by Wayne Crews

In the 2014 fiscal budget proposal, the White House praised regulation of auto safety, energy efficiency and credit cards, and claimed, “In fact, the net benefits of regulations issued through the third fiscal year of the first term have exceeded $91 billion.”

Well, let’s look at that.

Rules officially reviewed by OMB represent a small proportion of the total since before the turn of the century.

In the regulatory at year-end 2012, according to the Regulatory Plan and the Unified Agenda of Federal Regulations, 63 federal departments, agencies and commissions had just completed or were at work on 4,062 rules and regulations at various stages of planning and implementation (pre-rule, proposed, final, completed).

Of these, 224 were designated economically significant (loosely, $100 million-plus in annual impact). By calendar year end 2012, 3,708 rules were finalized.

Today’s official narrative maintains that “major” and “economically significant” rules account for the bulk of regulatory costs. OMB intones that “the benefits and costs of major rules, which have the largest economic effects, account for the majority of the total benefits and costs of all rules subject to OMB review.”

“Subject to OMB review” is a necessary qualifier — with its own “economically significant” implications.

The non-reviewed status of most rules small and large, such as controversial independent agency rules like the Federal Communications Commission’s net neutrality order, casts doubt on intimations that the annual Benefits and Costs report is inclusive and comprehensive.


OMB’s cost-benefit breakdowns incorporate only benefits and costs of “major” executive agency (not independent agency) rules that agencies or OMB have expressed in quantitative, monetary terms, omitting numerous categories and cost levels of rules altogether.

A few years back, OMB more explicitly recognized that costs could significantly exceed those made explicit in its annual reports, a stance far better situated for informing regulatory oversight.

For example, in 2002 OMB allowed that,

Based on the information released in previous reports, the total costs and benefits of all Federal rules now in effect (major and non-major, including those adopted more than 10 years ago) could easily be a factor of ten or more larger than the sum of the costs and benefits reported.

The “factor of ten” or more was repeated subsequently, for example in a February 3, 2003 OMB memo, “With New Information on Federal Regulations, OMB Releases Draft Cost-Benefit Report” as well as in the 2004 final report (but with ten year look-backs rather than six year).

The explicit emphasis that costs could be substantial was downplayed in subsequent reports, replaced by insistence that “reported monetized benefits continue to be significantly higher than the monetized costs.”

Note again the need for qualifiers, here “reported” and “monetized.”

For fiscal year 2012 (October 1, 2011– September 30, 2012), OMB reviewed 47 executive agency major rules with some combination of cost and benefit estimates (in contrast to the 3,708 rules great and small finalized in calendar year 2012).

Of these 47 rules, only 14 had cost and benefit calculations expressed in dollar terms. This is the block that OMB features prominently in its annual expositions of net benefits of the regulatory enterprise, and the one that gets reported publicly and used by regulatory advocates.

OMB estimates that this subset of rules sports anticipated costs of between $14.8 billion to $19.5 billion (in 2001 dollars). Twenty-two others implemented budgetary spending programs; such “budget rules” are officially considered transfers rather than regulations. Of the remaining 11 rules, agencies provided partial benefit and/or cost estimates.

It’s been noted that OMB’s cost-benefit roundup leaves out independent agencies’ compliance costs, and all rules costing under $100 million (not that anyone would know their cost). As expressed over a decade ago in its 2002 Report to Congress, “OMB does not review regulations of the independent agencies or any regulations that are not determined to be ‘significant’ under the E.O. 12866 definition.”

OMB has recently acknowledged the non-triviality of omissions. Here’s OMB’s 2011 disclaimer:

It is important to emphasize that the figures here have significant limitations. When agencies subject to Executive Orders 13563 and 12866 have not quantified or monetized the benefits or costs of regulations, or have not quantified or monetized important variables, it is because of an absence of relevant information. Many rules have benefits or costs that cannot be quantified or monetized in light of existing information, and the aggregate estimates presented here do not capture those non-monetized benefits and costs. In fulfilling their statutory mandates, agencies must often act in the face of substantial uncertainty about the likely consequences. In some cases, quantification of various effects is highly speculative. …In addition, and significantly, prospective estimates may contain erroneous assumptions, producing inaccurate predictions… While the estimates in this Report provide valuable information about the effects of regulations, they should not be taken to be either precise or complete.

I don’t know why the White House budget didn’t cite that instead of the $91 billion number.

The upshot is that major, significant or economically significant rules with cost estimates are a fraction of the total number of rules in the pipeline or finalized, and the number with cost and benefit analysis an even smaller proportion despite repeated annual assertions that “reported monetized benefits continue to be significantly higher than the monetized costs.”

A major problem with the annual roundup is the inherent randomness; had a different set of rules been “monetized,” it would have produced an entirely different net benefit range. Both benefit calculations and cost calculations for the regulatory enterprise rarely exist.

The reported ranges of net benefits we get annually, while widely quoted, are essentially meaningless as an annual assessment or for making inferences with year to year comparisons.