|
20 ELR 10419 | Environmental Law Reporter | copyright © 1990 | All rights reserved How Efficient Are EPA's Regulations?Ralph A. Luken and Lyman H. ClarkDr. Luken is Senior Environmental Advisor to the United Nations Industrial Development Organization (UNIDO) in Vienna, Austria. He is on leave from the U.S. EPA, where he was Chief of the Economic Analysis and Research Branch of the Office of Policy, Planning and Evaluation (OPPE). Mr. Clark is President of Environmental Economics Associates of Traverse City, Michigan. He was formerly chief of EPA's Cost and Economic Impact Analysis Branch in OPPE. The views expressed are those of the authors and do not represent the views of the U.S. Environmental Protection Agency. In the 1970s, the newly created U.S. Environmental Protection Agency (EPA) embarked on three different regulatory approaches for reducing pollution. First, the Clean Air Act1 mandated that EPA set ambient-based standards that protect human health and welfare. These standards require existing industrial sources of air pollution to install pollution control equipment only to the extent necessary to meet ambient standards. Second, the Clean Air Act mandated that EPA in some circumstances set benefits-based2 standards that trade off risks to society with the costs of risk reduction. These standards require existing industrial sources to install pollution control equipment only to the extent that there would be a reasonable balance between the benefits of pollution reduction and the costs of pollution control technology. Third, the Clean Water Act3 mandated that EPA set technology-based standards that reflect the availability and affordability of pollution control technology. These standards require existing industrial and municipal sources to meet uniform discharge limitations, even if the pollutants discharged did not result in violations of ambient standards. In the debate over how best to regulate the environment, each of the three approaches is championed by its own school of thought. The utilitarian school, which reflects the concern of the public health profession, supports the use of ambient-based standards as a scientific and reasonable way to protect health and welfare.4 The absolutist school, which reflects the interests of the environmental rights-oriented movement, supports the use of technology-based standards as a necessary and practical approach.5 The rationalist school, which reflects the economist's view of environmental management, supports the use of benefits-based standards.6 The utilitarian school dominated environmental management until the passage of the Clean Water Act in 1972, when the absolutist school achieved dominance.7 The Clean Air Act Amendments of 1977 and Clean Water Act Amendments of 1977 gave additional impetus to the absolutists' technology approach for existing industrial sources of pollution.8 More recently, the 1984 Resource Conversation and Recovery Act amendments, the 1986 Safe Drinking Water Act amendments, and the proposed amendments to the Clean Air Act for hazardous pollutant control represent new triumphs of the technology approach over the ambient and benefits approaches. The rationalist school has prevailed only with chemical regulating statutes because Congress recognized the importance as well as the hazards of chemicals to society. The few rationalist concessions in other pollution reduction statutes, such as odor control and lead in gasoline, appear to be an oversight rather than a deliberate decision. Although the debate over regulatory approaches has proceeded mostly on theoretical grounds, an empirical comparison of the relative efficiency of each approach is also possible. EPA's regulatory programs have been in place long enough to estimate both the costs of the regulations and the value of the environmental and social benefits that they have produced. Society ultimately pays for pollution reductions through increased costs for goods and services provided by polluting firms. To the degree that EPA's regulations are efficient, society benefits from the reduction [20 ELR 10420] or elimination of damages. EPA's regulations are efficient to the degree that they provide net benefits; that is, as long as the societal benefits of better health or a cleaner environment have greater value than the costs society pays to achieve them.9 To date, Freeman10 has attempted the only ex post national assessment of the efficiency consequences of EPA's water and air pollution control programs. Portney and Freeman11 recently reiterated the same findings with an update of Freeman's original effort. In both publications, the findings suggest that the benefits have exceeded the costs for ambient-based standards associated with the Clean Air Act and that the costs have exceeded the benefits for the technology-based standards associated with the Clean Water Act. As both Portney and Freeman acknowledge, their estimates are not systematic, spatially specific assessments of the efficiency consequences of environmental regulations. Their estimates, although insightful, suffer from the same limitations that would exist, for instance, if one did a water supply and demand analysis for the United States as a whole. Such an analysis would show that there is plenty of water in the aggregate, but would completely mask regional