Over the last thirty years, most legislative and executive branch efforts at regulatory reform have focused on analyzing and improving new regulations, and agencies seldom look back to evaluate whether existing regulations are having their intended effects. Section 610 of the Regulatory Flexibility Act provides for periodic review of regulations for their impact on small businesses, but researchers have found that most agencies “comply with the letter of the law for only a small percentage of their rules, and they rarely take action beyond publishing a brief notice in the Federal Register.”

Senator Mark Warner hopes to change that. He is drafting legislation focused on altering regulatory agencies’ incentives to issue new regulations and examine the effectiveness of existing regulations. His legislation “would require federal agencies to identify and eliminate one existing regulation for each new regulation they want to add.” Under his “regulatory pay-go system,” regulatory agencies, with oversight from the Office of Management and Budget (OMB) and either the Congressional Budget Office (CBO) or the Government Accountability Office (GAO), would catalogue existing regulations and develop estimates of their economic impacts. Then, before issuing a new regulation, agencies would be required to eliminate one outdated or duplicative regulation of the same approximate economic impact.

A regulatory pay-go shares similarities with a regulatory budget, a concept that attracted bipartisan interest in the 1970s and 1980s, but has not been championed in recent years. In 1980, President Carter’s Economic Report of the President discussed proposals “to develop a ‘regulatory budget,’ similar to the expenditure budget, as a framework for looking at the total financial burden imposed by regulations, for setting some limits to this burden, and for making tradeoffs within those limits.” The report noted analytical problems with developing a regulatory budget, but concluded that “tools like the regulatory budget may have to be developed” if governments are to “recognize that regulation to meet social goals competes for scarce resources with other national objectives,” and set priorities to achieve the “greatest social benefits.”

The analytical problems identified with the regulatory budget are non-trivial, and would also apply to a regulatory pay-go. Since the late 1990s, OMB has been compiling agency estimates of the costs (and benefits) of major regulations with mixed results. Estimating the opportunity costs of regulations is not as straightforward as estimating fiscal budget outlays, where past outlays are known and future outlays can generally be predicted with some accuracy. Some regulatory impacts will be harder to estimate than others. What are costs associated with homeland security measures that reduce airline travelers’ privacy? What are the costs of regulations that prevent a promising, but yet unknown, product from coming on the market? Even regulations whose costs appear to be straightforward, such as corporate average fuel economy standards that restrict the fleet of vehicles produced, depend on assumptions about consumer preferences and behaviors that may not reflect American diversity. EPA and DOT recently estimated that these rules will have large negative costs (even if benefits were zero), because, according to their calculations, the fuel savings consumers will derive from driving more fuel-efficient vehicles will outweigh the increased purchase price. Would agencies that estimate negative costs associated with their rules be able to issue even more?

Despite these analytical difficulties, a regulatory pay-go has the potential to impose needed discipline on regulatory agencies and generate a constructive debate on the real impacts of regulations. By focusing on the costs of regulations and allowing agencies to set priorities and make tradeoffs among regulatory programs, it might remove some of the contentiousness surrounding benefit-cost analysis. How it would affect agencies’ incentives for estimating costs is uncertain. In developing a baseline estimate of the costs of existing regulations, they may have incentives to overstate costs, particularly for regulations they may want to trade in exchange for new initiatives. Providing an entity outside of the executive branch (CBO or GAO) the resources and mandate to evaluate and critique agency estimates of regulatory costs could be critical to a regulatory pay-go’s success. While it will never be possible to estimate the real social costs of regulations with any precision, a regulatory pay-go should provide incentives for agencies, affected parties, academics, Congressional entities and non-governmental organizations to improve upon the rigor of regulatory impact estimates.

Susan E. Dudley directs the George Washington University Regulatory Studies Center. She served as Administrator of the Office of Information and Regulatory Affairs from 2007 to 2009.