In a rare act of regulatory self-restraint, the Environmental Protection Agency recently issued a proposed rule that would limit the flexibility the agency has shown when conducting benefit–cost analysis of proposed rules under the Clean Air Act. Among other provisions, the rule would formalize the analysis process, following the Office of Management and Budget’s Circular A‑4 and the EPA’s own Guidelines for Preparing Economic Analyses. It would also prohibit the inclusion of co-benefits when determining a rule’s net benefits, thereby avoiding the double-counting problem that infamously influenced the Obama administration’s Mercury and Air Toxics Standards.
This regulatory reform would force the EPA to adhere to a set of standards that — it is hoped — would not change much across administrations. The intention is to depoliticize the agency’s actions and focus its efforts on promulgating rules that indisputably benefit the country.
Preemptive caving / The mechanism within the federal government that produces and analyzes benefit–cost analyses — namely, the executive-branch agencies themselves and the Office of Information and Regulatory Affairs (OIRA) — can work only if all of the entities involved are guided by the same set of rules and are operating with a clear and consistent understanding of the process. Unfortunately, this is not always the case.
For instance, while the executive order mandates that agencies provide to OIRA a benefit–cost analysis for any proposed rule with an economic effect that exceeds $100 million, agencies tend to discover numerous proposed rules that appear to have an effect just below that threshold, making them immune to OIRA review. Research by Sam Batkins has found considerable evidence of this behavior.
Another problem is that the agencies typically perform the benefit–cost analysis on their own proposed rules, which creates a form of moral hazard. Agency staffers have a vested interest in their proposed rule passing muster, as their careers are advanced by achieving greater regulatory oversight. That puts the staff economists tasked with performing the benefit–cost analysis in a difficult position. The implicit understanding is that their analysis should validate the rule regardless of the underlying reality.
This moral hazard means that the quality of benefit–cost analysis can vary greatly across administrations. Members of a pro-regulatory administration will often work with the EPA to effectively let it know that they will pressure OIRA to be less critical of any benefit–cost analysis, which invariably leads to less rigorous work. The phrase OIRA staffers like to use to describe this strategy is “preemptive caving.”
Less-than-rigorous benefit–cost analysis can occur in administrations distracted by other exigencies as well, I have observed. In the aftermath of 9/11, OIRA — where I worked at the time — received an EPA analysis that had the first part of its printed title, “ENGINEERING ANALYSIS,” crossed out and “COST-BENEFIT” scribbled in pen above it. Needless to say, my colleagues and I found it to be of marginal value, but we were unable to delay the rule to conduct a more thorough analysis.
Reform ideas / Arguably the most important environmental legislation passed in the last three decades is the 1990 Amendments to the Clean Air Act. One reason for its effectiveness is that Congress worked for years in a bipartisan fashion to reach a broad consensus on its content, and that has resulted in a vastly improved environment. The most important rules that came from the legislation — and that the act clearly delineated — ended up with benefits that substantially outweighed the compliance and enforcement costs.
However, in recent years there have been rules engendered by the act that may not have been cost-effective. The act has been stretched to cover climate change as well as other perceived environmental exigencies that its authors did not conceive of when it was created three decades ago. We should have a process in place that ensures critical scrutiny of all regulatory proposals before they are enacted, and I fear that the current system does not achieve that.
One solution to this lacuna that would lessen the moral hazard and improve the analysis of regulatory proposals would take the responsibility for benefit–cost analysis from the agencies and give it to a new entity altogether. That would (partially) insulate analysis from the systemic pressures that exist when doing such tasks within the agency itself.
Such a reform has thus far failed to receive much support. Regulatory agencies will never be keen to surrender the modicum of leverage and budget that such a change would entail. Congress has failed to evince much enthusiasm for such a change as well.
Congress does have the ability to withdraw rules it does not like, using the Congressional Review Act. But that requires a majority in both the House and Senate, and is subject to veto by the president. And it applies only to rules published within 60 days of Congress adjourning sine die, so it is a very imperfect instrument.
However, the EPA’s proposal to standardize its benefit–cost analysis would constitute an acceptable compromise that achieves part of what would be gained from separating the analysis from the rulemaking. It would also strengthen the quality of these analyses in both Republican and Democratic administrations.
While it is a Republican administration that is enacting this rule (albeit a bit late in its tenure), it is worth noting that Cass Sunstein, a former OIRA administrator for the Obama administration and someone who has published a great deal of research on benefit–cost analysis, has written in support of the idea. According to Sunstein, EPA Administrator Andrew Wheeler’s 2019 guidance on the rule constitutes “an important memorandum that makes perfect sense.”
