President Obama's recent Executive Order regarding cost-benefit analysis and administrative procedure has drawn criticism both for what it does and for what it does not do. The Order provides little new guidance on how administrative agencies and the Office of Information and Regulatory Affairs (OIRA) should conduct cost-benefit analysis, and perhaps unavoidably it leaves many pertinent questions unanswered. But it does issue one significant directive: it requires agencies to formulate plans to perform "retrospective analysis" of existing significant regulations.
The stated purpose of these retrospective analyses is to determine when rules have become outmoded or ineffective. For this reason, pro-regulatory advocates have criticized the Obama administration for adopting an aggressively anti-regulatory stance and attacked the Executive Order as a concession to the business community. It is too early to know whether those criticisms are justified. But it is worth noting that the new directive offers OIRA the opportunity to do something it should have done decades ago: conduct follow-up studies of the accuracy of cost-benefit analysis.
Agencies have been conducting cost-benefit analyses at OIRA's direction since the beginning of the Reagan administration, and countless important regulatory decisions have been made on the basis of these studies. Yet across five presidential administrations, OIRA has never revisited these cost-benefit analyses. Agencies have little idea whether the projected costs and benefits actually materialized. (Several academics have studied the accuracy of cost-benefit projections, but agencies have never incorporated the results into their own analyses.) Retrospective study of past regulations (and the cost-benefit analyses) that accompanied them could provide extremely valuable information to agencies regarding when cost-benefit analysis errs and how best to correct it. (Eric Posner and I have a short discussion of this issue at page 47 of this paper.)
Importantly, this information could push agencies in pro-regulatory directions just as easily as anti-regulatory ones. It may be that agencies have consistently undercounted benefits, as many pro-regulatory scholars believe. It is also possible that they have consistently overestimated the costs of regulation to industry, and that firms are able to adapt to regulation at much lower cost than they claim. Advocates of regulation frequently make this argument as well. A great deal rests on agencies' selection of which regulations to review and on how agencies make use of the data they collect. If agencies approach the issue narrowly and look only for regulations to repeal, these gains will not materialize. But the fact remains that President Obama's Executive Order offers the opportunity to begin correcting the cost-benefit errors of the past, errors that may have been made in both pro- and anti-regulatory directions.