Controversies over Mercury Control Benefits

EPA came out last week with it’s rule for regulating mercury emissions from power plants. I haven’t read it yet – the draft rule plus preamble is 500 pages, double-spaced. But the WaPo noted today that the EPA was arguing that the costs of mercury controls outweighed the health benefits – at the same time an EPA-sponsored study conducted by Harvard University was coming to a different conclusion. According to the WaPo:

That analysis estimated health benefits 100 times as great as the EPA did, but top agency officials ordered the finding stripped from public documents, said a staff member who helped develop the rule. Acknowledging the Harvard study would have forced the agency to consider more stringent controls, said environmentalists and the study's author.

An EPA official said that the study was submitted to late to be considered in the rulemaking. But again according to the WaPo:

Interviews and documents, however, show that the EPA received the study results by the Jan. 3 deadline, and that officials had been briefed about its methodology as early as last August. EPA officials referred to some aspects of the Harvard study in a briefing for The Washington Post on Feb. 2.

The Harvard study concluded that mercury controls similar to those the EPA proposed could save nearly $5 billion a year through reduced neurological and cardiac harm. Last Tuesday, however, officials said the health benefits were worth no more than $50 million a year while the cost to industry would be $750 million a year.

It sounds like EPA didn’t even put the study into the docket (OAR-2002-0056). According to the WaPo, the study was sponsored by the Northeast States for Coordinated Air Use Management (NESCAUM), which administered the EPA grant. You can find the study on their web site.

Before I had heard about this issue, I had started a post about this study by Trasande et al., 2005, coming out soon in Environmental Health Perspectives. That analysis concluded the lost productivity due to neurodevelopmental impacts of methylmercury amount to $8.7 billion per year in the US. This estimate is based on the number of children born each year with methylmercury cord blood levels greater than the RfD of 5.8 μg/L. As discussed previously here, the RfD has been developed from an estimated methylmercury dose that doubles the prevalence of young children with scores on a test of intellectual development that would fall into the clinically subnormal range – shorter version: low-level methylmercury exposure is thought to reduce IQ. Of the total $8.7 billion per year in lost productivity, $1.3 billion each year is considered attributable to mercury emissions from American power plants (you can disagree with it, but it’s a pretty slick analysis – you should check it out).

So what’s on the cost side for mercury control? According to the WaPo, EPA says “$750 million per year” (maybe it’s in the preamble somewhere). The notice of data availability provides some economic modeling which indicates $1.6 billion/year in 2010 and $1.1 billion/year in 2020. Utility industry calculations show annual costs up to 10-fold higher (of course). But before you conclude that the benefits still outweigh the costs, note that the Trasande et al., 2005 paper didn’t include the costs from cardiovascular risks (too uncertain in the minds of the authors for calculation), or other societal costs, so there’s always the possibility that the productivity losses have been underestimated.

This isn’t the first time we’ve heard about lack of transparency (sounds so much more polite than “distortion” or “concealment”) in information regarding mercury rulemaking. And, mercury allies in Congress seem pretty frantic to stop the even minimal regulation that the Bush Administration is imposing on the utility industry, going to the lengths of manufacturing bogus science in an effort to cloud peoples minds. I wonder what the deal is here. Has anyone summarized what is the financial impact of mercury controls on the utility industry? Is it really that significant?