2.6. EPA Guidance
1. 1983 EPA Guidelines
In December 1983, the EPA issued Guidelines for Performing Regulatory Impact Analysis
as part of an effort to develop a consistent approach among its program offices in the way the agency responded to the requirements of Executive Order 12291. In outline form the guidelines call for the following information in agency RIAs:
The RIA should describe the environmental problem to be addressed as seen by the Agency, industry, labor and public interest organizations. What market imperfections will be addressed? Which pollutant is posing a problem? Where is this pollutant discharged and in what volumes? What is the impact of the pollutant on public health and the environment? What are available control measures and what is their effectiveness? How would the regulation affect these releases?
The RIA should examine the most important regulatory alternatives, including those of no action and alternatives to regulation such as negotiated voluntary actions, market mechanisms, and state and local regulatory action. Alternatives to consider within the legislative mandate include three broad variants: changes in the degree of control; changes in the effective compliance date; and methods for ensuring compliance.
The benefits resulting from decreases in pollution relative to baseline levels should be measured in monetary terms insofar as possible. Ordinarily one follows a chain of events from release of pollutants to changes in ambient quality, to changes in exposure, to resultant impacts on health and the environment, to monetary measures of these effects. The guidelines offer explicit instructions regarding the quantification of health effects and methods for valuing health and environmental impacts in monetary terms. Valuation should be in terms of willingness to pay.
Cost analysis should include more than estimated compliance costs. The proper measure, total cost to society, is the value of goods and services lost as a consequence of the regulatory action. This cost includes private sector real-resource costs, government regulatory costs, dead-weight welfare losses, and adjustment costs, all relative to the baseline of no regulatory action.
Benefits and costs must be evaluated. Generally the RIA should provide three types of analysis: estimates of the net benefits (benefits minus costs) for each regulatory alternative along with a discussion of the nonmonetized effects; a schedule showing when the benefits and costs would occur; and a cost effectiveness analysis of each major alternative for situations when many benefits are not easily monetized or when the governing statute specifies the regulatory objectives.
2. 1991 EPA Guidelines
Between 1983 and 1989 the agency prepared clarifications and extensions of various parts of the guidance document and issued these in its 1991 Guidelines for Preparing Regulatory Impact Analysis
as appendices to the 1983 guidance. The appendices included: Appendix A - Analysis of Benefits; Appendix B - Analysis of Costs; Appendix C - Analysis of the Choice of Discount Rates; Supplement to Appendix C - Analysis of the Choice of Discount Rate; and Appendix D - Analysis of Economic Impacts. These appendices are described briefly here.
Appendix A, Analysis of Benefits, describes the conceptual approach for estimating benefits, which is changes in producer and consumer surplus. Valuation is in terms of willingness to pay. The appendix also describes the available analytical methods for estimating benefits and their advantages and disadvantages. The principal methods include cost saving, damage function, travel cost, contingent valuation. Finally the appendix provides a review of some recent applications of methods for measuring benefits for health (both morbidity and mortality), materials, recreation, aesthetics, and ecosystems.
Appendix B, Analysis of Costs, describes two conceptual frameworks for estimating social costs: a dynamic general-equilibrium approach and a static partial-equilibrium approach. The static partial equilibrium approach is likely to be more practical to estimate. The appendix notes at the outset that cost analyses of the agency have focused almost exclusively on compliance costs of private parties affected by a regulation. While this element of social cost usually is the largest component, other elements of social cost also should be evaluated, including government regulatory costs, deadweight welfare losses and adjustment costs for displaced resources. The appendix also contains information on how to estimate the costs of atypical regulatory alternatives such as trading of pollution entitlements, pollution charges, recycling incentives, pollution indemnity, information and labeling requirements, and government cost sharing.
Appendix C, Analysis of the Choice of Discount Rates, discusses the concept of discounting, why it is appropriate to discount to net present values future benefits and costs, annualizing costs, and how to analyze the sensitivity of calculated net present values to changes in the discount rate. The appendix alseo describes several alternative methods for selecting a discount rate, including the opportunity costs of displaced resources, the social rate of time preference, and the shadow price of capital approach. The appendix concludes that discounting by the marginal rate of return on investment, which was approximated when the paper was prepared in 1984 by the OMB suggested 10 percent real rate, is most appropriate for environmental regulatory actions that have a relatively short time horizon. Various approaches having lower discount rates, such as the opportunity cost of capital, are recommended for environmental regulatory activities that span several generations.
