Aug 22, 2018

More Information Needed on the Benefits and Costs of Regulations

James Broughel Senior Research Fellow , Richard Williams Senior Affiliated Scholar

Each year, the Office of Information and Regulatory Affairs (OIRA) produces a report on the benefits and costs of federal regulations, summarizing information pulled from the regulatory impact analyses (RIAs) produced by federal agencies. The OIRA report and the underlying agency RIAs together provide an estimate of the effects that some regulations might have on the economy once they go into effect.

OIRA’s most recent draft report for 2017 estimated that major regulations agencies evaluated over the previous ten years were projected to produce benefits ranging between $287 and $911 billion (2015$), at a cost of between $78 and $115 billion.

At first glance, it might appear the regulatory system is working well for the American public by creating rules whose benefits far outweigh their costs. But these numbers are misleading.

As required by an executive order issued by President Bill Clinton, agencies must present an assessment of the potential benefits and costs for all regulations that are deemed to be “significant” by the Administrator of OIRA. There were 2,670 significant rules reviewed by OIRA between fiscal years 2007 and 2016. Among those rules, OIRA presents dollar estimates of benefits and costs for only a small fraction of the regulations the agency reviewed.

Of 36,255 final rules published between 2007 and 2016, only 137 rules had estimates of monetized benefits and costs in OIRA’s draft report. This represents about four tenths of one percent of all final regulations, an abysmal record. Furthermore, OIRA’s report includes no rules from independent regulatory agencies that have dollar estimates for both benefits and costs.Vast Majority of Regulations in 2017 Report Missing Monetized Estimates of Benefits or Costs Source: 2017 Draft Report to Congress on the Benefits and Costs of Federal Regulations and Agency Compliance with UMRAShare this

Even though many of the regulations issued by agencies are not “significant” in nature, the aggregate effects of thousands of “nonsignificant” regulations implemented year after year can be substantial, perhaps even greater than the cumulative impact of so-called “major” regulations.

Even for the 137 regulations with dollar estimates of both benefits and costs, the numbers provide little useful information. The headline numbers presented by OIRA understate the vast uncertainty surrounding these values. For example, benefit estimates rely on critical assumptions such as the monetary value of life, as well as key modelling assumptions about the effects of regulations on public health.  Risk estimates underlying benefits analyses, in particular, are notoriously uncertain.  If any of these assumptions turn out to be wrong, the costs of these regulations could easily dwarf the benefits.

OIRA deserves credit for identifying some of these key assumptions in its report, but given that most readers will focus on the aggregate benefit and cost numbers, without delving into the minutiae of the report, the annual report may end up misinforming more than it informs. In fact, this report, and the agency analyses the report is based on, are frequently invoked to declare that the administrative state is “working.”

OIRA’s benefit and cost figures provide an incomplete and misleading representation of the federal administrative state. It may be time to consider whether OIRA’s annual report on regulations itself passes a benefit-cost test.

Author’s note: There are several discrepancies in the OMB report related to the number of rules with monetized cost and benefit numbers over the last decade. Table 1-1 in the report summarizes information from 137 regulations with such cost and benefit estimates, presented by agency. Table 1-4 purportedly lists the same information by year, rather than by agency, and includes 140 regulations. Meanwhile, Appendix A-2 presents 125 individual regulations with monetized cost and benefit estimates from the last decade. It is unclear why these totals differ from one another. For the purposes of this chart, we used the data that appeared first in the report, from Table 1-1.

Photo credit: Wikimedia Commons, Almonroth

Support Mercatus

Your support allows us to continue bridging the gap between academic ideas and real-world policy solutions.Donate