EPA’s Flawed Science: From Pretense of Knowledge to Fatal Conceit
“The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design. To the naive mind that can conceive of order only as the product of deliberate arrangement, it may seem absurd that in complex conditions … adaptation to the unknown can be achieved more effectively by decentralizing decisions and that a division of authority will actually extend the possibility of overall order. Yet that decentralization actually leads to more information being taken into account.”
- F. A. Hayek, The Fatal Conceit: The Errors of Socialism (1988), p. 76
Whenever the U.S. Environmental Protection Agency (EPA) is confronted with evidence that its proposed regulations will kill jobs, risk blackouts, or otherwise harm economic growth, it typically seeks refuge in its own estimates of the amazing public health benefits the proposal will have.
By 2020, EPA rules “will prevent 230,000 early deaths,” one recent Administration report claims. EPA Administrator Lisa Jackson has gone so far as to testify before Congress that the new regulations would provide health benefits as valuable as a cure for cancer.
Such unequivocal declarations scare the public and can intimidate the skeptic. If EPA claims about saving lives were true, the case for its aggressive regulatory agenda would be compelling. How can society worry about higher electric rates or losing American jobs to foreign shores, after all, when thousands of human lives are at stake?
But as I document in my recent paper, “EPA’s Pretense of Science: Regulating Phantom Risks,” peeling back the layers of assumptions behind EPA’s conclusions reveal that the agency’s claims are misleading at best and deceptive at worst.
At the heart of EPA claims about health benefits is a methodological change made in 2009 to the way it calculates the health risks from exposure to fine particulate matter. At a stroke, this change more than tripled the number of deaths attributable to ambient particulates. Since then, EPA has used this vast reserve of claimed deaths to justify virtually every one of its major new regulations, regardless of whether the regulations were targeted at particulates or not.
Fine Particulate Matter, or PM2.5, is a catch-all category for natural and microscopic particles released from man-made activities, especially combustion. Wind, erosion, combustion and cooking are common sources of fine particles in the air. PM2.5 is also one of the six “criteria” pollutants listed in the original Clean Air Act, for which EPA must establish national ambient air quality standards (NAAQS) protective of public health with an adequate margin of safety.
The EPA has not always been obsessed with PM2.5 as if a lethal inhalant. Toxicological data on PM2.5 indicate that current ambient levels are too low to cause significant disease or death. Instead of weighing this evidence from toxicological studies of the direct effects of fine particulates, EPA has long based its health effect estimates on epidemiological studies that compare death rates for cities with different ambient particulate levels.
EPA current risk assessment for PM 2.5 favors two cherry-picked studies by authors with connections to EPA and disregards the many conflicting studies. On the basis of intricately manipulated statistical analysis and models, EPA concludes that there is a strong correlation between “early death” and ambient PM 2.5. And without subjecting these statistical associations to a causal framework, EPA assigns 100 percent probability – that is complete certainty- to the conclusion that PM2.5 causes premature mortality.
Going to Zero
On the basis of such implausible assumptions, EPA is now attributing mortal risks from exposure to pollutants below natural background levels approaching zero! In 2009, EPA began to extrapolate this correlation between PM2.5 and slightly elevated death rates beyond the statistical evidence in the relevant studies.
Simply put, EPA decided that there is no safe level of ambient PM 2.5 – however near to zero — at which risk of ‘early” death ceases. Statisticians call this analytic approach a “no threshold linear regression to zero analytic model.” Statistical extrapolation involves the assumption that a quantitative relationship persists beyond the evidence on which it was based. Through the no-threshold assumption, EPA now finds that mortal risks increase in proportion that a location’s ambient level of PM 2.5 exceeds natural background levels.
This methodology is not “data-free,” it also contradicts a foundational principle of toxicology, that it’s the dose that makes the poison. Almost any substance can harm you if taken in a large enough quantity, and in small enough quantities almost any substance is harmless. A tiny bit of radiation is actually good for you, whereas drinking a ton of water in afternoon will kill you.
Lending credibility to EPA’s new calculations leads to absurd results. If EPA’s calculations are to be believed, 13 percent of all deaths in America are due to particulate matter, and particulate matter is responsible for between 16 and 22 percent of all deaths in much of the eastern United States.
Oddly, EPA is capitalizing on the dramatically expanded risks from PM2.5 to justify almost all of the unprecedented air quality regulations collectively known as the train-wreck. These rules, however, do not directly target particulate matter. In justifying the slew of train-wreck rules, EPA is relying not on the benefits from reductions of the emissions directly regulated, but rather on “co-benefits” from coincidental reduction in fine particulates.
In January, for example, EPA implemented new standards for mercury and other hazardous emissions from power plants. With annual compliance costs of more than $10 billion, the rule is by EPA’s own admission the most expensive ever promulgated under the Clean Air Act. Yet only $6 million, or 0.004 percent, of the rule’s claimed health benefits come from reductions in mercury, while over 99.9 percent derive from the so-called co-benefits from reducing PM 2.5.
Over the last three years, EPA has released thirteen cost-benefit analyses for new regulations. In eleven of those cases, co-benefits from particulate matter were responsible for the majority of the claimed health benefits, and in six cases 100 percent of the claimed benefits came from reductions in particulate matter, not from the ostensible target of the regulation.
Even if EPA’s benefits calculations for PM2.5 were accurate, such heavy reliance on co-benefits would be troubling. The CAA obligates the EPA to determine a national standard for particulate matter fully protective of public health. If reducing particulate matter had the enormous benefits that EPA’s analysis claims, it has a responsibility to lower the national ambient standard to a level that is actually protective of human health.
Although the EPA is reviewing the current national standard for PM2.5, the Agency has in no way suggested lowering the current PM2.5 standard of 15 micrograms per cubic meter to concentrations as low as 1 or 2 micrograms. Yet, EPA is calculating risk of early death from levels that low in cost-benefit analyses.
EPA’s flawed scientific methodology is drawing increasing criticism from Congress and the National Academy of Science. Most citizens, however, would have no reason to or expertise to question EPA’s chilling declarations about early deaths. Environmental regulatory standards reflect a judgment about what is unacceptable societal risk.
What the current EPA contends are decisions driven by pure science are, obviously, policy decisions not legitimate scientific findings. If left unchallenged, the high priests of rare science at the EPA would fatally undermine our democracy. Elected officials, not scientists or federal employees, must make this country’s major policy decisions if our form of government is to survive.