The Role of Congress in Risk Management

Harold P. Green

Harold Green is Professor of Law and Associate Dean for Post-J.D. Studies at the National Law Center, The George Washington University, Washington, D.C.

[16 ELR 10220]

Congress is the primary authority guiding the risk-management decisions of the courts and regulatory agencies, whose chief role in risk management is implementation of the policies established by Congress. Federal regulatory agencies are creatures of Congress and exercise only such authority as Congress bestows upon them; they assess and manage risk according to criteria and standards enacted by Congress. The courts are subject to similar constraints, except, of course, when constitutional issues are involved.

Congress, moreover, can always reverse, overrule, or otherwise correct risk-management decisions made by either the courts or the agencies. Indeed, Congress can influence decisively the actions of regulatory agencies — and sometimes even of the courts — through measures less drastic than enactment of legislation. These measures include antagonistic hearings, adverse committee reports, and the power of the purse.

My topic, however, is not the exercise of political power by Congress, but rather the way Congress has gone about establishing criteria for managing risk.

Risk management by Congress is more a political than a scientific, or even a rational exercise. With respect to risk management for a particular activity, Congress must first determine whether the risk involved is either acceptable or unacceptable from the standpoint of society. Where the risk is deemed unacceptable, the most elementary mode of congressional risk management is flat prohibition of the activity. For example, statutes may prohibit the sale, use, or possession of items, such as heroin and firearms, or may forbid certain acts, such as the manufacture of nuclear weapons. These statutes reflect a legislative judgment that the risks associated with these activities are of such magnitude that the activity should be banned notwithstanding any possible offsetting benefits.

In most cases the activity banned is thought to have little, if any, social utility. There are, however, some instances in which an activity is flatly prohibited, even though substantial offsetting benefits are present. The Delaney Clause, for example, prohibits the use of any chemical in food that is known to cause cancer in animals.

The evidence upon which Congress bases its determination that an activity involves an unacceptable risk is often more anecdotal than empirical. Usually, Congress makes no attempt to quantify either risk or benefit. There is, moreover, often no satisfactory explanation why the legislature prohibits some activities involving risk but not others. It appears anomalous, to say the least, that a statute might prohibit the sale or possession of marijuana or pornography but not of liquor or tobacco.

In most situations involving benefit as well as risk, the resultant statute reflects an implicit balancing of the two. For example, although the use of a motor vehicle involves great risks, legislatures have recognized the importance of preserving the benefits of the technology, while at the same time minimizing the risks. This balancing process has led to numerous statutes that permit the use of motor vehicles, subject to various restrictions designed to reduce risk, such as minimum driving age and safety equipment requirements.

Although scientific analysis and quantification may have some input into legislative consideration of risk reduction strategies, the final statute will usually be influenced more heavily by subjective and political factors than by objective, scientific ones. Each legisator has his or her own experience with and attitudes towards risk and its consequences. These attitudes, coupled with the attitudes of constituents and pressures applied by special interest groups, shape the manner in which the legislator will vote.

Presumably, the way the legislature acts on a particular question reflects public sentiment about the risk and its acceptability. We are all prepared, I believe, to accept the fact that in a democracy Congress makes decisions and enacts statutes on the basis of emotional and political factors, and that this practice may result in a seemingly irrational fabric of law. We seem less willing to accept, however, that a regulatory statute almost always reflects a compromise among competing considerations. It will seldom be seen, therefore, as optimum from any one point of view.Sophisticated observers of the regulatory process know, for example, that when Congress enacts a coal mine safety statute, even though the rhetoric of the stated goals may be extremely tough, the operative provisions of the legislation may, in reality, be quite soft.

Flat prohibition of an activity is not the customary style of Congress. Rather, Congress usually prefers to enact regulatory statutes containing standards for risk management to be implemented by regulatory agencies. Examination of these statutes, however, will not help us to understand how Congress goes about the business of risk management. If there is any consistent thread running through these statutes, it is a thread of inconsistency.

This thread of inconsistency is neither accidental nor casual, but is inherent in the intensely political nature of the legislative process. Congress normally acts to regulate hazardous activities and products only after it has become abundantly clear to a sizable number of legislators that the risks require regulation. In most cases, the regulatory legislation languishes in Congress for several sessions before final action is taken. For example, it took at least six years to enact the Chemical Food Additives Act¹ and the Toxic Substances Control Act (TOSCA)².

