Best and Worst Practices
This Mercatus on Policy shows how regulatory process reforms should require agencies to thoroughly analyze alternatives and publish that analysis for public comment before they propose a regulation.
For more than three decades, presidents have instructed federal agencies to consider a wide variety of alternatives to regulation as well as alternative types of regulation. Agency compliance has been uneven at best—largely because agencies often decide what regulation to issue before they even consider alternatives. Agencies sometimes do examine the pros and cons of alternatives, but this is the exception rather than the rule. To remedy this problem, regulatory process reforms should require agencies to thoroughly analyze alternatives and publish that analysis for public comment before they propose a regulation.
Why Consider Alternatives?
Gordon Tullock, an internationally renowned economics professor long affiliated with George Mason University, often recounted the story of a Roman emperor who judged a singing contest. The first contestant sounded so horrible that the emperor immediately gave the prize to the second.1 The emperor’s foolishness is obvious: the second singer might have been even worse. The story illustrates the commonsense idea that policy makers need to hear all alternatives before they choose one.
Executive orders and Office of Management and Budget (OMB) guidance direct agencies to analyze alternatives to regulation and alternative types of regulation when they conduct a Regulatory Impact Analysis (RIA). Executive Order 12866, which governs regulatory analysis and review, states that, “Each agency shall identify and assess available alternatives to direct regulation ...” 2 President Obama’s Executive Order 13563 repeats this same language .3
OMB Circular A-4, which outlines best practices in regulatory analysis, explicitly directs agencies to analyze alternatives outside the scope of their current legal authority if such alternatives “best satisf[y] the philosophy and principles of Executive Order 12866.”4 Thus, even when legislation gives agencies little discretion, they are expected to assess a wide variety of alternatives to inform the president and Congress. Specifically, OMB Circular A-4 recommends that agencies consider the following approaches:
- Information measures rather than regulation
- Market-oriented approaches rather than direct controls
- Performance standards rather than design standards
- Different requirements for different geographic regions
- Different requirements for different-sized firms
- Different degrees of stringency
- Different enforcement methods
- Different compliance dates
- Different choices defined by statute
Unfortunately, agencies thoroughly assess a wide variety of alternatives for only a handful of major regulations. The Mercatus Center’s Regulatory Report Card—an in-depth evaluation of the quality of analysis for major regulations—evaluates agencies’ RIAs. Regulations receive a score ranging from 0 (no useful content) to 5 (comprehensive analysis with potential best practices) on questions related to agency consideration of alternatives, as well as other criteria.5 Only a few regulations assessed in 2008 and 2009 received the maximum possible 5 points for considering a wide range of alternatives and estimating the benefits of each. While many regulations do consider some alternatives, these alternatives often fail to receive the same level of scrutiny as the option chosen. The Report Card contains several diagnostic questions that assess how well the RIA evaluated alternatives to the proposed regulation. Below are examples of best and worst practices from these questions.
Best and Worst Practices in Analysis
Does the analysis enumerate other alternatives to address the problem?
Worst practice: Consider no serious alternatives to the proposed rule.
Seven regulations out of 87 scored a zero on this criterion in 2008 and 2009, meaning they did not identify any alternatives to the proposed regulation. The 2009 Department of Energy (DOE) rule on investing in innovation states that it considered a wide variety of alternatives, but never lists them. The 2008 Department of Health and Human Services Medicaid program premiums and cost-sharing rule simply states that Congress required this specific regulation, therefore no alternatives were considered.
Best practice: Consider numerous alternatives.
Naming alternatives is not in and of itself a best practice, but when agencies seriously consider a large number of varying approaches, this is a good start. DOE’s 2008 energy-conservation standards for general-service fluorescent lamps and incandescent reflector lamps considered nine alternatives, including non-regulatory alternatives.
Is the range of alternatives considered narrow or broad?
Worst Practice: Consider only no action or the proposed rule.
The Interior Department’s 2008 abandoned mine land program considered only the proposed rule or no regulatory action. The Department of Transportation’s (DOT) 2008 regulation establishing new maximum operating pressures for gas pipelines considered only its proposed rule and delaying the proposed rule. While it is advisable to consider the impact of no new regulation, agencies should consider further alternatives.
Best Practice: Present a wide variety of alternatives, including non-federal options and/or non-regulatory actions.