Supplemental guidelines for the choice of discount rates, prepared as an addendum to Appendix C, were issued in 1989. The supplement explains a procedure for obtaining a discount rate based on a blending of rates computed from returns on displaced private investments and the consumption opportunities foregone as a result of regulation. The so-called Kolb-Scheraga procedure first annualizes capital costs over the expected lifetime of the pollution controls using the marginal rate of return on private investment; then it discounts both cost and benefit streams using the consumption rare of interest (a proxy for the unknown social rate of time preference). To assist analysts, the Office of Policy, Planning and Evaluation developed a simple menu-driven computer program (included within the supplement) to perform the calculations.
Appendix D, prepared in 1984, provides guidance on how to evaluate the impacts of regulatory actions. The general approach should measure regulatory effects against a baseline situation for the affected industry in the absence of the regulatory action. The baseline should be defined in one of three ways: status quo if regulation is not enacted, present industry trends in pollution control continue, or the baseline differs from either of these alternatives due to changes in the relative prices of inputs or technological change. A discounted cash flow analysis should be undertaken to determine if individual facilities are likely to close as a result of regulation and the ability of an industry to pass costs through should be based on an assessment of supply and demand elasticities. The primary impacts of interest include: price effects, production effects, effects on profitability, employments effects and income distribution effects. Secondary effects to be analyzed include secondary employment effects, community effects and energy and balance of trade effects. Finally, intergenerational efficiency and equity effects should be assessed.
3. 1999 EPA Guidelines
Much has changed since the EPA issued its Guidelines for Performing Regulatory Impact Analysis in 1983. The original guidelines have remained largely unchanged, with modifications and additions to selected sections as noted above.
The September 1993 Executive Order 12866 "Regulatory Planning and Review" continues to require economic analyses involving costs, benefits and economic impacts and adds new requirements concerning distributional impacts. In itself, the new executive order would require relatively modest changes to the old EPA guidelines. However, several new mandates to perform other types of economic analysis have recently been enacted through law or executive order:
The Unfunded Mandates Reform Act of 1995 (PL 104-4) and Executive Order 12875, "Enhancing Intergovernmental Partnerships" instruct agencies to determine the effects of federal regulatory actions on state, local and tribal governments, and the private sector (in the UMRA), and to solicit input form state local and tribal governments for rules that contain "intergovernmental" mandates involving the transfer of $100 million or more in any one year.
The Regulatory Flexibility Act of 1980 (RFA) as amended by The Small Business Regulatory Enforcement Fairness Act of 1996 (PL 96-354) SBREFA) requires that federal agencies assess if a regulation will have a significant impact on small entiries including small businesses. If it will, then agencies must prepare a Regulatory Flexibility Analysis and consider regulatory options that minimize adverse economic impacts on small entities.
Executive Order 12898, "Federal Actions to Address Environmental Justice in Minority Populations and Low-Income Populations" requires that federal agencies determine and deal with as appropriate disproportionate adverse health or environmental effects of its programs or policies on low income and minority populations, including Native Americans.
Executive Order 13045
, "Protection of Children from Environmental Health Risks and Safety Risks" requires that agencies determine the health and safety effects of regulatory initiatives on children. In the case of regulations with a significant economic impact, agencies must explain why the alternative selected is preferred to other potentially effective and feasible alternatives.
Executive Orders 13083
on "federalism" require that agencies adhere to constitutional principles in the conduct of public policy. The Orders establish policy-making criteria including administrative flexibliity in regulations and accounting for unique circumstances that may be present in specific locations.
Executive Order 13084
on "Consultation and Coordination with Indian Tribal Governments" requires that agencies recognize unique legal relations with Indian tribal governments and that regular consultation and collaboration with Indian governments take place in the development of regulaitons, the imposition of unfunded mandates, and the process for seeking waivers from federal rules.
In addition, in the past 15 years methods for economic analysis, particularly benefits analysis, have improved significantly. For all of these reasons, the EPA decided in 1996 to begin a thorough review of the guidelines for performing regulatory impact analysis. This review is nearly complete. When new guidelines for preparing economic analyses are finalized, a summary of those guidelines and links to the relevant documents will be provided here.