March 29, 2011

Improving Pre-Proposal Regulatory Analysis

Testimony Before the House Judiciary Committee
  • Jerry Ellig

    Research Professor, George Washington University Regulatory Studies Center
Key materials
Contact us
To speak with a scholar or learn more on this topic, visit our contact page.

Somewhere along the line in our schooling, most of us learn about a few basic steps to take before making a major decision. These steps include: (1) define the goal or goals you are trying to achieve, (2) understand the nature of the obstacles to be overcome or problems to be solved to achieve the goal, (3) develop a list of alternative ways to solve the problem, and (4) assess the pros and cons of each alternative. After taking these steps, deciding on a course of action may still be challenging. The decision may involve difficult tradeoffs between some of the pros and cons. But at least the decision will be informed by knowledge of the likely consequences of alternative actions. Call this “Decision Making 101.” I don’t know any reasonable person who would argue we should not do these things when making decisions that have really big effects on our lives or the lives of others.

Regulatory impact analysis is nothing more than Decision Making 101 applied to regulation: (1) define the outcome or outcomes the regulatory agency seeks to achieve, (2) understand the root causes of the problem that stands in the way of achieving the desired outcomes, (3) develop a wide variety of alternative ways to solve the problem, and (4) assess the pros and cons of each alternative. If we cut through all the jargon and details, these are the four main elements of regulatory analysis.

We expect federal regulation to accomplish a lot of important things, such as protecting us from financial fraudsters, preventing workplace injuries, preserving clean air, and deterring terrorist attacks. And regulation also requires sacrifices. Depending on the regulation, consumers may pay more, workers may receive less, our retirement savings may grow more slowly due to reduced corporate profits, and we may have less personal freedom. Regulatory analysis is the key ingredient that makes these tradeoffs more transparent to decision makers. So, understanding the effects of regulation has to start with sound regulatory analysis.

For more than three decades, executive orders have instructed federal agencies to conduct regulatory impact analyses and consider the results of those analyses when making decisions. On January 18, President Obama issued Executive Order 13563, “Improving Regulation and Regulatory Review." Executive Order 13563 “reaffirms the principles, structures, and definitions governing contemporary regulatory review that were established in Executive Order 12866.” In fact, Executive Order 12866 reaffirmed the principles originally established 30 years ago with Executive Order 12991.

This reaffirmation is welcome. The analytical principles in Executive Order 12866 and the Office of Management and Budget’s accompanying guidance in Circular A-4 are sound. The administration’s reaffirmation of those principles may help quell some uncertainty about future standards for regulatory review that has existed since the administration announced in January 2009 that it planned to revise the executive order.

But how well do executive branch agencies do what presidents have been telling them to do for more than three decades? Scholarly research on regulatory analysis, including the Mercatus Center’s own Regulatory Report Card, finds that agency regulatory analysis is often incomplete and seldom used in decisions. This pattern persists across administrations, suggesting that the source of the problem is institutional, not political. Fundamental institutional reforms are necessary to ensure that agencies conduct high-quality regulatory impact analysis and use it in decisions. In short, regulatory impact analysis needs to be (1) required, (2) objective, and (3) used.

The body of my testimony documents current problems with the quality and use of regulatory analysis and suggests some solutions. Let me briefly summarize my recommendations:

1. Regulatory impact analysis should be required. Regulatory analysis needs to be legislatively required for all federal agencies, including independent agencies.

2. Regulatory impact analysis should be objective. All too often, regulatory analyses read as if the agency first made most of the major decisions about the regulation, then handed the regulation off to its economists to produce an analysis to get the regulation through the OMB review process. Agencies should publish regulatory analysis, along with all underlying data and research, before writing proposed regulations. Agency economists should have the independence to conduct objective analysis, instead of being expected to produce an advocacy document that justifies decisions that have already been made.

3. Regulatory impact analysis should be used. When Congress requires regulatory agencies to consider particular factors in designing regulations, such as costs or efficiency, agencies usually explain how those factors affected their decisions. Congress should require all agencies to explain, when proposing regulations, how the major elements of regulatory analysis affected decisions about the regulation.