problems. To define problems, to determine achievements, to relate achievements to costs incurred, and hence to determine the efficiency of environmental regulations, require systematic analyses at many locales and in terms of specific river reaches and airsheds. This Dialogue summarizes the first effort12 to assess the efficiency of EPA's programs by comparing the Agency's success in generating net benefits through the three different regulatory approaches; that is, the ambient approach (most of the air program), the technology approach (the water program), and the benefits approach (part of the air program). The effort compares the measurable environmental benefits and costs that have resulted from applying ambient, benefits, and technology approaches to the pulp and paper industry between 1973 and 1984. It also examines the potential of the three different regulatory approaches for achieving net benefits from air, water, and hazardous waste pollution control between 1984 and 1994. The Clean Air Act requires EPA to set primary and secondary national ambient air quality standards. Primary standards are intended to protect the health of the population, whereas secondary standards are meant to protect public welfare (materials, crops, forests, etc.). The 1970 Clean Air Act required EPA to propose national ambient air quality standards 30 days after its passage. Administrator Ruckelshaus was committed to meeting this deadline and turned to officials from the old National Air Pollution Control Commission, which was incorporated into EPA, for a full set of ambient standards.13 Three days before the statutory deadline, the officials gave him the standards, using scientific evidence and judgments that they had formed about the effects of air pollutants.14 Having neither the time nor the independent staff to review these standards, Administrator Ruckelshaus accepted the recommendations of these officials. He proposed national ambient air quality standards for six criteria pollutants, including total suspended particulates (TSP), on January 30, 1971, and promulgated final standards on April 30, 1971.15 The Clean Air Act requires EPA to set new source performance standards, which apply to new sources, for new or modified stationary source categories whose emissions cause or significantly contribute to air pollution that may endanger public health or welfare. New source performance standards are to be based on the best demonstrated technology, considering costs and economic and social impacts. If a new source performance standard is set for a pollutant not regulated by an ambient standard, the Clean Air Act requires that states develop regulations to control existing sources of the same pollutant. EPA regulations specify, however, that EPA will issue guidance to the states on the best retrofit technology for existing sources, considering costs and economic and social impacts. If the pollutant is health-related (e.g., sulfuric acid mist from sulfuric acid plants), states must provide strong justification for adopting emission limits less stringent than the guidance specified by EPA. For pollutants that are welfare-related (e.g., total reduced sulfur (TRS) from pulp and paper mills), states have more flexibility in adopting regulations for existing sources. Under these circumstances, the standard becomes benefits-based rather than technology-based. EPA promulgated a new source performance standard for TRS from new sources in 1978.16 The standard reflected EPA's review of the best demonstrated technologies, tempered by EPA's consideration of technology costs and economic impacts. As required by the Clean Air Act, EPA issued guidance for reducing TRS emissions from existing sources in 1979.17 The TRS emission limits in the emission [20 ELR 10421] guidelines were less stringent than the mandatory technology standards for new sources. The TRS emission guidelines for existing sources are examples of benefits-based standards for two reasons. First, and most important, EPA allowed states considerable flexibility in requiring controls on TRS emissions through its decision to designate TRS as a welfare-based, rather than as a health-based, pollutant.18 For welfare-related pollutants, states could balance required emission reductions and time required for compliance against other factors of public concern. For example, the evidence on odor problems around pulp mills in the mid-1980s suggests that states considered the severity of the odor nuisance and the economic conditions of mills in requiring reductions in TRS emissions. Better data and models would have undoubtedly contributed to an even better balance between the benefits and costs of odor reductions. Second, EPA's Office of Air Quality Planning and Standards paid particular attention to cost effectiveness (dollars per ton of pollutant removal) in issuing TRS emission guidelines. It exempted individual sources and precluded control techniques because the cost of control exceeded a certain level. The Clean Water Act requires EPA to set technology-based standards that restrict the amount of pollutants discharged by industrial plants. The standards are different for each industry and for subcategories within an industry. Three technology-based standards apply to existing industrial sources: best practicable control technology, best