Does Cost-Benefit Analysis Matter?

By Cindy Skrzycki
Tuesday, June 12, 2007

The practice of estimating the costs and benefits of U.S. government regulations is "frequently done poorly," with scant evidence that it makes a difference on policymaking.

That's the dismal finding of a recent report concluding that economic analysis had little substantial impact on decisions made by regulators. The assessment is important because the White House budget office estimates that major rules can cost industry $40 billion annually, while benefits to society from regulation can range from $94 billion to $449 billion a year.

The conclusion by the AEI-Brookings Joint Center for Regulatory Studies underscores the continued disagreement over the relative value of regulation and the role that cost-benefit analysis should play in making politically sensitive decisions, such as writing clean-air rules.

"Given the forces of law and politics, one should not expect the role of economic analysis to be any more than modest," said John D. Graham, a proponent of cost-benefit analysis who headed the Bush administration's regulatory review office until last year.

Economic analysis has been a key tool in making regulatory policy since 1981, when Ronald Reagan became president, with cost-benefit reviews increasingly being applied to initiatives.

The idea is to assign monetary values to regulation, such as installing pollution-control technology or overhauling assembly lines to avoid repetitive-motion injuries to workers. That number is then weighed against benefits, such as how many illnesses could be avoided or lives saved.

Although the report's general assessment is gloomy, it suggests that putting rules under an economic microscope is worthwhile since analysis can save billions of dollars at the margins.

"It certainly does matter in selected instances,'' said Robert W. Hahn, a scholar at the American Enterprise Institute and co-author of the report with Paul C. Tetlock, an assistant professor at the University of Texas at Austin. "Our findings suggest that government regulatory analyses are a lot like chicken soup: They typically do no harm, and in some instances, they help a lot."

Hahn is executive director of the Joint Center, which published their study. The center is a collaborative effort between the American Enterprise Institute and the Brookings Institution, two District-based research groups.

The report cited the lead-in-gasoline rule as an effective use of cost-benefit analysis. The calculations forced the Reagan administration to adopt a stringent rule that ordered about 90 percent of the lead still in gas to be removed by 1988.

In that case, cost-benefit analysis showed potential improvements in health, including lowering blood pressure, to be 10 times the cost to refiners from 1985 to 1988. The numbers -- $20 billion in benefits and $2 billion in costs -- were too lopsided to argue with.

The recent report has special relevance for the Bush administration, which has been an enthusiastic practitioner of cost-benefit analysis. President Bush's budget office has instructed federal agencies in the finer points of the methodology and appointed to the regulatory-review office two ardent believers in economic and scientific assessment of rules -- Graham and Susan E. Dudley, who is from a research group at George Mason University in Arlington.

Other cases show that the cost-benefit equation can result in the right decision being made for the wrong reasons, according to Frank Ackerman, director of the research and policy program at the Global Development and Environment Institute at Tufts University.

Ackerman cites a 1967 Rand Corp. analysis advising that dams not be built on the Colorado River just above and below the Grand Canyon, not because they might ruin its natural beauty but because nuclear power at the time was cheaper. Environmentalists also helped kill the project.

Outcomes like this are used to argue against using such a tool as a deciding factor in rulemakings intended to protect safety, health and the environment.

Critics also point to how economic analysis can be twisted to reach a certain conclusion. In 2001, when the Bush administration eliminated a rule from President Bill Clinton's administration to stem ergonomic injuries in the workplace, the broad reach and costliness of the regulation were cited as the reasons. What was overlooked was that the economic analysis found the $4 billion in cost was outweighed by $9 billion in benefits.

Hahn doesn't get involved in the moral debate over cost-benefit analysis. He cites other reasons why the approach often doesn't have a significant impact on policy -- politics and the quality of the reports themselves.

"There is no reason to think you are getting the straight poop from a government analysis," he said. He cited research that showed 74 rules issued across the Reagan, George H.W. Bush and Clinton administrations "do not provide some very basic economic information," including "net benefits and policy alternatives."

The report may be a cautionary note to policymakers about depending on economic analysis too much.

Charles L. Schultze, who was head of the Council of Economic Advisers in the Carter administration, said that over the past 30 years, economic analysis has made the debate over rulemaking more sophisticated.

"You never hit home runs or good clean singles," said Schultze, now a senior fellow emeritus at Brookings. "But you can narrow the range of what's acceptable and screen out a lot of junk."

Cindy Skrzycki is a regulatory columnist for Bloomberg News. She can be reached atcskrzycki@bloomberg.net.