Environmental Risk Analysis: A Review of Public Policy Issues IV

98-618 ENR

CONTENTS FOR THIS SECTION

Analysis

Issues in Environmental Risk Management

How Valuable Is the Information Provided?
Is It a Scientific Basis for Environmental Decisions?
Should It Be Used to Quantity Environmental and Health Benefits?

Analysis

Issues in Environmental Risk Management

According to panelists in the 1993 CR5 workshop on risk and cost-benefit analysis, several developments have spurred congressional interest in the potential utility of environmental risk analysis for informing risk management decisions, including: complaints about inflexible, unprioritized, and "unfunded federal mandates" imposed on state and local governments; the growing cost of compliance with environmental requirements to regulated industries; the need to reduce the budget deficit while reauthorizing several of the major environmental statutes; and consideration of proposals to elevate EPA to departmental (that is, cabinet) status amidst allegations of inefficient and ineffective EPA programs. Such developments led many to conclude that federal managers responsible for environmental protection were not doing enough to control costs and should be held more accountable; they believe EPA should be required to use risk analysis and economic analysis to demonstrate that proposed regulations will efficiently reduce serious risks to human health or the environment.

These proponents of risk analysis suggested that it could serve risk management in two general ways: 1) as a basis for comparing and ranking environmental hazards, permitting assignment of priorities for regulatory action, and 2) as a basis for evaluating the effectiveness of specific regulations in reducing risks relative to the costs of compliance and implementation. Debates about these general approaches and more specific legislative proposals promoting risk analysis revolved around five general issues: 1) How valuable is the information provided by risk analysis for policymakers? 2) Is risk analysis a scientific basis for environmental decisions? 3) Should risk analysis be used to quantify environmental and health costs and benefits? 4) Should EPA priorities be based on relative risks and risk reduction opportunities? and 5) Given that EPA already does risk analysis and economic analysis, would additional requirements for analysis improve risk management?

How Valuable Is the Information Provided? Most observers seem to agree that risk analysis is a potentially valuable tool for summarizing scientific information about the potential human health effects of exposure to an environmental hazard.31 EPA Administrator Browner and Dr. Lynn Goldman, Assistant Administrator for Pollution Prevention, Pesticides and Toxic Substances, have testified repeatedly that they believe risk analysis is a useful tool. Former Administrator Reilly and EPA's Science Advisory Board (SAB), a group of independent scientists appointed by the Administrator who review the scientific bases for EPA's decisions, also have praised risk analysis. The National Academy of Sciences, U.S. General Accounting Office, several state and local government associations, regulated industries, and many academics support the use of risk analysis by federal agencies.

The apparently widespread agreement regarding the potential value of risk analysis breaks down, however, when questions arise about how risk analysis should be used and how much influence it should have on federal regulatory decisions. Opinions range along a continuum. At one end is the belief that risk analysis and economic analysis should be conducted to inform every regulatory decision; EPA should first regulate hazards found to pose the greatest risks, and the reduction in risk due to the regulation should be large enough to justify the cost of implementation and compliance. At the other end of the continuum is the view that EPA should regulate all environmental hazards posing risks unacceptable to the public, unless those responsible for creating the risk (or those profiting from it) demonstrate that it is insignificant or justified, for example, by the benefits provided and a lack of safer alternatives. In this view, chemical releases to the environment are presumed to be unacceptable and should be eliminated and the polluter should internalize the costs of compliance as well as for toxicity testing and risk analyses, regardless of cost.

Other opinions about the value of risk analysis fall between these two. For example, environmentalists and environmental justice activists generally oppose legislation promoting EPA analysis of risks, costs, and benefits, because they fear increased attention to risks will reduce consideration of other important information, for example, about pollution prevention opportunities or potential ecological or aesthetic impacts of management options. Many of these groups prefer the current level of attention to risk, or less. In contrast, state and local governments and regulated industries generally support a greater role for risk and economic analyses in EPA's risk management decisions which they believe will help control compliance costs.

The CRS workshop panelists generally agreed that decisionmakers need better information about risks, costs, and other aspects of regulatory decisions, including consequences of regulatory options that are not quantifiable, and thought that risk analysis might provide some of that needed information, if it is appropriately adapted to decision makers' needs. For example, they noted that risk analysis is useful for such analytic purposes as clarifying trade-offs (that is, choices among mutually exclusive options each of which has both good and bad potential consequences.)