conventional technology, and best available technology economically achievable. New source performance standards are similar to best available technology. EPA promulgated best practicable technology standards to control the familiar, primarily conventional, pollutants for most of the pulp and paper industry in 1974 and 197719 and promulgated the remaining best practicable technology and new source performance standards for conventional pollutants in 1982.20 The best practicable technology standards for biological oxygen demand (BOD) and total suspended solids (TSS) are based on biological treatment for all but a few nonintegrated subcategories of the pulp and paper industry where only primary treatment is the basis for the effluent limitations. These limitations for BOD and TSS are uniquely specified for the 25 production/process subcategories within the industry. The new source performance standards for conventional pollutants were established based on the same technology as best practicable technology, plus additional flow-reducing, in-plant process controls. The first part of this ex post evaluation characterized the costs and pollutant discharges of the pulp and paper industry in 1984 with and without the Clean Water and Clean Air Acts. The "with" characterization represents the actual costs and pollutant discharges in 1984. The "without" characterization represents hypothetical costs and pollutant discharges calculated by combining 1984 production levels with 1973 pollution reduction technology in-place. The without costs were then subtracted from the with costs to eliminate the costs incurred before 1973, plus any increase in costs due to expanded production. The without pollutant discharges were then subtracted from the with pollutant discharges to capture the increase in pollutants due to increased production. As seen in Table 1, implementation of the standards made a significant difference in pollutant discharges. Table 1: Summary of National Pollution Control Costs and Pollutant Loadings
The second part of the evaluation assessed the benefits of the reductions in pollutant loadings that resulted from the implementation of the three approaches. Due to the difficulty of assessing ambient quality changes resulting from reduced pollutant loading, benefit estimates were calculated for only a representative subset of the mills included in the cost characterization. The benefits assessment of the technology-based standard was based on the increased value of recreational activities resulting from improvements in water quality and covered 68 of the 306 directly discharging pulp and paper mills covered in the cost analysis. The benefits assessment of the ambient-based standard was based on both health and welfare benefits21 resulting from reduced exposure to particulate emissions and included 60 of the 157 pulp-producing mills covered in the cost analysis. The benefits assessment of the benefits-based standard was based on people's willingness to pay to avoid sulphur odors and included 60 of the 129 draft-pulp-producing mills covered in the cost analysis. As seen in Table 2, the technology-based standards for water pollution management failed as a strategy for generating net benefits. The costs clearly exceeded the benefits in the aggregate, [20 ELR 10422] as well as in the specific, in most situations. Benefits exceeded costs at only seven of the 68 mills investigated in the study. The ambient-based standards for ambient air pollution management succeeded as a strategy for generating net benefits. The benefits exceeded the costs in the aggregate as well as in the specific for about one-third of the mills (20 of 60 mills). We did not expect emission reductions at all mills to exhibit benefits greater than costs because the ambient standard was uniformly implemented across the country. The benefits-based standards for air pollution management also succeeded as a strategy for generating net benefits. The benefits exceeded the costs in the aggregate as well as in the specific for about one-half of the mills. Benefits exceeded costs at 29 of the 60 mills investigated. The benefits-based standard failed in many specific situations, partly because the EPA guidance in 1979 did not require rigorous quantitative analysis of both benefits and costs. Table 2: Benefits and Costs Attributable to the Three Approaches — 1973-198422 ($ 1984 10<6>)
The relative effectiveness of the ambient approach calls into question the harshness of the rationalist school's critique of the utilitarian school. The rationalist school rejects not only the technology approach, but also the ambient approach because the ambient approach fails to incorporate population exposure and the costs of meeting a given standard. This preconceived failure did not happen in the implementation of the ambient-based TSP standard around pulp and paper mills, however. In fact, the ambient-based standard produced results that were remarkably similar to the benefits-based standard for TRS in the aggregate and in the specific. The ambient-based standard succeeded as well as it did, we think, because local environmental managers did not require reductions in TSP emissions from pulp-producing mills where the costs per ton of emission reduction were high and the potential human health risks were small. The