By the time Congress gets around to considering risk management legislation, the benefits of the technology have been well established, and there are powerful vested interests to resist the contemplated regulation. When I speak of vested interests, I do not use that term pejoratively. I am referring [16 ELR 10221] not only to business interests but also to the interests of consumers, for example, cigarette smokers. Congress thus finds itself squeezed by pressures from the Administration, from business interests, and from citizen and consumer groups. Almost inevitably, the legislation represents a compromise that gives industry some comfort and throws tough issues to the agencies and courts, where they will remain unresolved for several years. In this connection, I invite you to consider Justice Rehnquist's dissent in American Textile Manufacturers Institute v. Donovan.³

I cannot, therefore, offer any concrete exposition on how Congress manages risk. Instead, I will offer several examples of how Congress has approached risk management in the past. These examples demonstrate that the management of risk by Congress is a haphazard affair, dependent largely upon the strength of vested interests and the prevailing political winds. Before I proceed, let me emphasize that these remarks are not intended to be critical of Congress. The process I describe is par for the course in a democratic society and probably represents the best that we can expect.

Let us turn to several illustrations of the legislative approach to safety regulation:

Chemical Food Additives.

The Chemical Food Additives Amendments to the Food, Drug and Cosmetics Act were enacted in 1958⁴. Under this legislation, a chemical — except one "generally recognized as safe" — is "deemed to be unsafe" and may not be added to food until it has been pretested and falls within a Food and Drug Administration regulation prescribing the conditions under which it may be "safely used."⁵ Although the statute does not attempt to define "safely used," the report of the Senate Committee on Labor and Public Welfare⁶ defines "safety" to require "proof of a reasonable certainty that no harm will result from the proposed use of an additive."

Despite the Committee's gloss on the definition of "safely used," considerable ambiguity and latitude remain in interpretation. What seems clear is that an additive may be used only if its proposed use is shown with reasonable certainty to cause no harm. One can, however, debate the meaning of "reasonable certainty" and "use." The word "harm," moreover, embraces a broad spectrum of possible, probable, and/or actual consequences.

The statutory language can be interpreted further by reference to the specific language of the Delaney Amendment, which flatly prohibits the use of any chemical additive "found to induce cancer when ingested by man or animal."⁷ As is generally known, this provision has been interpreted to prohibit the use of any chemical found to cause cancer in animals, even where animals have been fed enormously large quantities of the chemical (relative to quantities likely to be ingested by man). We can infer, therefore, that a somewhat more relaxed standard may apply to noncarcinogenic chemicals.

Of equal importance in interpreting the concept of safety found in the Act and in the legislative history is the basic procedure for approval of additives set forth in the statute itself. The statute provides that an additive may not be used until the FDA promulgates a rule allowing its use. An applicant has the burden of showing that an additive may be "safely used." As long as its decision is not arbitrary or capricious, the FDA has no burden to show that any harm will in fact result. The approval process, in effect, reflects a bias against additives and a lack of concern for the loss of any benefits incident to their use.

Flammable Fabrics.

The Flammable Fabrics Act⁸ banned the movement in interstate commerce of fabrics that do not meet the flammability standards prescribed by the agency. The Act requires that the standard for determining the acceptability of risk be based on findings that the standard "is needed to adequately protect the public against unreasonable risk of the occurrence of fire … (and) is reasonable, technologically practicable, and appropriate."⁹

This standard of acceptable risk is obviously less severe than that of the Chemical Food Additives legislation. The Chemical Food Additives Amendments bar any harm, while the Flammable Fabrics Act bars only unreasonable risk of harm. The former, moreover, prohibits treatment of food with chemicals whose safety has not been proven, whereas the latter requires treatment of fabrics with chemicals to achieve adequate safety, where such treatment is "reasonable, technologically practicable, and appropriate …."¹⁰ Thus, while economic considerations are not relevant to approval of food additives, they must be taken into account in the regulation of flammable fabrics.

As with the Chemical Food Additives Amendments, consideration of the locus of the burden of proof will be instructive. The Flammable Fabrics Act is in no sense a licensing statute. No permission or license is required to ship flammable fabrics in interstate commerce, but one will be subject to criminal prosecution if one's products do not meet the flammability standards. The government therefore has the burden of proof; moreover, the government has the burden of showing that its flammability standards are "reasonable, technologically practicable, and appropriate."¹¹ One would thus expect the standard of acceptable risk to be much less severe with respect to fabrics than to food additives.