The CRS workshop panelists disagreed about whether risk analysis as currently practiced should be promoted through legislation. Disagreements stem in part from the fact that the information value of risk analysis is highly variable, depending on factors discussed in detail in a later section of this report.

Is It a Scientific Basis for Environmental Decisions? Some promote risk analysis because they believe it is an objective scientific basis for environmental policies and management. In their opinion, risk analysis should be used to inform federal agencies, Congress, and the public, in the hope that it will lead to rational decisions and environmental protection strategies, and replace what these observers regard as a piecemeal environmental policy that developed in response to real and imagined crises. They favor legislation mandating risk analysis by EPA and reports to Congress.

Opponents of such legislation argue that a mandate for risk analysis will not improve environmental decisions, because it is neither pure science nor entirely objective and, they assert, it is easily manipulated for political or venal purposes. Due to inherent uncertainties, risk analysis is not, and probably can never be, entirely objective or fact-based, although it generally is characterized as a scientific activity.32

The CRS workshop panelists agreed that the results of risk analyses are always debatable, and therefore, reliance on the results of risk analyses in regulating perpetuates debates over how to regulate. In addition, they noted that when risk is managed by comparing risk estimates for different problems or for regulatory options, value judgments are necessary. This issue is discussed under the heading "Should Priorities Be Based on Relative Risks and Risk Reduction Opportunities?"

Opponents of legislation mandating risk analysis by EPA further contend that the science used in risk analysis is immature and only is validated for assessing the risk of developing cancer. In addition, they maintain that, for most chemicals, health effects, and ecological effects, data do not exist because scientists have not done the necessary studies, and without data, risk analysis is meaningless. Even when data are available, they argue that human data are usually only from studies of adult white males with occupational exposures, and animal studies are insensitive to risks affecting fewer than 1% of test animals.

CRS workshop panelists agreed that risk analysis focuses attention on the few chemicals that have been tested for toxicity and currently ignores chemical mixtures, possible synergistic effects, and the effects of exposure to multiple emissions sources.

The validity of this claim is further discussed in the section of this report titled "The Information Value of Risk Analysis,".

CRS workshop panelists agreed that there is cause for concern about the quality of quantitative estimates of risk, costs, and benefits. They said this is because the quality of underlying data is questionable, analyses are difficult to do well and vulnerable to human error, and results are presented without accompanying information about the range of possible estimates that might be produced by different scientists. For example, one panelist claimed that although it is generally thought that risk analyses are deliberately conservative and therefore protective of human health, this is not always true; some risk estimates probably underestimate risk.

Many who promote risk analysis acknowledge that it has limitations but believe they can be overcome through research and development of improved analytic methods or through the establishment of guidelines for the conduct of analysis and presentation of results. Even the imperfect information produced by risk analysis today is valuable, some argue, and should be considered by decision makers. Clear explanations of the assumptions and uncertainties associated with the risk estimates and standardized methods would reduce the chance of misuse or misunderstanding, proponents of the legislation believe. The consensus among CRS workshop panelists was that risk analysis is quite sophisticated and constantly improving.

Should It Be Used to Quantity Environmental and Health Benefits?

Many policymakers would like risk analysis to be used to quantify risk reduction potential of environmental management strategies. Some also want to quantify benefits to permit comparison with the costs of management options. Such information, they argue, would help identify economically efficient choices, that is, how to get the "biggest bang for the buck." Some of these policymakers believe that spending for environmental protection should be managed more efficiently because it is a considerable amount of money, too much to spend wastefully. Others believe spending for environmental protection is excessive, squandering too many public and private resources to produce small or uncertain gains in environmental protection and public health. Legislation addressing these concerns would require EPA to conduct risk assessment and economic assessment (either net benefit or cost-effectiveness assessments) of regulations.33

EPA's most severe critics are of the opinion that environmental regulations adversely impact the national economy and international competitiveness of American businesses. (For an analysis of the economic consequences of environmental regulations, see CRS Report 97-459, Environmental Protection: How Much It Costs and who Pays.)

Many environmentalists and others object to quantitative cost-benefit assessment of environmental or health and safety laws and regulations on moral or ethical grounds, claiming that benefits such as life, health, and an aesthetically pleasing environment should not be equated with commodities bought and sold in the market and valued in monetary terms. They want benefits described fully, in qualitative as well as quantitative terms.

