Given its recent advance notice, EPA should implement retrospective analysis of its regulations in several ways.
How well do environmental regulations perform? Are environmental standards “job-killing regulations” that merit elimination? Does environmental policy strike a balance of reducing pollution without undermining economic growth?
Thoughtful policy debates about the future of environmental regulation require rigorous, transparent evidence addressing these questions. Evaluating the performance of regulations—through retrospective analysis of their impacts, benefits, and costs—can play a critical role in informing these debates. Indeed, these analyses can illustrate whether fully-implemented regulations are making the whole of society better off.
In June, the U.S. Environmental Protection Agency published an advance notice of proposed rulemaking soliciting suggestions on how to increase “the consistency and transparency in considering costs and benefits in the rulemaking process.” To its credit, EPA raises the topic of retrospective review of regulations in this proposal. Indeed, the most significant progress EPA could make in promoting regulatory transparency lies in undertaking analysis of and communicating to the public the resulting information about the impacts of its regulations in practice.
Regulatory policy scholars have provided guidance on how to implement and institutionalize retrospective review of regulations, yet the norm—in every administration dating back to the Carter Administration—has been to undertake short-lived, ad hoc lookbacks of existing regulations.
Today’s EPA can implement retrospective review in its rulemaking process in several ways.
As one such way, EPA should identify rules for retrospective analysis based on a value of information standard, given resource constraints. Several types of cases could arise in which the value of information from a retrospective analysis could be high. First, a regulation could be issued under an authority that requires periodically reviewing and updating the standard. This is the case with the National Ambient Air Quality Standards (NAAQS). Understanding regulatory performance could improve the design and setting of future NAAQS. In addition, EPA could leverage additional resources—such as the Clean Air Scientific Advisory Committee—in the retrospective analysis of NAAQS rules. Second, understanding the performance of a regulation in one context could yield important learning spillovers for other regulatory contexts. For example, the success of the sulfur dioxide cap-and-trade program has established the foundation for subsequent trading programs. Third, just as EPA applies more rigorous prospective analyses to economically significant rules, retrospective analysis could also target these larger rules, since learning how to improve economically “large” regulations could yield greater social benefits.
In addition, EPA could dramatically improve the quality of retrospective analysis by planning for the analysis in the design of new rules. At the proposal stage, the agency could draw up a research plan that both reflects the proposal’s design and shows how the agency can credibly execute an empirical strategy to enable the causal identification of the rule’s impacts. Even though the typical government regulator may have little experience thinking about causal identification, experts in program evaluation—at government statistics agencies as well as in academia—can aid the design of these types of research plans.
Research plans should account for the agency’s ability to employ discretionary authority to craft rules that simultaneously meet statutory objectives and enable their rigorous evaluation. For example, economists have long exploited the variation in rules’ application by season, geography, or pollutant source for their research on the impacts of environmental regulations. The research plan should explicitly describe the data that should be collected—both through new data collections and by compiling data from existing sources such as the U.S. Energy Information Administration and the U.S. Census Bureau—necessary to execute the empirical strategy. This may require a more expansive approach to data collection, such as acquiring information from non-regulated entities—like “control groups” in the context of quasi-experimental empirical methods—that may play a critical role in estimating the rules’ impacts.
EPA should also establish a clear framework and guidelines for how it prioritizes and implements retrospective analysis. This framework could be elaborated in a guidance document akin to the agency’s high-quality guidelines for preparing economic analyses. By incorporating research plans for retrospective analysis at the proposal stage, the agency provides the public with an opportunity for comment and feedback. In addition, the data collected and compiled by EPA should be made available for external evaluation. This can leverage the resources and expertise among academic scholars and stakeholders while facilitating transparency on regulatory performance.
Finally, an effective approach to regulatory evaluation will require institutionalizing a “culture of retrospective analysis.” Better alignment of incentives could help promote retrospective analysis. Planning for a retrospective evaluation at the rule development stage both lowers the resource costs of future analyses and increases the political costs of failing to implement the research plan. Integrating the research plan in the rule’s preamble and conditioning a sunset date in the rule on the implementation of a retrospective analysis could likewise orient political and bureaucratic incentives toward regulatory evaluation. If understanding the impacts of regulations in practice is a priority for the agency leadership—as well as for the White House and the U.S. Congress—then agencies should appropriate additional resources for retrospective analysis.
The June EPA proposal also raises questions about the role of co-benefits—such as reducing premature mortality related to fine particulate matter as a by-product of regulating power plant carbon dioxide emissions—in regulatory evaluation. President Donald J. Trump’s White House, through its annual report on the benefits and costs of regulation to Congress, stated that “the consideration of co-benefits, including the co-benefits associated with reduction of particulate matter, is consistent with standard accounting practices and has long been required under” guidance from the Office of Management and Budget. In the context of retrospective review, the agency should undertake analyses after rules have been finalized to assess co-benefits as well as the benefits of directly targeted pollutants. Regulatory evaluations should fully account for co-benefits because co-benefits help illustrate whether the American people are better off under the envisioned regulation.
A well-designed approach to transparency can enable continuing improvement in the design and implementation of environmental policy. Institutionalizing retrospective analysis of environmental regulations represents the most thoughtful action this Administration can take to promote greater regulatory transparency and enhance the welfare of our society.
This essay is part of a series, entitled Consistency and Transparency in Environmental Cost-Benefit Analysis.
Professor Aldy’s research on regulatory policy has been supported by the Alfred P. Sloan Foundation, the Administrative Conference of the United States, Resources for the Future, NBER, BP, the Mossavar-Rahmani Center for Business and Government at the Harvard Kennedy School, and the Taubman Center for State and Local Government at the Harvard Kennedy School.
