February 23, 2017

Regulatory Process, Regulatory Reform, and the Quality of Regulatory Impact Analysis

  • Jerry Ellig

    Research Professor, George Washington University Regulatory Studies Center
  • Rosemarie Fike

    Instructor of Economics, Texas Christian University
Usage RIghts

The paper to which this summary refers has been accepted for publication in the Journal of Benefit-Cost Analysis. Readers may use the typescript version of the full paper for private research purposes but may not redistribute it. The published version of record is available at http://dx.doi.org/10.1017/bca.2016.20.

A regulatory impact analysis (RIA) identifies the nature and cause of the problem a regulation is intended to solve, develops alternative solutions, and assesses the benefits and costs of the regulation and its alternatives. This analysis is a key element of the regulatory process in developed and developing nations alike. However, evaluations of RIAs by independent scholars have found that their quality often falls short of expectations. A poor RIA increases the risk that a regulation addresses a nonexistent problem or fails to solve the problem at a reasonable cost.

In “Regulatory Process, Regulatory Reform, and the Quality of Regulatory Impact Analysis,” recently published in the Journal of Benefit-Cost Analysis, Mercatus Senior Research Fellow Jerry Ellig and Texas Christian University economist Rosemarie Fike combine newly gathered data on regulatory processes with RIA scores from the Mercatus Center’s Regulatory Report Card to assess how RIA quality varies with the type of effort expended by agencies and the Office of Information and Regulatory Affairs (OIRA).

The study’s results suggest that three reforms could noticeably improve the quality of RIAs: (1) require agencies to seek public comment on an analysis of alternatives before proposing a regulation, (2) require agencies to consult with stakeholders and experts before proposing regulations, and (3) expand the resources and influence of OIRA, which reviews executive branch regulations and their accompanying analyses.

ANALYSIS

The effects of agency and OIRA effort on the quality of RIAs are evaluated by combining measurements of RIA quality with new data on regulatory procedures.

  • RIA quality is measured by the Mercatus Center’s Regulatory Report Card, a project that scores the quality of RIAs using evaluation criteria that mirror the Office of Management and Budget’s Regulatory Impact Analysis Checklist.
  • These measurements are combined with a new dataset that tracks specific types of activities that agencies and OIRA have devoted to the production and review of RIAs for each proposed rule.
  • This study tracks agency activities including the agency’s prior public requests for comments or information related to the analysis; advance consultation with state and local governments; use of expert advisory committees; public meetings with stakeholders; and agency commitments to hold public hearings on the proposed regulation in the future. OIRA’s effort and influence are measured by the number of days OIRA reviewed the proposed regulation and whether OIRA was headed by a presidentially appointed administrator or an acting administrator.

The analysis uses several different econometric methods appropriate to the data. The principal results are as follows:

  • Regulations for which the agency previously requested the public to provide comments or information tend to have more thorough RIAs, as do regulations for which agency officials consult with state and local governments.
  • Use of an advisory committee appears to be positively correlated with RIA quality, but this result depends on the econometric estimator used.
  • Regulations proposed after a public meeting with stakeholders tend to have lower-quality RIAs.
  • An acting OIRA administrator is associated with lower RIA quality than a presidentially appointed administrator.
  • The time it takes OIRA to review a regulation is positively correlated with RIA quality. The number of review days is subject to diminishing marginal returns, but longer OIRA review time is associated with higher-quality analysis for all but one regulation in the sample.

EXAMPLES

Because correlation need not imply causation, the paper includes several case studies that demonstrate how agency and OIRA effort led to higher-quality analysis:

  • Prior notice. The Department of Energy has an explicit policy of seeking public comment on the analysis that informs its decisions on energy efficiency regulations before it proposes a specific regulation. In response to comments on a preliminary analysis for a regulation establishing efficiency standards for residential refrigerators and freezers, the department gathered additional data on engineering and costs to substantiate, and in some cases revise, its estimates of the effectiveness and cost of regulatory alternatives.
  • State consultation. The state of Colorado worked with the US Forest Service on a 2008 rule regarding roadless national forest land that allowed some exemptions to address state and local concerns regarding fire hazard, insect and disease treatment, construction of electric and water facilities, and development—concerns the Forest Service may never have become aware of without this consultation with a state government. The Forest Service’s RIA then compared the effects of this 2008 rule with the effects of the federal government’s 2001 rule for roadless areas and the Forest Service’s land management plans, which would have taken effect if litigation had overturned the 2001 rule.
  • Advisory committees. The Department of Labor’s 2010 proposed revisions to rules governing miners’ exposure to dust benefited from reports from two dust task groups, one of which recommended several alternative margins that could be altered in the existing dust program to help reduce the health risks faced by miners.
  • OIRA review. During the OIRA review process, the FDA made two major improvements to the RIA for its proposed graphic warning labels on cigarette packages. First, it added an uncertainty analysis acknowledging that the benefits of the regulation may be zero. Second, the FDA calculated the incremental cost-effectiveness of alternative versions of the regulation.