Toxic Substances.

The Toxic Substances Control Act¹² (TSCA) authorizes the Environmental Protection Agency (EPA) to regulate any chemical upon a finding of "unreasonable risk of injury to health or the environment."¹³ Here, some degree of risk is clearly acceptable, since only "unreasonable" risk triggers regulation. Since it specifies that the regulation should impose the "least burdensome requirements" necessary to protect against "unreasonable risk," the statute would seem to contemplate some degree of EPA tolerance toward risk.

The criterion for "unreasonable risk" is identical to that found in the Flammable Fabrics Act, but the report of the House Commerce Committee¹⁴ provides a rather explicit definition of "unreasonable risk" for the purposes of TSCA:

[the determination of] unreasonable risk [involves] balancing the probabilities that harm will occur and the magnitude and severity of that harm against the effect [16 ELR 10222] of proposed regulatory action on the availability to society of the benefits of the substance or mixture, taking into account the availability of substitutes for the substance or mixture which do not require regulation, and any other adverse effects which such proposed action may have on society.

The report expressly negates any requirement for a "formal cost-benefit analysis under which a monetary value is assigned to the risks … and to the cost to society."

These three statutes reflect the typical congressional approach to safety regulation. Each was enacted after several years of legislative effort. In all three cases, moreover, the legislative history includes ample documentation — a veritable parade of horribles — of the evils facing the public under the status quo. In each case, the industry to be regulated fought the legislation, and, typically, the ultimate legislation contains compromise language designed to appease industry and to curb the zeal of the regulatory agency.

Despite statutory language and legislative histories that contemplate the applicability of varying standards of acceptable risk, the regulatory agency in each of the examples above clearly has a mandate to prevent and to correct evil. The basic presumption underlying al three regulatory programs is that elimination of harm to society is more important than societal enjoyment of the full benefits of the regulated activity. Within the framework of this presumption, there are, nevertheless, some interesting distinctions that may be drawn among the three cases.

In the instance of chemical food additives, the presumption is that an additive is unsafe until it has been shown to be safe. The burden of persuasion rests with industry, and under the statute, FDA has virtually unrestrained discretion to determine the conditions under which use of an additive will be permitted. Under TSCA, the burden likewise falls on industry, but the statutory language constrains EPA and limits its discretion.

Under the Flammable Fabrics Act, on the other hand, a fabric may be used unless and until its use has been prohibited by regulation. The agency, not the industry, must take the initiative; the agency must also assume the burden of establishing that the regulation is necessary to achieve adequate safety, and that it is "reasonable, technologically practicable, and appropriate."

Finally, it should be observed that the considerations outlined above are subject to judicial review and, more importantly, to constant monitoring by Congress — or, at least, by those members of Congress who have an interest from the standpoint of either public health and safety or of industry.

The statute, once enacted, is not the end of the lawmaking process. Frequently, legislators who are dissatisfied with the compromise language included in a statute will seek in succeeding years to improve the legislation by amendment. Moreover, the words used in the statute are subject to interpretation through application by the agency and the courts on a case-by-case basis. The agency's actions are subject to continuing legislative oversight, and an interpretation that conflicts with the views of an influential member of Congress is likely to meet resistance in the form of a legislative hearing, a public rebuke, or an effort to cut the agency's funds.

The agency's discretion is thus constrained not only by the language of the statute but also by the necessity to keep agency decisions within the ambit of political reality. This necessity arises from the desire of agency personnel to protect themselves and to avoid jeopardizing their chances for advancement in the bureaucracy or the world of politics. The ultimate constraint, of course, is Congress's power to enact legislation to correct what it deems to be improper regulation.

The legislative examples above may be compared instructively with the atypical case of nuclear power. The Atomic Energy Act of 1954¹⁵ evidences an obsessive concern with the "health and safety of the public." This phrase, or a close equivalent, is used no fewer than 30 times in the original Act, and numerous additional references appear in the amendments that have been enacted since 1954.

The Atomic Energy Act, nevertheless, contains only the most general standard as to safety: it provides that no license for a nuclear power plant will issue if the Nuclear Regulatory Commission determines that issuance of the license would be "inimical to the health and safety of the public." This standard seems to require a higher threshold of acceptable safety, since "inimical" refers to the tendency to cause harm rather than to actual harm itself.

