30 Years of Regulatory Oversight: Lessons Learned, Future Challenges
Susan E. Dudley, Administrator
Office of Information and Regulatory Affairs
Office of Management and Budget
Executive Office of the President
Cost-Benefit of Regulations: Lessons Learned, Future Challenges
The Searle Center
Northwestern University School of Law
October 11, 2007
I've been a student of regulation for much of my career, so I'm honored to be among this august group of some of the finest scholars doing analytical research on regulatory matters today. Some of you I'm delighted to meet for the first time, while others have provided me solid advice, support, and even mentoring over the years. I thought I would talk a little bit about my experiences in my current position as Administrator of OIRA (for the last 6 months), and about OIRA's role in regulatory analysis and oversight over the last 3 decades. Then I'd like to speak directly to the topic of this conference, about what lessons we've learned, and what challenges lie ahead.
Former OIRA Administrators, many of whom are here today, say that this is the best job in Washington, but I think that's an understatement. It may be the best job in the world. I have to admit there are times when I need to be reminded of that though. The pace is very different from what I knew in academia, and the next year, with the inevitable increase in regulatory activity as we enter the midnight period, promises not to be dull.
What makes OIRA the best job is first its mission – to understand the consequences of different regulatory options before they are in effect, and ensure that the selected option maximizes net benefits to the public, to the extent possible subject to constraints. Those constraints of course can be formidable, but I'll talk about that when we get to "future challenges."
This clear mission attracts very capable and principled people, which is another major factor in making OIRA Administrator the best job in Washington. I work with a smart, motivated group of analysts who truly believe in getting the analysis right so that regulations really make the public better off. A few have been at OIRA since its inception in 1981, and worked with each of us OIRA Administrators over the years.
OIRA was created by the Paperwork Reduction Act of 1980, but it was President Reagan's Executive Order 12291 that gave the office the mandate to analyze regulations. E.O. 12291 required, to the extent permitted by law, that administrative decisions be based on adequate information concerning the need for and consequences of proposed government action, and that regulatory actions should maximize the net benefits to society.
When President Clinton took office in 1993, he replaced Executive Order 12291 with Executive Order 12866. In many ways, E.O. 12866 mirrors its predecessor. It reinforces the philosophy that regulations should be based on an analysis of the costs and benefits of all available alternatives, and that agencies should select the regulatory approach that maximizes net benefits to society, unless otherwise constrained by law. We continue to operate under E.O. 12866 today, with a few amendments I can talk about later, if you'd like. Indeed Congress has buttressed these analytical requirements through several statutes, including the Small Business Regulatory Enforcement Fairness Act (SBREFA), the Unfunded Mandates Reform Act, and the Regulatory Right to Know Act.
So, what lessons have we learned over the last 26 years, studying regulation under 4 different presidents representing both parties?
- Regulatory Analysis is here to stay.
Perhaps the most important lesson is that regulatory analysis has emerged as an integral part of government accountability – a non-partisan tool for understanding the likely effects of regulation. There are those who still object to attempts to quantify the impacts of regulation, and certainly those who also object to OIRA's role in regulatory coordination, but nevertheless, the principled approach to regulation articulated by Presidents Reagan and Clinton has withstood the test of time.
Other countries are emulating our approach. The U.S. was the first in the mid 1970's to implement a Regulatory Impact Analysis (RIA) requirement for major regulations. As late as 1990, only five other countries had RIA requirements. The success of our program as well as the strength of our economy soon led other governments to emulate as least parts of our approach. A recent report for Canada estimates that 23 of the 30 Organisation for Economic Co-Operation and Development (OECD) countries and eight non-OECD countries have RIA requirements in place. In addition, the European Commission has established an Impact Assessment Board that evaluates new legislation before it is implemented. It is still a new concept, and not all members of the EC and European Parliament are fully supportive. They are concerned that these analytical techniques may undermine their political role, but analysis doesn't need to supplant policy judgment, but rather can help inform it.
Those of you who know me well know I don't think quantitative benefit-cost analysis has all the answers, though I certainly think it has its place. Sometimes in our zeal for better quantification of costs and benefits, we lose sight of the more fundamental question of whether each problem we seek to address really calls for regulation.
- A good regulatory analysis is more than benefit-cost analysis.
Both President Reagan & President Clinton's Executive Orders directed agencies and OIRA first to identify the need for the regulatory action before undertaking benefit-cost analysis. President Clinton was more explicit than President Reagan regarding this first step, stating in Executive Order 12866, Section 1, in the Statement of Regulatory Philosophy and Principles:
Federal agencies should promulgate only such regulations as are required by law, are necessary to interpret the law, or are made necessary by compelling public need, such as material failures of private markets to protect or improve the health and safety of the public, the environment, or the well-being of the American people.
President Bush's recent amendments to Executive Order 12866 left that language in place, but made the "market failure" language more prominent in a subsequent subsection of Section 1:
Each agency shall identify in writing the specific market failure (such as externalities, market power, lack of information) or other specific problem that it intends to address (including, where applicable, the failures of public institutions) that warrant new agency action, as well as assess the significance of that problem, to enable assessment of whether any new regulation is warranted.
Increased emphasis on first identifying the compelling public need before launching into a benefit-cost analysis perhaps reflects a growing awareness that the best benefit-cost analysis in the world cannot improve upon an outcome if the initial decision was one that need not have been made collectively.
