On matters of environmental protection and regulation, free-market conservatives have two chief principles to which they claim to adhere: "sound science" and "cost-benefit analysis." As John D. Graham, cost-benefit guru and director of the Office of Management and Budget's (OMB) Office of Information and Regulatory Affairs, has put it, "The Bush administration supports federal regulations that are based on sound science and economics."

In the abstract, such a statement is difficult to argue with. When you start looking closely at the scientific and economic rationales for Bush administration actions, however, the impressive rhetoric quickly diverges from reality. As a case in point, consider the issue of mercury pollution (a timely example because this week, the Environmental Protection Agency, or EPA, will finally announce how it plans to regulate the toxic substance). In the face of concerted lobbying by their pals in the electric-utility industry, conservatives have done a miserable job of adhering to the principles of either "sound science" or "sound economics."

In a previous column, I showed how right-wingers in Congress have mercilessly distorted science in order to downplay mercury risks. In the process, they have even ignored findings by the National Academy of Sciences. And as now seems clear, conservatives do no better on the economics of mercury regulation than there they do on the science.

Stop and think about it for a second. Mercury is a toxic substance, emitted from power plants (among other sources), which contaminates fish and poses the strongest dangers to pregnant mothers and their unborn children. Clearly, there are going to be non-negligible economic benefits to be reaped from protecting children from the kinds of neurological and developmental problems that can result from being exposed to mercury in utero. In deciding how to regulate mercury, you probably wouldn't want to downplay such considerations, much less leave them out of your regulatory calculus.

Yet the Bush administration has rigged economic analyses -- or simply not performed them at all -- in order to obscure the clear benefits of getting a toxic metal out of the environment and the food supply without further delay.

Last week, the Government Accountability Office (GAO) released a report taking a hard look at the cost-benefit studies used by the EPA to choose between two competing mercury regulation options -- the tougher and more traditional "maximum achievable control technology" approach (which would require pollution reductions at every offending plant) and the administration's lax, market-based "cap and trade" plan (which dovetails with President Bush's Clear Skies legislation, and which early news reports suggest is the administration's choice). What the GAO found is stunning: The EPA cooked the books to favor cap and trade.

First, in its economic analysis, the agency yoked the mercury cap-and-trade plan to the recently enacted Clean Air Interstate Rule, then counted the benefits of both together. Not surprisingly, that made cap and trade look better. The GAO also found that the EPA failed to explain in a transparent fashion how it assessed the economic merits of cap and trade to begin with.

And maybe most stunningly, the GAO added that the EPA had failed to "quantify the human health benefits of decreased exposure to mercury, such as reduced incidence of developmental delays, learning disabilities, and neurological disorders." In short, perhaps the most obvious source of benefit from regulatory action -- and an area where the tougher technology-based approach clearly bests cap and trade -- got short shrift. And despite this, the technology-based approach still had a net economic benefit of $ 13 billion annually!

But wait, there's more. In a February report, the EPA inspector general highlighted some of the same problems as the GAO, but also noted that the EPA had failed to select the best technology-based approach to begin with. In effect, the EPA cherry-picked a technology-based standard that would result in mercury reductions comparable to the administration's Clear Skies plan, rather than a standard that would actually achieve the maximum possible pollution reductions (as the Clean Air Act requires). To do so, the agency had to attempt multiple model runs just to find a technology-based scenario that was bad enough that it would match the cap-and-trade approach. In short, once again the EPA rigged its analyses to make the Bush administration's politically favored approach seem like it could match the competition.

How could all this happen at one of the finest regulatory agencies in the world? First, media reports suggest that in the case of mercury, EPA political appointees told their staff "not to undertake the normal scientific and economic studies" typically required for analyzing and justifying an agency action. Moreover, the White House branch typically charged with scrutinizing agency proposals to make sure they're based on rigorous economic analyses -- John Graham's office within the OMB --- was curiously "asleep at the cost-benefit switch," as legal scholars Lisa Heinzerling and Rena Steinzor of the Center for Progressive Regulation explain in a lengthy article on the corruption of economic analysis in the mercury-regulatory process.

Remember the big picture here: Conservatives are the ones who have long yelped about the need to impose an economic check on federal-agency regulations. But along came a proposed agency action with massive benefits, and the hardheaded economists at the OMB simply shrugged while a politicized EPA rigged its analyses. Whatever regulatory option the administration ultimately chooses for mercury next week, we now know that it will have emerged from a truly dubious process.

Chris Mooney is a Prospect senior correspondent whose TAP Online column appears each week. His book on the politicization of science will be published later this year by Basic Books. His daily blog and other writings can be found at www.chriscmooney.com.