Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Keeping Pace with Science and Engineering. 1993. Pp. 243-250. Washington, DC: National Academy Press. Science' Engineering' and Regulation Richard D. Morgenstern The principal problem posed in this volume is that the policy and regu- latory process does not incorporate new scientific information very well or in a timely manner. No matter how good or poor the state of understanding was about the costs, risks, or benefits associated with a particular environ- mental issue, sooner or later that understanding is likely to change, presum- ably for the better. When it changes, the new understanding may call an earlier decision into question. In light of this inevitability, we should be concerned that the regulatory system is sufficiently flexible to be able to adapt. In thinking about this issue, we should recognize that all regulations are not created equal. Depending on the situation, different levels of scientific evidence are required to promulgate a regulation. Driven by statutory man- dates and deadlines, tight budgets, and other directives, a type of "triage" system is applied to different environmental problems. The case studies in this volume all of which deal with changes in the information base suggest that there are three distinct but related circum- stances in which the U.S. Environmental Protection Agency (EPA) may be called upon to revisit decisions: 1. The original decision was based on the best science at that time, but subsequently the science changed. *The author is indebted to Frederick Allen, Devra Lee Davis, Albert McGartland, and Myron F. Uman for invaluable comments. 243
244 RICHARD D. MORGENSTERN 2. The agency was pressed to make decisions in such a short time frame or under such a tight budget constraint that it was not possible to evaluate the science thoroughly. Economic and other pressures have now raised concerns about the scientific validity of the original decision. 3. The original decision was based solely on the availability of pollu- tion control technology, and more recent evidence has brought to light some environmental or public health problems that were not previously foreseen. The rationale for change is the clearest when the original decision was based on the best science then available and now there are new data sug- gesting that the decision should be revisited. Particularly in cases where the science is making significant new advances, we clearly need to incorporate the new information into our regulations and policies. In this volume, the case study on dioxin comes closest to this descrip- tion. As described in the case study, EPA is in the process of reconsidering its risk assessments of dioxin, with an eye to what EPA should do next. The agency has gone through a public process but in the interim has committed to enforcing the regulations until any changes are made. Chlorofluorocarbons (CFCs) are another good example. The best scien- tific data available were brought to bear when the Montreal Protocol was drafted in 1987. The drafters knew, however, that the science was rapidly evolving and wisely wrote in a provision to the protocol that allows for revisions based on new scientific information when it becomes available. In 1990, based on such evidence, the 50 percent reduction in CFCs that had been agreed to just three years earlier was transformed into a phaseout by the year 2000, and several additional ozone-depleting chemicals were added to the agreement. Two years later, in 1992, the phaseout date was moved up to 1994 and several additional chemicals were added to the agreement. The second category involves situations where, often because of the pressure for quick action, the original decision was not always based on the most thorough understanding of the science. An example here is the body of drinking water regulations. Congress charged EPA with promulgating a large number of regulations in a short time but did not correspondingly expand the resources to accomplish the task. Now, state and local govern- ments face increased financial burdens to comply with these regulations. The key question here is how to remedy the situation? In 1992 the Na- tional Governors' Association convened several high-level meetings and ul timately issued a series of recommendations that called for slowing the implementation of certain drinking water regulations. After considering a moratorium on implementing already promulgated regulations, Congress enacted the Chafee-Lautenberg Amendment to the EPA Appropriations Act, which mandated a major study on the implementation of the Safe Drinking Water Act, including analyses of the costs and benefits, state implementa
SCIENCE, ENGINEERING, AND REGULATION 245 lion, and small systems issues. Whether this will ultimately lead to signifi- cant programmatic or statutory changes remains to be seen. Some Superfund remedies fall into this category too. In a number of instances states have raised concerns about the wisdom of the proposed site remedies, 10 percent of the costs of which are typically assumed by the states themselves. This issue will inevitably be considered in the reauthori- zation of the Comprehensive Environmental Response Compensation and Liability Act. The third category, in which the original decision was based solely on the availability of pollution control technologies, is illustrated by the case study on the municipal waste combustor New Source Performance Stan- dard. In that instance, EPA had the flexibility to regulate under either Section 111 or 112 of the Clean Air Act. Using the latter would have involved a risk-based determination. Based on its discretion, the agency chose, instead, to issue the regulations under the authority of Section 111, which applies a technology-based approach to controlling emissions. In a similar vein, the standards for hazardous waste treatment generally rely heavily on engineering solutions, often ignoring a balancing of risks, ben- efits, and costs. THE LAWS AND THE REGULATORY PROCESS Some provisions of the laws EPA administers require the agency to revisit decisions periodically. The most notable example is the National Ambient Air Quality Standards the engine of the Clean Air Act which, by statute (Section 109), are to be reconsidered every five years. In fact, these reconsiderations have consistently taken longer than five years. Four- teen years elapsed between the last two formal decisions on the ozone standard and, in the end, the standard was simply reaffirmed. The primary SO2 standard was revised in 1979 and it is still being reconsidered. Another example is the reregistration of pesticides. EPA is required under the 1988 amendments to the Federal Insecticide, Fungicide, and Ro- denticide Act to reregister the approximately 700 active ingredients within 10 years. Many of these active ingredients were registered when risk as- sessment methodology was extremely primitive, and the data requirements for registration were minimal. Because of this, there are many registered active ingredients that have not undergone rigorous human or ecological risk assessments. To date, the EPA has reregistered only a handful of active ingredients; however, the pace should pick up. When originally enacted, many technology-based standards, such as the New Source Performance Standards (Section 111) of the Clean Air Act, were intended to be technology forcing. The paint industry, for example, has reduced its rate of emissions of volatile organic compounds (VOCs) by ~7 .
