used to assess risks posed by formaldehyde and benzene).
167. See supra notes 15-17 and accompanying text. 138 See id.
168. The history of this trade dispute is described in George H. Rountree, Note, Raging Hormones: A Discussion of the World Trade Organization's Decision in the European Union-United States Beef Dispute, 27 Ga. J. Int'l & Comp. L. 607, 611-12, 632-33 (1999).
169. The European Union, like other jurisdictions, requires labeling for genetically modified foods. Ruth MacKenzie & Silvia Francescon, The Regulation of Genetically Modified Foods in the European Union: An Overview, 8 N.Y.U. Envtl. L.J. 530 (2000).
170. Science of Banning American Cows Too, Birmingham Post (Eng.), Oct. 14, 1999, at 26, 1999 WL 25286371.
171. Communication on the Precautionary Principle, supra note 17, at 11. For a description of international environmental policies relying on the precautionary principle, see id. at 26-28.
172. Editorial, Fear of the Future, Wall St. J., Feb. 10, 2000, at A18.
173. Alan P. Larson, Remarks at Iowa Governor's Conference on International Trade (transcript available in U.S. Dep't of State Dispatch, June 1, 1999, at 18, 1999 WL 19912583).
174. Communication on the Precautionary Principle, supra note 17, at 9. The European Commission initiates proposals for all new legislation in the European Union.
175. Id. at 20.
176. Id. at 8; see also id. at 13 (discussing same basic criteria for applying precautionary principle).
177. Id. at 18.
178. Id.
179. Id. at 19.
180. See id. at 20.
181. Fear of the Future, supra note 143, at A18.
182. when in Doubt . . ., Bus. Eur., Feb. 9, 2000, at 6, 6.
183. See supra Part I.
184. Communication on the Precautionary Principle, supra note 17, at 15.
185. The probability and severity of injury could be expressed as probability distributions over the range of possible outcomes, with each outcome weighted by the probability of its being true. That is merely one approach to the problem, however, as discussed below.
186. Justice Breyer advocates this approach to risk regulation. See Stephen Breyer, Breaking the Vicious Circle 55-81 (1993).
187. See supra note 143 and accompanying text.
188. The relevant tort norm, by contrast, tries to apportion the burden of factual uncertainty equally between the parties, and can justify (nonideal) outcomes in which potential victims disproportionately bear the cost of factual uncertainty. See Mark Geistfeld, Scientific Uncertainty and Causation in Tort Law, 54 Vand. L. Rev. (forthcoming 2001) (arguing that tort norms justify requirement that plaintiffs establish causation with epidemiological proof in cases of scientific uncertainty).
189. See id. (explaining why disclosure of scientific uncertainty about risk can be material to average consumer's decision of whether to purchase or use particular product).
190. See supra note 144 and accompanying text.
191. The safety threat could be intergenerational if the hormones affect future generations. But consumers would consider any adverse health affects on their progeny, making it unnecessary to account for those interests separately in the quantification of costs and benefits.
192. See Rountree, supra note 139, at 610 (describing studies); id. at 624-25 (describing finding by WTO panel).
193. See id. at 609.
194. The requirement also can be justified on grounds of allocative efficiency. See Geistfeld, supra note 159.
195. See supra note 140 and accompanying text. Consistency in labeling has been difficult to achieve due to widespread but different uses and levels of genetically modified food ingredients. See Steve Stecklow, 'Genetically Modified' on the Label Means . . . Well, It's Hard to Say, Wall St. J., Oct. 26, 1999, at Al.
196. See Carol Kaesuk Yoon, Altered Salmon Leading Way to Dinner Plates, but Rules Lag, N.Y. Times, May 1, 2000, at Al (stating: A recent study showed, for example, that populations of wild fish could, in theory, be wiped out by mating with certain kinds of genetically engineered fish, should they escape. In addition, there is the possibility of unpredictable environmental disruptions, like those that occur when non-native species invade ecosystems, as the zebra mussels have the Hudson River.).
197. See Brandon Mitchener, Biosafety Agreement Raises Question, Wall St. J. Eur., Jan. 31, 2000, at 4, 2000 WL-WSJE 2944258 (discussing role of Precautionary Principle in Biosafety Protocol); see also supra note 17 (providing sources on Biosafety Protocol).
198. See supra note 27 and accompanying text.
199. v). When the potential victim does not receive the WTA risk proceeds (WTA =0?), this equation makes no sense.