Cost-Benefit Analysis: The Stunning Triumph of a Flawed Tool
by Katherine McFate, 9/18/2012
UPDATE (9/21/12): On Wednesday, Jim Harper at the Cato Institute published a response to this post with a title that completely misrepresented our position on regulations.
Of course we support standards and safeguards that have demonstrable benefits; in fact, we believe there are some safeguards – like improved airline safety rules – that are important enough that they should be put in place even if they don’t conform to a narrow, formulaic cost-benefit analysis.
Public policy choices are about values. And many of the core values that Americans hold dear – safety, security, freedom, fairness – are not easily quantified and monetized. The tradeoffs and choices we make should be the result of democratic debate, not hinge on the assumptions built into mathematical models. Anyone can generate cost-benefit numbers, but an over-reliance on econometrics in public policy means we "know the price of everything and value of nothing."
* * *
Last Thursday, Cass Sunstein, the former administrator of the Office of Information and Regulatory Affairs (OIRA), argued that “cost-benefit analysis has become part of the informal constitution of the U.S. regulatory state” and that this represents a “stunning triumph.” While it’s true that cost-benefit analyses are being applied to rulemaking across an array of laws and programs, we believe that this represents the triumph of a flawed analytic tool and is not a triumph for American citizens. It is simply not appropriate to apply cost-benefit analysis to many aspects of policymaking, and the results from such analyses should not be the final determinant of the value of many proposed standards or safeguards.
Many regulatory experts and members of the public interest community believe cost-benefit analysis exaggerates the costs of new rules and underestimates their social benefits. Cost-benefit analyses are heavily dependent on the assumptions built into the quantitative models used and the data on which the models are applied. The data is provided by regulated businesses. If these businesses are resisting the need to change their production processes or business practices to comply with a new standard or regulation, they will tend to overestimate the compliance costs of the rule. But once a new rule goes into effect, American businesses quickly find the cheapest and most efficient way to comply, so costs tends to be significantly less than advertised.
Benefits, while easy to describe (safer toys, safer food, less risk of cancer, better health), are often difficult and expensive to quantify and to monetize. Take the case of lead paint. It is difficult to estimate the exact reduction in a child’s IQ that results from continuous exposure to lead paint. (Individuals have different susceptibilities; buildings have different concentrations of lead.) But assume a defensible “average drop in IQ by exposure level” is generated. Next, the researcher would have to estimate the number of children who experienced particular exposure levels over a certain period of time. Then to monetize the costs, the researcher would have to estimate the costs of remedial education and/or the drop in earnings experienced over a lifetime due to the lower IQ level. A huge amount of data and work would be involved to generate a dollar figure to justify a rule to prevent young children from being exposed to a toxic substance. Once science has demonstrated that lead paint damages children’s brains, is there a question of what the public policy response should be?
There are times when a cost benefit-analysis is not only inappropriate, but can be offensive. A stark illustration: the Department of Justice (DOJ) planned to conduct a cost-benefit analysis on a rule designed to reduce the incidence of prison rape. Professor Lisa Heinzerling criticized this effort as “bizarre and unfortunate.”
Requiring cost-benefit analyses before rules can be changed can produce undesirable and counterproductive results. Take, for instance, airline safety standards. The Federal Aviation Administration (FAA) has been remarkably successful at protecting the flying public over the past decade. But under a purely economic cost-benefit analysis requirement imposed on the agency, too few people have died to justify further safety improvements – even if updated standards would save more lives.
Previous Congresses understood that the standards, rules, and safeguards we establish define who we are – or aspire to be – as a nation and should not be boiled down to a dollars-and-cents calculation. As a result, many federal environmental and worker safety laws actually prohibit the use of cost-benefit analysis in setting public protection standards.
Americans have a right to safe drinking water, clean air, safe workplaces, safe food, safe drugs, and safe toys. They expect their government to ensure these basic protections. End of story. The value of maintaining and improving the health and safety of the American people? Priceless.
Image courtesy of Harvard Law Record, used under a Creative Commons license