The agency rules provide two other criteria of acceptable safety, which may be inferred from the language of the statute. In the case of a construction permit, the criterion is "reasonable assurance that … the proposed facility can be constructed and operated at the proposed location without undue risk to the health and safety of the public."¹⁶ In the case of an operating license, the criterion is "reasonable assurance that the activities … can be conducted without endangering the health and safety of the public."¹⁷

"Without undue risk" is probably a more relaxed standard than that found in the Chemical Food Additives Amendments, but it is substantially equivalent to the standard of "unreasonable risk" found in the Flammable Fabrics Act and in TSCA. Indeed, "undue" seems to invite a balancing of risks against benefits. On the other hand, "without endangering" seems to be at least as severe a standard as "safely used," and approaches in severity the standard of "inimical to the health and safety of the public."

The Atomic Energy Act has thus spawned three separate criteria for acceptable safety, which from the semantic standpoint, at least, possess differing value connotations. In actual practice, the courts have treated these criteria as virtually identical.¹⁸ The failure to distinguish among these criteria is less a product of analysis than of political circumstance. Indeed, the prevailing view is that determination of acceptability of risk does not require any balancing of benefit against risk, since the mere fact that the statute itself seeks to promote nuclear power constitutes a congressional determination that each nuclear power plant will have compelling benefits. The licensing process thus focused solely on the question, "how safe is safe enough?"

The Atomic Energy Act of 1954 stands in sharp contrast to the three statutes discussed above and, indeed, to all other health and safety regulatory schemes. Unlike the typical regulatory statute that spends many years in gestation, less than two years elapsed between conception and enactment of the Atomic Energy Act, and only seven months elapsed between introduction of the first bill and its enactment. Moreover, the Act was in no sense the product of perceived [16 ELR 10223] abuse of the public interest: there was no parade of horribles and no record of injuries. Indeed, despite the Act's apparent obsession with the health and safety of the public, there is scarcely a word in the legislative history about the nature or magnitude of the possible hazards.

When the 1954 Act established a regulatory scheme for nuclear power, nuclear power was not yet in existence. The basic purpose of the Act was more promotional than regulatory. The Act established a framework for dealing with risks that were expected to arise when a nuclear power technology was created. The regulatory provisions of the Act were intended to ensure that, when the time was ripe, the industry would be regulated appropriately. It was clear, as well, that the regulation was to sufficiently benign so as not to strangle the embryonic technology.

Notwithstanding the constructive influences of its technically competent arms, such as the Office of Technology Assessment, the Congressional Research Service, and the General Accounting Office, Congress, at least when judged by the statutes it enacts, does not assess or manage risk in a scientific, consistent or even rational manner. In my view, this state of affairs should not be cause for criticism or concern; it merely reflects the reality that, in the public policy arena, risk assessment is, and should be, essentially a political rather than a scientific exercise.

In conclusion, it should be stressed that in the context of public policy decisionmaking, risk management is intrinsically a socio-economic political question. Although the elements of the risk management process may be stated in objective, scientific terms, these terms cannot be reduced to meaningful numbers having more than ephemeral value in the decisionmaking process. There is no doubt that our nation could benefit from improved methods for determining the acceptability of risk, but in a democratic society such improvement can come about only through an overall betterment of the political process.

1. 21 U.S.C. § 321

2. 15 U.S.C. §§ 2601-2629, ELR STAT. 4135.

3. 452 U.S. 490, 11 ELR 20736 (1981).

4. Pub. L. No. 85-929, 72 Stat. 1784 (1958).

5. 21 U.S.C. § 348(a).

6. S. Rep. No. 2422, 85TH Cong. 2d Sess. (1958).

7. 21 U.S.C. § 348(c)(3).

8. Pub. L. No 90-189, 81 568-573 (1967).

9. 15 U.S.C. § 1193(b).

10. Id.

11. Id.

12. 15 U.S.C. §§ 2601-2629, ELR STAT. 4135.

13. 15 U.S.C. § 2603(a).

14. H.R. Rep. No 1341, 94TH Cong. 2d Sess. (1976).

15. 42 U.S.C. §§ 2011-2282, ELR STAT. 41805.

16. Power Reactor Development, Co. v. International Union, 367 U.S. 396, 407 (1961).

17. Maine Yankee Atomic Power Co., 6 A.E.C. 1003, 1004 n.4 (1973).

18. Citizens for Safe Power v. NRC, 524 F.2d 1291, 6 ELR 20095 (D.C. Cir. 1975).