Clean Air Act focused the attention of these environmental managers on achieving areawide ambient air quality, and as a result allowed them to take into account the cost-effectiveness of emission reductions among different sources (primarily manufacturing and utilities). Our investigation found considerable variation in the cost per ton of TSP reduction at pulp-producing mills in different areas. Other investigations have found considerable variation among different source categories.23 To our knowledge, no one has attempted to relate these variations in specific areas to the relative cost-effectiveness of emission reduction and population exposure. Our analyses also investigated the relative efficiency of the three regulatory approaches in view of the projected growth in industrial capacity. We estimated that the annual capacity for pulp production and for paper and paperboard would increase by 8,100 tons (15 percent) and by 9,800 tons (15 percent), respectively, in the 1984-1994 time period. We assumed that 60 percent of the expansion would occur at existing mills; the remaining 40 percent would occur at new mills. In spite of the requirement to install more stringent pollution control equipment (new source performance standards) at existing mills with significant capacity expansions and at new mills, we estimate that water effluents and air emissions will increase based on current Clean Water Act and Clean Air Act policies. Our findings about the three regulatory approaches for reducing the environmental risks associated with industrial growth generally make a strong argument for a benefits approach. (See Table 3.) The benefits approach is the only approach that would result in significant net benefits for both conventional water and criteria air pollutants. Neither the technology-forcing nor antidegradation approach would generate net benefits. Additionally, the benefits approach is the most efficient approach for reducing the environmental risks associated with increased levels of TRS emissions. Table 3: Benefits and Costs Attributable to the Three Regulatory Approaches Assuming Industrial Growth24 ($ 1984 10 <6>)
These findings about the limitations of the ambient approach deserve special attention because they contradict past findings about the efficiency of the ambient approach. We speculate that a future-oriented ambient approach is not efficient because our evaluation of the future-oriented approaches does not consider how local environmental managers, particularly those dealing with air pollution problems, will make allowances for variations in cost-effectiveness among sources and population exposures. This evaluation points to the limits of critiquing regulatory approaches only from an ex-ante perspective, as is most often done by the rationalist school, and strongly supports the necessity for more rigorous post evaluations of regulatory programs. In addition, we recognize that ambient-based standards for all pollutants may not produce net benefits. According to the U.S. Office of Technology Assessment,25 partial implementation of the ozone ambient standard by 2004 would not generate net benefits. Our findings caused us to ask why Congress has not passed pollution control statutes that call for more benefits-based standards. Congress has only mandated benefits-based standards for regulation of chemicals usedby society. The risk-benefit provisions of the Toxic Substance Control Act26 and the Federal Insecticide, Fungicide, and Rodenticide Act27 show that Congress is concerned with achieving a reasonable balance between the benefits and risks when the risks are believed to be inherent in the products themselves. Apparently, Congress does not want to acknowledge, except in some specific provisions of the three pollutant-reduction statutes, that the manufacture of desirable products, such as paper, includes desirable as well as undesirable attributes and that elimination of these undesirable attributes in the extreme can be very costly in relation to the environmental protection achieved. An extreme outcome of congressional directives is the hazardous air pollutant mandate of the Clean Air Act. EPA recently issued regulations to implement the Act's hazardous air pollutant provisions that would reduce risks at a cost of over $ 150 million per cancer incidence.28 Supporters of the rationalist school of environmental management should not despair, however. They should remember that Congress mandated water development projects for over 100 years before it institutionalized in 1936 the requirement that the benefits should exceed the costs of such projects.29 The bureaucracy then took until 1951 to issue the first guidance (Green Book) on this subject. The latest guidance was issued in 1983.30 Even this guidance requires that economic efficiency be only one of three criteria for judging the merits of a water development project. Also, supporters of the rationalist school should recognize that benefit-cost evaluations are incorporated to some degree in major EPA decisions because of a series of executive orders that date back to the Nixon administration.31 The current executive order32 calls for an explicit benefit-cost analysis of major regulations. Unfortunately, it has generated considerable controversy about the legality and wisdom of the Office of Management