The Magazine

Not Worth the Paper It’s Printed On

The folly of OMB’s annual cost-benefit report.

Aug 19, 2013, Vol. 18, No. 46 • By IKE BRANNON and SAM BATKINS
Widget tooltip
Audio version Single Page Print Larger Text Smaller Text Alerts

Every spring the Office of Management and Budget releases the president’s proposed budget for the upcoming fiscal year. While Congress invites senior administration figures to testify before various committees, and the media pore through the document to elucidate the administration’s priorities, by the end of a week everyone agrees that most of what’s in the budget has little chance of becoming enacted. Afterwards, Congress goes through the motions of passing a budget of its own, with scant regard to what the White House has proposed.

OMB director Sylvia Burwell

OMB director Sylvia Burwell


At the same time it releases the budget, OMB produces another document, one that’s a manifestation of the management half of its portfolio: The Report to Congress on the Benefits and Costs of Federal Regulations. While it garners much less attention than the president’s budget, the cost-benefit report deserves much closer scrutiny—because it perennially misrepresents the activities of the regulatory agencies as being above reproach.

Since Ronald Reagan was in office all presidents have had an executive order in place mandating that “economically significant” regulations issued by executive agencies (i.e., those with an economic impact in excess of $100 million) be subject to a formal cost-benefit analysis showing that the benefits outweigh the costs.

The OMB report compiling these costs and benefits—which was begun by the previous administration—is in essence a league table that compiles the costs and benefits of a variety of regulations formally issued during the previous year. Every edition has reported that all is well with our regulatory world, with the benefits accruing to society from our government’s regulatory activities far outweighing the costs imposed.

The problem is that the report is of little use in discerning whether this is, in fact, the truth. First, and most important, the report includes only a few regulations. While 3,700 regulations were issued in fiscal year 2012, with 80 of them being categorized as “major,” only 14 regulations were included in OMB’s analysis.

It is, of course, the case that the costs and benefits of some regulations (think of the one issued in late 2001 to strengthen the doors to airline cockpits) can’t be quantified. Also, as it currently stands, some entities that issue regulations don’t bother, and are not required, to do cost-benefit analyses, and most regulations considered to be “non-major” never have their costs and benefits measured. Still, OMB is cherry-picking less than 1 percent of regulations issued in 2012 to present in its report.

Another problem is that the various executive branch agencies aren’t all that good at measuring costs and benefits. When EPA tasks its economists to do a cost-benefit analysis on a regulation, it expects them to deliver an analysis that supports the regulation, and their economists have every incentive in the world to deliver the desired answer. The idea that an agency’s economists would get in the way of its regulators is laughable: Any economist worth his salt can torture data until they justify whatever the agency wants to do.

One example: An important part of measuring the benefits of many regulations is the value of a life saved. If a new regulation were to save 10 lives a year, how should we value this gain? Economists have come up with several different ways to assess the value that people implicitly place on their lives: For instance, there are numerous studies that estimate the value of a “statistical life” by observing how much money people demand to take a slightly more dangerous job, or how much they are willing to pay for a safer bike helmet, or on safety accessories for a car.

About a decade ago two enterprising economists, Janusz Mrozek and Karen Taylor, compiled a comprehensive list of studies that tried to measure the value of a statistical life (VSL) and did a meta-analysis to see if they could reach a broad conclusion on the topic. The representative number they arrived at was $2.5 million, which happened to be much lower than what EPA used in its cost-benefit calculations. Adopting the lower figure would dramatically reduce the benefits for most EPA regulations, which would mean that fewer potential regulations could pass a cost-benefit test.

Recent Blog Posts