2.5. OMB Guidance
On January 11, 1996, the Office of Management and Budget issued a guidance document
regarding the conduct and implementation of economic analysis required by Executive Order 12866
. Entitled "Economic Analysis of Federal Regulations," the memorandum was prepared by a Regulatory Working Group chaired by Joseph Stiglitz of the Council of Economic Advisors and Steven Kaplan, then-general counsel of the Department of Transportation. The report represents the group's effort to describe "best practices" for economic analysis (EA) to comply with the analytic requirements of the executive order. The report cautions that it is not a "mechanistic blueprint," rather it is designed to help agencies think through issues and methods and to prepare useful regulatory analyses.
The OMB guidelines are divided into an introduction and three sections: (1) a statement of the need for the proposed action; (2) an examination of alternative approaches; and (3) analysis of benefits and costs. The introduction states that an Economic analysis (EA) should provide sufficient information for decisionmakers to determine:
there is adequate information demonstrating a need for and the consequences of the proposed regulatory action,
the potential benefits justify the potential costs, considering that not all benefits can be expressed in monetary terms,
the proposed action will maximize net benefits to society, unless a statute mandates another approach,
where a statute mandates a specific approach, the proposed action will be the most cost-effective, and
agency decisions are based on the best reasonably available scientific, technical economic and related information.
(1) Statement of need for proposed action
To establish the need for proposed regulatory actions, OMB's guidance calls on agencies to evaluate whether the problem stems from a significant market failure. If market failure is not the origin of the problem, the analysis should make a case for compelling public need such as distributional concerns or improving government processes. If the rational for action is a statutory or judicial directive, that should be stated.
If the proposed action stems from market failure, the guidance calls on agencies to evaluate whether the problem could be addressed at modest cost by market participants instead of through a federal regulatory process. If a significant market failure is identified, the analysis should demonstrate how regulatory alternatives under consideration would address the problem.
Four major types of market failure are described. (1) Externalities occur when the action of one party impose costs on another. Environmental pollution provides a good example. Another type of externality may occur with common property resources; the costs to society are greater than the costs borne by individual users. Public goods are a third type of externality; it is impossible to exclude nonpaying individuals from the benefits. (2) A second type of market failure arises when a market is served at lowest cost by a single provider. (3) Firms in monopolistic and oligopolistic industries may exercise market power and reduce output below what would be provided by a competitive industry. (4) Market failure also may follow from the inadequate or asymmetric distribution of information. Inadequate information can result in various social costs such as lack of investment in innovation or the deception of consumers.
Because government actions themselves may contribute to inefficient market outcomes, the guidance recommends a particularly heavy burden of proof for regulations that involve price controls, production or sales quotas, mandatory product or service quality standards, or limitations on employment or production.
Even in the case of market failure, there may be other or better means of dealing with the problem than federal regulatory action. Such alternatives could include antitrust enforcement, workers' compensation systems, and the judicial system (e.g., litigation over environmental harms). Other nonregulatory alternatives such as economic instruments also might be superior means of dealing with pollution externalities and other forms of market failure.
(2) Examination of Alternative Approaches
The EA should consider important alternative approaches and provide the agency's rationale for selecting the preferred alternative. The number of alternatives to consider is a matter of judgement, recognizing a trade-off between the thoroughness of each analysis and the number of alternatives considered. Alternative regulatory actions to be investigate should include:
performance oriented standards where there truly are alternative ways of meeting the standard;
different requirements for different portions of the regulated community;
alternative levels of stringency;
different effective dates of compliance;
various means of ensuring compliance, such as inspections, penalties for noncompliance, and periodic reporting;
means to improve the availability of information, including mandatory disclosure; and
considering specific statutory requirements.
(3) Analysis of Benefits and Costs
Following a preliminary analysis to identify a set of alternatives for review, the agency needs to pay particular attention to several features of the analysis.
Baselines are the benchmark against which the costs and benefits of regulatory alternatives are measured. The baseline should represent the best assessment of the situation without the proposed regulation. The assessment may incorporate many factors, such as population and income growth, evolution of the market, probable changes in other regulations that may affect regulated entities, and the likely degree of compliance by the regulated community. When more than one baseline appears appropriate for analysis, the agency may estimate costs and benefits against more than one baseline as a form of sensitivity analysis.
Agencies should identify alternatives that satisfy the criteria of the executive order as well as identify the statutory requirements that affect the choice among alternatives. Benefits and costs should be presented in discounted constant dollars. Distributional impacts should be described. In choosing among alternatives, the benefit-cost ratio must be used with care as the highest ratio may not identify the preferred alternative that maximizes net benefits. Further the internal rate of return is not a proper criterion for choosing among alternatives. When monetization of benefits or costs is not possible, other quantitative and qualitative descriptions of the outcomes should be used. When the level of benefits is specified by statute, cost-effectiveness analysis should be used to compare alternatives.
Future benefits and costs should be discounted to a present value equivalent for the calculation of net benefits. The basic guidance on discount rates is specified in OMB circular A-94. The Circular A-94 rate averages returns on low yielding capital investments such as housing with higher-yielding corporate capital and averaged 7 percent in real terms in early 1996. OMB stated that even hard to quantify benefits should be discounted. Future price changes that reflect increasing relative scarcity or abundance of a resource should be taken into account to the extent feasible.
Risk and uncertainty need to be reflected in two separate and distinct steps: the risk assessment and a subsequent valuation of the levels and changes in risk experienced by affected portions of the population. Important issues in risk assessment include the quality and reliability of the data, models and assumptions, and methods of scientific inference. All of these areas involve uncertainties that need to be reflected in the analysis. Making the analysis transparent and fully disclosing all assumptions and methods is essential for making informed judgments. Overall uncertainty is a reflection of uncertainties at many stages. These cumulative uncertainties may be reflected in (1) Monte Carlo analysis and other simulation methods, (2) sensitivity analysis, (3) Delphi methods, and (4) meta-analysis. Risk and changes in risk should be valued in certainty equivalents.
Opportunity cost is the proper principle for valuing both benefits and costs. Either willingness-to-pay or willingness-to-accept can capture the notion of opportunity cost, though most economists prefer WTP because of empirical difficulties in estimating WTA. A variety of methods exist for measuring benefits: direct market measurement; indirect measurement through market transactions; and direct surveys of WTP. Costs also should be measured using the principle of opportunity cost. The methods of quantifying both benefits and costs are elaborated extensively in the guidance.