Others criticize proposals to rely more heavily on quantitative risk assessment for political reasons. They charge that the complexity of the analytic process allows analysts and interest groups to conceal important value judgments and questionable assumptions. Thus, critics of risk analysis legislation argue, the results of risk analysis may be misleading to policymakers, who are the ones charged with the responsibility for making the decisions. Moreover, complexity allows an intellectual elite and those wealthy enough to hire their expertise to dominate discussions and decisions, according to these opponents of legislation.

Scientific objections also have been raised to quantitative assessment of the benefits of risk reduction by those who believe that the economic theories and methods employed to express the value of benefits in dollars are inadequate and unreliable. For example, they contend that surveys asking how much one is willing to pay to obtain a hypothetical reduction in risk are irrelevant to important decisions about real choices in daily life. They argue that for real decisions many factors are considered in addition to the magnitude of risk reduction. For example, people may consider the quality of the benefits and costs, characteristics of those who bear the costs and receive the benefits, the degree of choice available to those exposed, timing of the decision and health or environmental consequences, economic and social status of individuals who benefit or potentially suffer prior to and following the decision, certainty of the risk and cost estimates, and the perceived necessity of choosing among the proffered options. Some believe the only valid approach to quantifying society's values is case-by-case.

EPA's Science Advisory Board agrees with these critics, in part. It has criticized EPA's methods for assuming that the future value of an ecological resource must be less than its present value. It concluded in a 1990 report that this policy inevitably leads to depletion of irreplaceable natural resources.34 Moreover, reliance on measures such as the public's "willingness to pay" exacerbates this problem, according to the Board. For example, although many members of the public may not care about wetlands, these nonetheless contribute to the larger ecosystem and are valuable now and in the future, for such purposes as waterfowl and fish habitat and to filter pollutants from water. Therefore, the Board concluded, techniques need to be developed to assess the real long-term value of ecosystems.

Critics of quantitative economic analysis also claim that the costs of environmental and health and safety regulations may be exaggerated, because data often are provided by regulated industries, and EPA ignores the "learning curve"; the critics believe real costs tend to go down with time as companies gain experience in complying. In contrast to monetary values, unquantifiable benefits are excluded from analysis, according to critics who are particularly concerned about the discounting of benefits to future generations. CRS workshop panelists noted that EPA has been unable to quantity benefits such as clean ground water and that the Agency claims it does not know how to evaluate costs and benefits of some of its programs.

The insensitivity of quantitative analysis to the value of environmental and health benefits to future generations is an example of the general inability of economic methods to account for uneven distributions of environmental risks and regulatory costs and benefits among people, according to this view. Advocates for environmental justice who seek to eliminate alleged disproportionate risks borne by low-income and minority communities argue that those subgroups may be burdened and other groups may reap a disproportionate share of the risk reduction, while taxpayers and consumers bear the cost of implementation and compliance. Instead of, or in addition to, weighing total costs against total benefits, they want EPA to describe distributions of risks and benefits and to consider inequities in developing environmental regulations. Existing statutes may limit EPA's authority to address this concern in regulations, however. For example, statutes sometimes direct states to design remedies for risks.

Many promoters and critics of environmental risk analysis agree that the results of quantitative risk assessments should not be reported as a single number known as a "point estimate." Point estimates of risk often are used, they believe, (as did CRS workshop panelists) to focus attention on relatively small risks to large populations (for example, the U.S. population as a whole) rather than on large risks to smaller groups, such as workers, the economically disadvantaged, or ethnic minorities. EPA often assesses and reports risk estimates for vulnerable individuals, so this concern may be overstated. Nevertheless, CRS workshop panelists agreed that the results of risk analyses too often are reported as single numbers (point estimates). They would prefer analysts present a range of risk estimates that might be obtained by different scientists with different values and deliberately discuss scientific uncertainty so that decisionmakers are not misled. The quality of the data underlying the risk assessments also should be revealed, they agreed. According to some panelists, cost data and estimates often are as uncertain as risk estimates and should be presented as a range of plausible values.