As Nobel Prize-winning economist Friedrich Hayek showed, decentralized market processes are superior to centralized regulatory solutions because decentralized markets focus dispersed information—information that no one individual (not even a regulator) can obtain—and convey it efficiently to market participants.1 James Suroweicki's delightful book, The Wisdom of Crowds, provides lots of evidence that individuals with diverse, localized knowledge can make choices, generate ideas, and solve problems far better than small groups of experts, no matter how well intentioned, knowledgeable, or intelligent.2
Government intervention substitutes the judgment of a small group of experts for the wisdom of crowds, and is best used in a limited way, such as in cases of a clear "market failure" that cannot be adequately addressed by other means. Even then, regulatory intervention may have unintended results. Government intervention, no matter how well intentioned, can never be as dynamic or responsive to individual choices and preferences as market forces.
Anecdotes about outcomes we don't like do not indicate a market failure, nor do they present a sufficient argument for government intervention. To be successful, good policy must be based on an understanding of why the wisdom of crowds aggregated through market transactions fails to allocate resources optimally in a given situation. Regulatory actions that do not explicitly recognize the market failure or systemic problem underlying the need for action are bound to be less effective than those that identify and correct the fundamental problem.
- Beware the Planner's Paradox
It is important to recognize that while markets are not perfect, neither are governments, and efforts to address perceived problems can often have unintended consequences. The "planner's paradox" is that planned solutions always look better on paper than unplanned solutions, because the planner sees only his "data, assumptions, biases and understandings of the way the world works . . . All of the unseen difficulties with the planned solution—the data, assumptions, biases, and understandings of the world that turn out to be wrong—are invisible to the analyst because the data he considers are his own."3 The most carefully analyzed regulations may result in unanticipated changes in behavior that undermine the desired effects of the regulation.
More regulation is not always in the public interest
The media has been intrigued lately by the observation that companies are asking the federal government to regulate them. Tobacco companies support legislation requiring FDA approval of cigarettes. Food and toy companies want more regulation to ensure their products' safety. Energy companies support cap and trade for greenhouse gas emissions.
This is not new. For decades, economists who have studied regulation have observed that regulation can provide competitive advantage, so it is often in the self-interest of regulated parties to support it. This is yet another reason to focus on identifying a compelling public need or market failure as a first step in considering regulatory action. It is easy to say "even industry wants regulation to protect consumers so it must be in the public interest," but such regulation may not really benefit consumers.
If we aren't careful to understand the core problem we're trying to address, regulation is likely to be biased toward benefiting interest groups that are better organized and have more to gain from the wealth redistribution generated by regulation, at the expense of other, perhaps larger interest groups with weakly felt preferences.
- While analysis has improved over last few decades, there is room for further
Circular A-4 guides agencies first to identify the core problem they are trying to address, and then examine alternatives that address that core problem. The analysis should be developed before decisions are made, and used to inform policy decisions. An after-the-fact analysis of a selected option may be helpful for understanding the likely consequences of the regulation, but is not as useful as an analysis of the consequences of several alternative approaches to achieving policy goals.
- Homeland security
Probably our biggest analytical challenge today is in analyzing homeland security regulations. They not only involve a difficult balancing of security and privacy goals, but many regulations to improve homeland security have benefits that are difficult to quantify. Consider that benefits are a function of the likelihood and severity of a hypothetical future terrorist attack and you can easily understand that forecasting, quantifying, and monetizing those benefits is a major challenge. The ten major, economically significant homeland security rules that have been finalized since 2004 are estimated to impose estimated costs of $2.2 billion to $4.1 billion a year on the economy (2001 dollars). The corresponding benefits of those regulations were not quantified. The Department of Homeland Security is increasingly attempting to address this challenge by performing break-even analysis – estimating the security enhancement a proposal would have to provide to justify estimated costs.
- Legislative impediments to Regulatory Analysis
Many statutes that grant agencies regulatory authority still don't call for analyses of the impacts of those regulations, and some expressly preclude consideration of the likely costs and benefits. The Safe Drinking Water Act of 1996 was an exception. It emphasized sound science and risk-based standard setting, as well as balancing of benefits and costs, and flexibility for small water supply systems.
In addition, very short statutory deadlines for issuing regulations often preclude any analysis of the likely effects, or unintended consequences.
- Global impacts
The global consequences of domestic regulations are increasingly important. The agenda of the Transatlantic Economic Council, established by then-European Union President Merkel and President Bush in April 2007, includes regulatory coordination goals, such as better information-sharing on regulatory activities, risk assessment, and sector-specific coordination and mutual recognition. Circular A-4 currently reminds agencies to consider the role of federal regulation in facilitating U.S. participation in global markets. My office is working with our European counterparts to consider expanding that analysis to ensure we consider international trade & investment effects when developing new regulations.
Regulatory analysis will be particularly important over the next year. We are entering the final phase of this administration, with the corresponding pressure to complete regulatory agendas. The careful oversight and analysis that being stressed at this conference will be essential if we are to understand the possible consequences, intended and unintended, of these regulatory activities.
1 Friedrich A. Hayek, The Use of Knowledge in Society, American Economic Review, XXXV, No. 4, 519-30 (September 1945).
2 James Surowiecki, The Wisdom of Crowds: Why the Many are Smarter than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations (2004).
3 Brian Mannix, The Planner's Paradox, Regulation (Summer 2003), available at