246 RICHARD D. MORGENSTERN more than 20 percent by substituting water-based solvents for petroleum- based solvents, largely as a result of the technology-based emission limits for that industry. Technology-based standards, however, can also inhibit innovation. Stan- dards are often based on a particular technology, which at the time may be considered to be at the cutting edge. Permit writers, knowing that the standard setters relied on that technology to determine the limit, rarely allow compliance by the use of other technologies except after rigorous (often costly) demonstrations by the applicant. Thus the technology may become entrenched in the industry and there is little incentive to innovate or to achieve emissions beyond those required by the standard. Most of the nation's environmental laws allow EPA to reconsider its decisions at its discretion-in other words, it is up to EPA. Critics note that the agency does not have a strong record for meeting deadlines for the reevaluation of regulatory decisions. Often this is due to the budget con- straints under which EPA operates and pressures on the agency to address as yet unregulated environmental problems. Another factor to consider is that the EPA's regulatory process is typi- cally analytical, inclusive, and very public especially when compared with the procedures used in other countries. In contrast to most parliamentary systems, the U.S. government is organized around a system of checks and balances on the theory that adversarial tensions are required to ensure fair and balanced outcomes. The Administrative Procedures Act mandates a complex set of steps, including analysis, development of regulatory options, formal proposal, public comment, reassessment, and final rule. The key participants in the process include EPA, the regulated community, environ- mental organizations, the media, and the research and engineering commu . . . Sties. The case studies in this volume show how our open, democratic process must accommodate not only scientific concerns, but also political, legal, and institutional issues. Many legal scholars would undoubtedly argue that the process itself is as important as the regulatory outcome. THE ROLE OF SCIENTISTS AND ENGINEERS Science, engineering, and their practitioners should be, and in many cases are, an integral part of the development, drafting, and implementation of regulations. In practice, scientists and engineers assist in ensuring that regulations are based on the best technical understanding available. That is, they help to translate technical concepts into implementable, enforceable, realistic regulations. Clearly, many would argue for strengthening the role of scientists and engineers in the regulatory process. More technically defensible decisions inevitably lead to better alloca . .
SCIENCE, ENGINEERING, AND REGULATION 247 lion of the resources available to be devoted to environmental protection. At the same time, reconsidering decisions involves costs that can be reduced if we get it right the first time. Changing regulations creates uncertainty and a less stable business environment. Since EPA has severely limited resources, revisiting past decisions often means other work is delayed. As Robert M. White notes in his introduction to this volume (p. 6~: . . . the environmental regulatory system should . . . keep pace with the changes in our understanding of the technical aspects of the issues, and it should remain stable on a time scale sufficient for regulated parties to comply with some measure of economic efficiency. It is evident that these two normative characteristics can be, and frequently are, in conflict. In this regard, the most important thing that scientists and engineers involved in the regulatory process can do is to understand policymakers' information needs. Scientists and engineers should consider the following: 1. Quantify and measure outcomes that have meaning in a policy con- text. What does a reduction in forced expiratory volume mean to decision makers? How does that translate into lifestyle or health changes that people understand? Similarly, hazards research that is too general and fails to de- velop dose-response relationships is often difficult to use. 2. Don't be afraid to make best judgments. Decisions often have to be made under tight deadlines. It is usually better that judgments about sci- ence and engineering be made by scientists and engineers rather than by laymen. EPA often uses the National Research Council or its own Science Advisory Board to get best judgments about tough issues. The Clean Air Scientific Advisory Committee (CASAC) has played a key role in helping the agency to set ambient air standards. This type of mechanism should be expanded to encourage scientists and engineers to make best judgments on unresolved technical issues. 3. Work on the right technical issues. This is perhaps the hardest prob- lem of all to solve. Good research and development usually involves a long lead time and the EPA process often cannot provide sufficient notice to get it going. However, just as EPA can, to some extent, shape its agenda to reflect the work of researchers in the field, so could the technical experts take EPA's needs into account as they set their research priorities. EPA's regulatory needs should not be the only determinant of the research and development agenda, but its needs should play a significant part in shaping that agenda. To help establish research agendas for both intramural and extramural research, EPA has begun an interactive consultation process in . valving program managers and scientists inside the agency as well as ex- perts from outside.