The RIAs for DOE’s energy efficiency-regulations regularly consider a wide variety of alternatives. For example, when proposing standards for pool heaters and water heaters in 2009, the department considered eight alternatives to the proposed regulation:
- No new regulatory action
- Consumer rebates
- Consumer tax credits
- Manufacturer tax credits
- Voluntary energy-efficiency targets
- Early replacement subsidies (similar to “cash for clunkers”)
- Bulk government purchases
- Allow states to incorporate requirements for high-efficiency storage water heaters in building code
Does the analysis evaluate how alternative approaches would affect the amount of the outcome achieved?
Worst Practice: Dismiss alternatives without evidence to support decision.
The analysis for the Department of Labor’s 2008 regulations for the use of cranes and derricks in construction mentions several alternatives, but dismisses them. Firms with unsafe practices may find that they have to pay compensating differentials in wages or higher insurance premiums, but the agency summarily dismisses the effects of these market-based incentives. The agency acknowledges that some states have safety plans approved by the Occupational Safety and Health Administration, but it does not consider whether any states already address crane and derrick safety. No empirical evidence or scholarly literature is cited to support dismissing the alternatives.
Best practice: Describe outcomes for each alternative and monetize them to facilitate comparison.
The RIA for the Environmental Protection Agency’s (EPA) 2008 effluent limitations guidelines for construction sites shows how different alternatives will affect the amount of outcome achieved. The agency estimates the reductions in costs for dredging sediment and treatment of drinking water under each alternative. Evidence is presented in the text as well as in a table. Outcomes are presented in physical as well as monetized form and discounted at multiple discount rates.
Does the analysis identify and quantify incremental costs of all alternatives considered?
Worst Practice: Consider no costs at all, or only the cost of the alternative chosen.
RIAs almost always contain some information about the cost of the proposed regulation. However, many fail to include thorough cost analysis of the alternatives, if alternatives are considered. This conveys the impression that the RIA’s main purpose is to justify a decision already made, rather than to inform decisions.
Best Practice: Identify all costs, both direct and indirect, for each alternative.
The analysis for the 2008 EPA effluent limitations guidelines for construction sites breaks down costs borne by firms and the government, as well as increases in the cost of single-family housing under each alternative. EPA estimates the direct costs to industry under each alternative, as well as economy-wide effects on consumers (see Table 1).
Table 1: EPA Comparison of Social Costs and Benefits for 2008 Effluent Limitations GuidelinesDoes the analysis identify the alternative that maximizes net benefits?
Worst Practice: Do not compare net benefits of alternatives.
Twenty-one regulations scored a zero on this question in 2008 and 13 scored a zero in 2009. This is nearly 40 percent of all regulations evaluated for the Mercatus Regulatory Report Card. These RIAs had no content related to identifying alternatives that maximize net benefits.
Best Practice: Calculate the net benefits associated with each alternative.
Table 1 shows a best practice for presenting information on net benefits. After monetizing the costs and benefits, EPA calculates net benefits for each alternative. (Estimated net benefits are negative because EPA notes several significant benefits were not monetized.)
Does the analysis identify the cost-effectiveness of each alternative considered?
Worst Practice: Do not compare cost-effectiveness of alternatives.
Seventeen regulations scored a zero on this question in 2008 and 21 scored a zero in 2009, meaning over 43 percent of regulations scored for the Mercatus Regulatory Report Card failed to assess the cost-effectiveness of alternatives.
Best practice: Calculate cost-effectiveness, defined as outcomes divided by total costs, for each alternative.
A 2009 Department of Agriculture meat-labeling rule includes a cost-effectiveness analysis showing costs-per-life-saved of the various alternatives. These costs vary from $291,000 to $2.2 million depending upon the estimated effectiveness of the alternatives. The department deserves credit for calculating the cost-effectiveness as well as the net benefits of each alternative and for accounting for uncertainty about the success rates of each alternative.
The best practice examples clearly demonstrate that agencies can thoroughly assess the pros and cons of alternatives to a proposed regulation. But such thorough analysis is the exception rather than the rule.
Agency analysts often lack the incentive and opportunity to evaluate diverse alternatives. In many cases, agencies decide on the favored regulatory option before analysts even evaluate alternatives. 6 One expert characterized the problem succinctly: “RIAs may serve primarily as a mechanism for promoting agency decisions rather than scrutinizing them.” 7
Agencies should be required to assess alternatives before they decide on proposed rules. The analysis and underlying data should be published for public comment before the regulation is written.8 This will ensure that America’s regulatory system avoids the mistake of the Roman emperor who awarded the prize before he heard the second contestant sing.
View the paper and footnotes as a PDF