and Budget's effort to oversee and coordinate the regulatory process. How can EPA focus more of the nation's resources as well as its own administrative resources on implementing regulations that generate net benefits? Certainly, EPA's technology approach has not eliminated significant risks and has resulted in some needless investments by public and private enterprises. Also, EPA's ambient approach has worked only when pollutant discharge is great and local environmental managers temper, for better or worse, the actual implementation of ambient standards. Further, EPA uses a benefits approach infrequently, only when required by statute, because its organizational culture is dominated by meeting statutory requirements rather than by achieving measurable environmental results. In reality,EPA can do little, given its current statutory authority. Congress must give EPA new direction if EPA and its state counterparts are to protect health and welfare. Congress should require that EPA have a focused mission statement that encompasses its major statutory authorities. EPA's current mission statement — reduction of risks to human health and the environment — does not set any priority among the multitude of environmental risks that exist in society. Instead, it justifies working on reducing any risks, significant or insignificant. A slight modification, although one that would require statutory approval and have revolutionary implications, would fundamentally change how EPA operates. The mission of EPA should be modified to read: "reduction of unreasonable risks to human health and the environment." Reduction of unreasonable risks33 would focus EPA's actions on those regulations and programs where the benefits to society would be comparable to the costs imposed on individuals and public and private enterprises. Congress should also give EPA more flexibility in implementing its numerous statutes. It should require EPA to prepare and systematically revise a strategic plan that would indicate the Agency's view of the most significant unreasonable risks confronting U.S. society.34 Congress should then allow EPA to implement individual statutory [20 ELR 10424] provisions consistent with its strategic plan. In so doing, Congress would free EPA from arbitrary timetables and individual statutory provisions that are of secondary importance in protecting human health and welfare. If these two recommendations are too radical, Congress could make a modest reform in the EPA statutes that require technology-based standards. The technology approach fails to reduce environmental risks because it does not take into account the impacts of pollutant removal on ambient quality or the size of the population that would benefit from an improvement in ambient quality. It also does not take into account the costs of pollutant reduction. Congress could require that EPA incorporate all of these factors in its decisions about the degree of stringency of environmental standards. Although this requirement would not mandate that there be a reasonable balance between benefits and costs, it would temper EPA's tendency to pursue technology for technology's sake alone.35 Nowhere is this requirement morenecessary than for hazardous waste management.36 EPA's recently proposed standard for municipal solid waste management would impose requirements so stringent that, because of the few people exposed, the cost per cancer incidence avoided would be $ 30 billion.37 Similarly, many of EPA's Superfund cleanups cost millions of dollars, yet are estimated to result in only minimal improvement in the quality of groundwater sources that no one will use for the foreseeable future. Hazardous waste management has only recently become a focus for environmental regulation. Hence, it provides an excellent opportunity to apply the lessons we have learned about the necessity of balancing benefits and costs when setting environmental standards. Although administering benefits-based standards is more complex than administering uniform technology-based standards, the administrative tasks are no more complex than those already undertaken for ambient-based standards. Balancing benefits and costs will require difficult and sometimes uncertain analysis, but EPA's evaluations of the current ambient-based standards indicate that much of this analysis is being taken into consideration already, at least implicitly.38 Thus, applying the benefits approach may not be such a radical departure from current practices. The change will be more in the decision criterion used (net benefits versus ambient quality), than in the way things are done. As our analysis shows, simply changing the decision criterion could lead to a vast improvement in the net benefits that society realizes from environmental management. 1. 42 U.S.C. §§ 7401-7626, ELR STAT. CAA 001-052 (1970). 2. We use "benefits-based" to refer to standards that are established by comparing potential benefits with potential costs. These standards might more precisely be referred to as net-benefits-based standards. The term benefits-based is used for simplicity and readability, however, and should be interpreted as implying an evaluation of net benefits. 3. 33 U.S.C. §§ 1251-1387, ELR STAT. FWPCA 001-065 (1972). 4. See generally W. RODGERS, ENVIRONMENTAL LAW: AIR AND WATER (1986). 