248 RICHARD D. MORGENSTERN WHAT OTHER INSTITUTIONS CAN DO I have covered some things that EPA and the science community might do to make use of new scientific information in the regulatory process. Let me turn to the issue of how other key institutions can become helpful cata- lysts for change as well. 1. The mainstream media. Both print and broadcast media can provide more complete coverage of technical issues relevant to environmental regu- lation. A number of newspapers and journals are beginning to acknowledge that environmental policy cannot be based just on what Senator Moynihan has called "middle-class enthusiasms." The media should focus more on facts and not just perceptions. Scientists and engineers can help reporters by taking the time to speak clearly and frankly to the media. In many technical quarters media coverage is shunned. Yet scientists lose their right to complain about scientific illiteracy if they do not themselves contribute to a more informed public. 2. Congress. The Washington Post reported that the House Space, Sci- ence, and Technology Committee, formerly obscure, was the second most popular pick for freshman members in 1993, largely because of its potential for pork barrel politics (Washington Post, January 26,19931. In fact, pork barrel spending does the double harm of funding often unneeded programs and forcing aside programs founded on good technical assessments of sci- entific and policy priorities. In fiscal year 1993 the Congress added more than 100 specific items to the EPA budget while approving essentially flat spending levels relative to 1992. Of course, in such circumstances, some- thing has to give. Many programs of national interest, based on scientific risk assessment, were cut to make room for these congressionally mandated programs, many of which had not been subject to technical review and did not carry high technical priority. 3. Environmental groups. As environmental groups become better versed in technical issues, their contributions to the debate in science-based policy questions are becoming more useful. As many as 15,000 environmental groups currently exist in this country, most at the community level. It would be unrealistic to expect many of these groups to be dispassionately concerned with technical analysis. Some of the national groups, such as the Nature Conservancy, are becoming more conversant in scientific issues and are using science as part of their own priority setting. The Environmental Defense Fund is gaining expertise in both economic and science issues. But a staff study done at the EPA in 1992 indicated that science most often comes into the loop in these organizations after the membership, board, or staff decides what problems to target. Clearly, further progress is called for. A?
SCIENCE, ENGINEERING, AND REGULATION 249 Building a commonly accepted science base will enhance trust and help ensure that we are not working at cross-purposes. 4. The regulated community. Finally, the regulated community can fund more high-quality research and development. The most effective arguments are based on good technical and economic analyses. Too often industry channels its resources to lobbying efforts at the expense of collecting real data or even contributing relevant data already in its possession. EPA has established procedures for accepting proprietary data and has now a solid track record with such data. One can see the dawning of a more progressive and involved view in many parts of the business community, but clearly the time has come for industry to make a larger contribution to developing relevant scientific and engineering information and making it readily avail- able to EPA. DISCUSSION The way a problem is posed often dictates its solution. Thus, the problem considered in this volume of case studies is how to improve the use of scientific and technical information in the regulatory process. The solu- tions presented in the case studies generally offer a variety of approaches regarding more and better scientific information. As described above, en- hanced mechanisms to improve the use of technical information in the regu- latory process are clearly needed. It is possible that this focus on better data and analysis has some down- sides as well. Two types of risks need to be examined the risk of falsely regulating something that needs no such control, a false positive, and the risk of falsely exonerating something that poses a hazard. While many observers argue that the costs of false positives are quite high, at least one analyst, Carl Cranor (1993) of the University of Califor- nia, suggests that the costs of false negatives particularly in the environ- mental field are significantly greater than the costs of false positives. Re- search scientists are characterized as aspiring to find more and better data, to withhold judgment until sufficient empirical information is obtained, to be cautious about inferences about causation, to be ever skeptical regarding conclusions, and to guard against basing decisions on false positives, that is, evidence suggesting something is a problem, when it, in fact, is not. This type of "better science," as Professor Cranor notes, "when . . . used inappropriately or insensitively . . . can produce substantial bad conse- quences as well as good." To reduce these undesirable consequences, he argues for expanded administrative discretion regarding scientific and tech- nical information: A better approach would be to address uncertainties by policy choices
250 RICHARD D. MORGENSTERN determine the minimum amount of information needed to guide decisions and then use it, develop rebuttable interim standards for protecting public health, find an appropriate balance between false positives and false nega- tives for the context, expedite the evaluation of known carcinogens [and other hazards], and increase the rate at which substances are identified as toxins (Cranor, 1993, p. 103~. While one can dispute the frequency of false negatives it may be most relevant where the data on risk and other aspects are particularly lacking- Congress clearly thought that the problem of air toxics emissions repre- sented a false negative; Congress dropped the risk approach in favor of a technology standard as an initial basis for regulating a specified list of pollutants. Even acknowledging such examples of false negatives, how- ever, it is probably not the principal problem in the field of environmental protection. Yet, it is based on a cogent assessment of the complex and changing nature of environmental sciences. Certainly we should be mindful of the issues raised by such an approach as we consider our future paths. REFERENCE Cranor, Carl F. 1993. Regulating Toxic Substances: A Philosophy of Science and the Law. New York: Oxford University Press.
