In our study, we estimated the costs and benefits of the U.S. Environmental Protection Agency’s (EPA) power plant mercury regulation. To estimate the benefits of mercury reduction, we considered each link in the pathway, including the reduction of emissions from U.S. power plants; reductions in mercury deposition; reductions of methylmercury in U.S. freshwater and marine fish; reductions of methylmercury consumption from U.S. fish by U.S. residents; reductions of methylmercury in U.S. women of childbearing age; and IQ improvements in U.S. children. For each link, we used the best available evidence and, if anything, tended to err on the side of overstating benefits. Only at the end did we monetize estimates of IQ improvements, based on a study of parental willingness to pay for IQ increases through chelation.

Zeller and Booth contend that our estimate of the benefits of mercury reduction is “grossly understated” based on their claim that our estimate of the value of an IQ point is flawed. They cite a study by Trasande et al. (1) claiming that benefits of mercury reduction are $1.3 billion per year. Unfortunately, they are comparing apples with oranges. The $1.3 billion estimate (1) is for the benefits of eliminating all U.S. power plant mercury emissions. Zeller and Booth apply this annual measure of complete elimination of power plant mercury emissions to each year from 2005 to 2020. It is incorrect to compare the costs of EPA’s regulation that eliminates a fraction of the power plant emissions to the benefits of eliminating all power plant emissions of mercury (which would cost considerably more to achieve).

Zeller and Booth suggest that the monetized benefits we use for IQ may be understated. We agree that the willingness-topay numbers for IQ may understate the benefits of IQ. The value of an IQ point suggested by Trasande et al. (1) is about an order of magnitude greater than our estimate. However, as we noted in our Letter, using their estimate does not change our finding that the costs of the regulation are likely to exceed benefits.