5. See generally Latin, Ideal Versus Real Regulatory Efficiency: Implementation of Uniform Standards and "Fine-Tuning" Regulatory Reforms, 37 STAN. L. REV. 1267 (1985). 6. See generally Freeman, Technology-Based Effluent Standards: The U.S. Case, 6 WATER RESOURCES RESEARCH 21 (1980); Krier, The Irrational National Air Quality Standards: Macro and Micro Mistakes, 22 UCLA L. REV. 323 (1974); Pedersen, Turning the Tide on Water Quality, 15 ECOLOGY L.Q. 69 (1988). 7. See generally Baum, Legislating Cost-Benefit Analysis: The Federal Water Pollution Control Act Experience, 9 COLUM. J. ENVTL. L. 75 (1983). 8. See generally LaPierre, Technology-Forcing and Federal Environmental Protection Statutes, 22 IOWA L. REV. 323 (1977); Randle, Forcing Technology: The Clean Air Act Experience, 88 YALE L.J. 1713 (1979). 9. See generally E. GRAMLICH, BENEFIT-COST ANALYSIS OF GOVERNMENT PROGRAMS (1981); E. STOKEY & R. ZECKHAUSER, A PRIMER FOR POLICY ANALYSIS (1978). 10. A. FREEMAN, AIR AND WATER POLLUTION CONTROL: A BENEFIT-COST ASSESSMENT (1982). 11. A. FREEMAN, Water Pollution Policy, in PUBLIC POLICIES FOR ENVIRONMENTAL PROTECTION (P. Portney, ed.) (1990). 12. R. LUKEN, EFFICIENCY IN ENVIRONMENTAL REGULATION: A BENEFIT-COST ANALYSIS OF ALTERNATIVE APPROACHES (1990). 13. A. MARCUS, PROMISE AND PERFORMANCE: CHOOSING AND IMPLEMENTING AN ENVIRONMENTAL POLICY (1980). 14. The development of the original total suspended particulates (TSP) national ambient air quality standard did not follow the more rigorous and transparent process that characterized the 1987 particulate matter standard. Zaragoza reviewed the original particulate matter criteria document and found that community observational or epidemiological studies played the major role in setting the primary (health) standards. L. Zaragoza, The Use of Biological Information in the Development of National Ambient Air Quality Standards (1982) (Ph.D. thesis, UCLA). 15. 36 Fed. Reg. 8186 (1971). In 1987, EPA revised the particulate matter ambient air quality standards. 52 Fed. Reg. 24634 (1987). The revisions changed the focus of the standard from total suspended particulate matter to particulate matter less than 10 microns in diameter. 16. 43 Fed. Reg. 7568 (1978). In 1986, EPA issued minor revisions in the new source performance standards for TRS. 51 Fed. Reg. 18538 (1986). 17. 44 Fed. Reg. 58642 (1979). 18. U.S. ENVIRONMENTAL PROTECTION AGENCY, KRAFT PULPING: CONTROL OF TRS EMISSIONS FROM EXISTING SOURCES, EPA-450/2-78-0036 (1979). 19. 39 Fed. Reg. 18742 (1974); 42 Fed. Reg. 1398 (1977). 20. 47 Fed. Reg. 52006 (1982). 21. The benefits of reduced mortality were based on estimates of people's willingness to pay to avoid increased risk. The benefits of reduced morbidity were based on cost of illness estimates taking into account loss of work time, the value of reduced activity, and medical costs. Welfare benefits included an estimate of cost savings from reduced household soiling and material damage. 22. Data from R. LUKEN, supra note 12. 23. See R. CRANDALL, CONTROLLING INDUSTRIAL POLLUTION: THE ECONOMICS AND POLITICS OF THE CLEAN AIR ACT (1983). 24. Data from R. LUKEN, supra note 12. 25. U.S. OFFICE OF TECHNOLOGY ASSESSMENT, CATCHING OUR BREATH: NEXT STEPS FOR REDUCING URBAN OZONE, OTA-0-412 (1989). 26. 15 U.S.C. §§ 2601-2671, ELR STAT. TSCA 001-056. 27. 7 U.S.C. §§ 136-136y, ELR STAT. FIFRA 001. 28. 55 Fed. Reg. 6482 (1990). 29. See A. KNEESE, MEASURING THE BENEFITS OF CLEAN AIR AND WATER (1985). 30. U.S. WATER RESOURCES COUNCIL, ECONOMIC ENVIRONMENTAL PRINCIPLES AND GUIDELINES FOR WATER RELATED RESOURCES: IMPLEMENTATION STUDIES (1983). 31. See U.S. ENVIRONMENTAL PROTECTION AGENCY, EPA's USE OF BENEFIT-COST ANALYSIS: 1981-1986, EPA-230-05-87-028 (1987). 32. Exec. Order No. 12291, 3 C.F.R. § 127 (1982), ELR ADMIN. MATERIALS 45025. 33. See The Conservation Foundation, The Environmental Protection Act (second draft 1988) (defining unreasonable risk). 34. EPA recently instituted a strategic planning process. U.S. Environmental Protection Agency, EPA Strategic Planning and Budgeting (memorandum of Administrator William K. Reilly, 1989). The stated intention of the process is to promote and facilitate a more rational allocation of the limited resources available to the agency among its activities to reduce risks to health and the environment. The process will be difficult because the managers of the process have not secured a focused mission statement, the first step in a strategic planning process. 35. For similar approaches see Ackerman & Stewart, Reforming Environmental Law, 37 STAN. L. REV. 1333 (1985); Haigh, Harrison & Nichols, Benefit-Cost Analysis of Environmental Regulations: Case Studies of Hazardous Air Pollutants, 8 HARV. ENVTL. L. REV. 395 (1984). 36. See OFFICE OF POLICY, PLANNING AND EVALUATION, U.S. ENVIRONMENTAL PROTECTION AGENCY, UNFINISHED BUSINESS: A COMPARATIVE ASSESSMENT OF ENVIRONMENTAL PROBLEMS (1987). 37. TEMPLE, BARKER & SLOANE, INC., DRAFT REGULATORY IMPACT ANALYSIS OF PROPOSED REVISIONS TO SUBTITLE D CRITERIA FOR MUNICIPAL SOLID WASTE LANDFILLS (Dec. 11, 1989) (report to the Economic Analysis Staff, Office of Solid Waste, U.S. Environmental Protection Agency). 38. See U.S. ENVIRONMENTAL PROTECTION AGENCY, supra note 36. 20 ELR 10419 | Environmental Law Reporter | copyright © 1990 | All rights reserved |