Regulatory Reform: Build on What Works

Regulatory rollbacks are all the rage in Washington, from President Trump’s executive orders on regulatory reform to congressional resolutions disapproving major regulations issued by the Obama administration. But whether you think we have too much or too little regulation, reasonable people ought to be able to agree that we should have the right kind.

Unfortunately, it’s not at all clear that we do. One of us (Ellig) authored a study in 2016 revealing that less than half of the major non-budget executive branch regulations proposed between 2008 and 2013 were accompanied by evidence showing that the regulations addressed significant problems. And for about one-third of those regulations, regulators failed to consider the benefits or cost-effectiveness of alternative approaches.

When regulators do not understand the problem they are trying to solve or the merits of alternative solutions, they are more likely to issue regulations that are ineffective, unnecessary, or overly broad. In a recent study published in the Journal of Benefit-Cost Analysis, we identified three ways to increase the likelihood that federal agencies will understand what problems they’re trying to solve and consider the pros and cons of alternative solutions: (1) require agencies to assess the relevant problems and possible alternatives before proposing any regulations; (2) require agencies to consult with affected parties and experts before writing regulations; and (3) give the Office of Information and Regulatory Affairs (OIRA), which reviews regulations before they are proposed, additional resources.

Agencies tend to produce more thorough analyses of regulations when they seek public comments on prior proposals, publish preliminary analyses, or ask the public for data before proposing new regulations. For example, public comments provided in response to a request for information regarding the Family Medical Leave Act (FMLA) helped the Department of Labor identify vague definitions that were the source of confusion among affected parties — such as whether an employee with a severe cold or flu was entitled to medical leave.

These results suggest that the first reform — having Congress require agencies first to assess the problem and consider alternative solutions — will improve regulation. It would also help combat agencies’ documented tendency to treat their analysis as supporting arguments for regulatory actions they have already decided to take, rather than as information to help them decide whether or how to regulate. If Congress wants agencies to issue economically sound rules, regulatory reform should discourage this practice.

In some cases, there is already legislation requiring agencies to consult with state governments, small businesses, or expert advisory committees before proposing regulations. Our second proposed reform would standardize this practice with the aim of producing more thorough analyses.

This is what occurred, for instance, when the Department of Transportation (DOT) developed four stakeholder groups to assess alternative ways of improving safety for gas pipelines. The official analysis accompanying the regulation issued by DOT included significant portions of the assessment of alternatives performed by the stakeholder groups.

Our research also found that regulatory analysis tends to be more thorough when it has undergone a longer review by OIRA. Moreover, agency analysis tends to be more thorough when OIRA review is concluded under a presidentially appointed OIRA administrator, who has more clout than an acting administrator. But OIRA’s staff and budget have shrunk as its workload has increased, making such reviews more difficult to carry out effectively.

More specifically, the number of OIRA full-time equivalent employees has fallen by more than half since it was created, from 97 in 1981 to 47 in 2016. Yet the number of major regulatory actions reviewed by OIRA has more than doubled from 262 in its first four years to 588 between 2012 and 2016. And since 1981, OIRA’s budget has fallen by 43 percent in real terms. This is why a bipartisan group of former OIRA administrators recently recommended that OIRA’s budget be restored to a level that will allow it to handle all of its responsibilities effectively.

There is no panacea for all that ails our regulatory system. But these three reforms would increase the odds that federal agencies will rely on evidence, rather than merely good intentions, when adopting regulations. This, in turn, makes it more likely that regulations will end up solving real problems at reasonable costs.

These reforms have enjoyed bipartisan support in the past. Washington’s new-found fervor for regulatory reform may finally provide the impetus needed to get them enacted.