Still others charge that if the goal is to reduce costs of regulation, then Congress should revise the statutes, not just add requirements for quantitative analysis of risks and economic impacts, because EPA cannot consider costs of regulation under some major environmental statutes that dictate the degree of protection to be achieved.35 In this view, a mandate to analyze in itself is inefficient, because analysis will consume scarce EPA resources, sometimes without purpose. In many cases, statutes that impose the greatest regulatory compliance costs (and therefore would be the preferred targets for analytic requirements meant to increase efficiency) prohibit consideration of compliance costs or health, human welfare, and environmental benefits, according to OMB analyst Arthur Fraas.36

Finally, many argue that formal quantitative cost-benefit analysis would delay EPA's issuance of some regulations, and delays will mean that lives or habitats might be lost that could have been saved had the regulation been in effect. Delays may even increase the cost of analysis, for example, if the Agency misses statutory or judicial deadlines and environmental groups respond by filing lawsuits, critics argue. Thus, they claim the net benefit of environmental regulation might be reduced. On the other hand, if analysis is cut short to meet deadlines, industries may charge that EPA regulations are arbitrary and capricious. Depending on the statutory requirements, this may also lead to legal challenges. Some critics believe that regulation based on risk is too time consuming, resulting in delayed implementation of statutes enacted to protect human health and the environment. They cite the rationale for the Clean Air Act Amendments of 1990 (CAAA) as an example. Before enactment of the CAAA, Section 112 of the Clean Air Act required EPA to regulate emissions of hazardous air pollutants based on risk. Only 6 pollutants were regulated in 13 years under that Act. Dissatisfied with the pace of regulation, Congress amended the Act in 1990 to require EPA regulation of emissions of 189 hazardous pollutants based on available pollution control technology. 37

CRS panelists agreed that data collection and risk analysis can delay regulatory action and consume resources that might otherwise be used to prevent, reduce, or redress environmental pollution. They noted a tendency for agencies to analyze more than is necessary to inform risk managers, and advised that risk analysis should cease when the cost of conducting the analysis and of delayed decisions outweighs the potential value of additional information to decision makers.

Critics of cost-benefit assessment are not necessarily opposed to designing efficient environmental protection strategies; many argue that cost-effectiveness is a more acceptable form of analysis of alternative management strategies. Cost-effectiveness analysis begins after the health and safety or environmental goal (that is, the level of risk reduction desired) has been established; the analysis then identifies the least cost means of obtaining the benefits.

Those who practice cost-benefit analysis have responded to some of these criticisms by developing new methods. Policymakers also have responded by modifying proposals, for example by limiting requirements to rules with an annual impact of $100 million or more. In recent years, approximately 3.5% of EPA's published rules were expected to have an economic impact greater than $100 million.

Presidents Reagan and Bush required and President Clinton requires (to the extent permitted by law) analysis of risks and costs for federal regulations likely to have a "major" or "significant" effect, respectively, on the nation. The Clinton Administration's requirements govern the current level of risk analysis at EPA. Historically, this level of analysis has been found to yield a return on investment of 1,000 to 1, due to cases where analyses have led to choices that increased net benefits to society from environmental regulation.38 Between February 1981 and February 1986, EPA invested between $210,000 and $2,380,000 to analyze a major rule. The average cost of analysis was $67 5,000.39 There are no figures available for more recent years or for the preparation of less comprehensive analyses for rules that were not "major" rules.

For more information about the strengths and limitations of cost-benefit analysis, see CRS Report 95-760, Cost-Benefit Analysis: Regulatory Issues.

ENDNOTES

31 The information summarized in risk analysis generally is obtained from animal experiments, studies of the effects on humans who have been exposed to hazards, short-term tests on bacteria or living tissues of people or animals, and knowledge and theories about the structure and behavior of chemicals.

32 For example, for three case studies illustrating the difficulty of separating science from policy in risk assessment, see Brown, Halina Szejnwald and Robert L. Goble, "The Role of Scientists in Risk Assessment," RJSK Issues in Health & Safety, v.283, (Fall) 1990, p.283-311.

33 Various forms of economic analysis are discussed on previous pages and in endnotes 16 and 25.

34 U.S. EPA, Science Advisory Board. Reducing Risk: Setting Priorities and Strategies for Environmental Protection. Washington, U.S. Environmental Protection Agency, September 1990. p.8.

35 See CRS Report, 98-619, Risk Analysis: Background on Environmenta/ Protection Agency Mandates.

36 Luken and Fraas. The US. Regulatory Analysis Framework: A Review. p.100.

37 U.S. Library of Congress. Congressional Research Service. A Legislative History of the Clean Air Act Amendments of 1990. Volume 1. Washington, U.S. Govt. Print. Off., Nov. 1993. p. 860-863.

38 U.S. EPA; Economic Studies Branch, Office of Policy Analysis. EPA's Use of Benefit-Cost Analysis 1981-1986. Washington, U.S. EPA, August 1987. p.1 and 2.

39 Ibid. p. 6-5.