By Catherine A. O'Neill and Amy Sinden

Cass Sunstein, President Obama's nominee to serve as the federal government's so-called regulatory czar, is scheduled for a confirmation hearing today. While on the hot seat, Sunstein can expect to face questions about his vision for the Office of Information and Regulatory Affairs, which he would lead - in particular, whether he'll continue the increasingly controversial use of cost-benefit analysis to evaluate proposed regulations.

Beginning in the Reagan administration, any regulation with a significant impact has had to pass through Information and Regulatory Affairs' doors for approval. The office's role, frankly, has been to water down health, safety, and environmental regulations - if not drown them entirely.

One way it accomplishes that is by subjecting rules to cost-benefit analysis. This process reduces the benefits and costs of a proposed regulation to monetary terms - no matter how implausible the assigning of dollar values may be - and then weighs them against each other. If the calculated benefits don't outweigh the calculated costs, the office typically weakens or scuttles the rule.

Take mercury regulation, for example. Mercury is a potent neurotoxin that is particularly hazardous to the fragile neurological systems of developing fetuses and young children. When emitted by coal-fired power plants and other sources, mercury can end up in the fish we eat.

In 1990, Congress directed the Environmental Protection Agency to clamp down on mercury emissions. By the end of the Clinton administration, EPA appeared poised to require power plants to reduce their mercury emissions by 90 percent by 2007. But the Bush administration reversed course, offering up a "Clean Air Mercury Rule" that would allow plants to wait until at least 2020 to reduce emissions by just 70 percent.

Pursuant to the regulatory office's requirements, EPA economists hammered out a cost-benefit analysis for the proposed rule to justify the change. That required some fancy math.

The office's staff began by deciding that the chief benefit of mercury regulation would be preventing neurological damage in utero. To translate this benefit into dollars, staffers measured the expected loss in future earnings as a result of each affected child's lower IQ.

But they also found a supposed silver lining in neurological damage: Children with lower IQs seek fewer years of education, and so save society the costs of "excess" education, as well as "opportunity costs." (Those who waste time in the classroom, after all, are missing the "opportunity" to work and earn money.) After a few more such calculations, the economists concluded that each IQ point sacrificed to mercury poisoning was worth $8,807.

How, um, rigorous. Notably absent from this equation was the point that nobody has a right to inflict such harm on our children, no matter what government economists think about it.

Fortunately, an appeals court struck down EPA's mercury rule last year, and the agency is back to the drawing board. (Note that, despite that defeat, industry and its allies have managed to delay meaningful regulation for years.) But the incident is vivid proof of just how vulnerable cost-benefit analysis is to gross manipulation - and how fatally flawed it is as a tool for measuring the value of a regulation.

There's more: The dirty little secret of cost-benefit analysis is that, in most cases, it's not what the law calls for. Of the 31 separate statutory provisions directing various agencies to regulate matters of health, safety, and the environment, only two call for it. A handful tolerate its use as one of several options for regulators. But 23 - the vast majority - call for some other standard, such as reducing pollution levels as much as is technologically feasible. In other words, the law doesn't force cost-benefit analysis on an administration; an administration chooses to use it, often regardless of the law.

Once confirmed, Cass Sunstein will face a choice: rely on cost-benefit analysis with the zeal his past writings suggest he would, modify the process in the hope that it can somehow be mended, or abandon it in favor of a better method. The decision he makes will have profound consequences.

Let's hope we find out what his choice will be during his confirmation hearing. And let's hope he chooses wisely.