The Black Box
EPA defies Congress on “secret science” used to justify environmental regulations (Green Watch, November 2013 – PDF here)
By James M. Taylor
Summary: Environmental Protection Agency rules greatly affect the quality of life in the United States—not only in terms of air pollution, but in terms of economic growth and job creation. Yet EPA’s decisions are rooted in information that is hidden away from both Congress and the American people.
“Sunlight is said to be the best of disinfectants; electric light the most efficient policeman.” – Louis Brandeis
The House Committee on Science, Space and Technology has issued its first subpoena in 21 years. The target: “secret science” used by the Environmental Protection Agency.
In the face of repeated requests over the past two years, the EPA has refused to provide Congress with the secret information that the agency employs to justify draconian restrictions—federal regulations that are shutting down power plants, raising fuel prices, and creating a tremendous burden on the country’s economy.
Amazingly, “clean air” rules that critically affect every American are being formulated by unelected, unaccountable bureaucrats, based on data (purported data, anyway) that not even the people’s elected representatives are allowed to see.
“By denying the Committee’s request, the agency prevents Congress from fulfilling its oversight responsibilities and denies the American people the ability to verify EPA’s claims,” said Science Committee Chairman Lamar Smith (R-Texas). “The EPA’s lack of cooperation contributes to the suspicion that the data sets do not support the agency’s actions. The American people deserve all of the facts and have a right to know whether the EPA is using good science.”
Said Smith: “Over the past two years, the committee has repeatedly requested the data the agency uses to justify virtually every Clean Air Act regulation proposed and finalized by the Obama administration. . . . In September 2011, the EPA’s then-Assistant Administrator Gina McCarthy committed to make the data sets available to the Committee. Even though Ms. McCarthy now leads the agency, she has yet to provide the promised data to the Committee.
“This subpoena could have been avoided. Unfortunately, we’ve been put in this position by an agency that willfully disregards congressional requests and makes its rules using undisclosed data. After two years of failing to respond, it’s clear that the EPA is not going to give the American people what they deserve—the truth about regulations.”
EPA’s refusal to provide the requested information, even in the face of a Congressional subpoena, reinforces concerns that bureaucrats are cooking the books to justify a growing budget, more employees, and more power over the U.S. economy.
EPA studies deeply flawed
The subpoena was issued August 1. On August 19, EPA sent the House committee files containing information the committee already possessed, but refused to provide the secret data supporting two key studies that the agency frequently cites to justify ever-tightening emissions restrictions.
For more than a decade, EPA has cited two studies as the backbone justification for its ever-tightening restrictions on power plants, automobiles, and other sources of emissions into the air. The two studies are known as the Harvard Six Cities study and the American Cancer Society study (particularly Cancer Prevention Study II, or CPS-II). The Six Cities study, funded by EPA and published in 1993, claimed modestly higher concentrations of fine particulate matter in the air is correlated with higher death rates. The 1995 American Cancer Society reported similar results.
(Fine particulates are those with a diameter no greater than 2.5 microns, which is 1/400,000th of a meter or 1/30th the width of a human hair. The term “PM 2.5” refers to particles of 2.5 microns in size or, in EPA use, to particles of that size or smaller.)
EPA seized upon the two studies, the Harvard Six Cities study and the ACS study, to justify more stringent emissions restrictions during the Clinton, Bush, and Obama administrations. Scientists, however, have documented serious flaws in the data and methodology of the studies that cast severe doubt on the studies’ conclusions.
The Six Cities study, for example, produced no correlation between fine particulate matter levels and death rates among people with more than a high school education. Correlations could be produced only within the subgroup of people with no higher education, which is a red flag that the study did not properly filter out external factors.
The Six Cities study also did not find any correlation between fine particulate matter and mortality in the western United States. Again, the ability to produce correlations only in certain parts of the country called into question the study’s effectiveness in filtering out external factors.
In an eye-opening surprise which raises additional questions regarding the study’s methodology, the Six Cities study found correlations between higher concentrations of fine particulate matter and important human health benefits. That’s right: benefits.
As scientist Joel Schwartz, a former staffer for the Coalition for Clean Air, reported in formal comments to EPA, the Six Cities study “found no association between PM 2.5 and mortality for people with more than a high school education. The [Six Cities] study also reported a statistically significant decrease in mortality due to respiratory causes in areas with higher PM 2.5 levels” (emphasis in the original).
Jon Heuss, principal scientist with Air Improvement Resources, reported similar flaws. According to Heuss, “the increased mortality was experienced in the portion of the cohort that had a high school education or less, and that there was a significant spatial heterogeneity in the association, with no effect seen in western U. S. cities. All these additional findings tend to raise questions concerning the interpretation of the PM 2.5 associations as a universally applicable chronic PM health effect.”
The American Cancer Society study has even greater problems. As Schwartz noted in a 2006 policy report published by the John Locke Foundation, the American Cancer Society study claimed higher particulate matter concentrations killed men but not women, killed people with no more than a high school degree but not higher educated people, and killed people who are moderately active but not people who are very active or sedentary. In short, there was no correlation between higher concentrations of fine particulate matter and mortality for a majority of the groups studied. Schwartz concluded that the American Cancer Society’s reported correlation between fine particulate matter and mortality among certain population subgroups was likely due to other factors affecting those subgroups.
“Reanalysis of the ACS data has also shown that considering additional factors in the statistical analysis of the data can make the apparent PM 2.5 effect disappear,” Schwartz explained. “For example, when migration rates into and out of cities was added to the statistical model relating PM 2.5 and premature death, the apparent effect of PM 2.5 declined by two-thirds and became statistically insignificant.”
“These results are biologically implausible and suggest problems with the researchers’ statistical model, rather than a real cause-effect relationship,” Schwartz explained.
Other scientists confirmed the flaws reported in the two studies.
“Two re-analyses of these studies have been published since, each of which included authors from the original studies,” reported Jerome Arnett, a pulmonologist and editorial board member of the Journal of American Physicians and Surgeons. “These confirmed the weak associations found in the original studies.”
“Later, in a 2005 comprehensive review of the numerous studies published since 1996, prominent epidemiologist Suresh Moolgavkar, a professor of epidemiology at the University of Washington, found the case for EPA’s PM 2.5-mortality hypothesis was even weaker than it had been in 1996. He concluded the standard was not defensible,” Arnett explained.
Additional studies contradict EPA
Beyond the Six Cities and American Cancer Society studies, scientific evidence from other sources indicates EPA’s ever-tougher restrictions on fine particulate matter do not reduce human mortality.
Writing in the peer-reviewed journal Inhalation Toxicology, UCLA research professor James Enstrom reported on a study of nearly 50,000 elderly people exposed to varying concentrations of fine particulate matter between 1973 and 2002.
Enstrom observed, “The fully adjusted RRs [relative risk of death] were not elevated above 1.0 for any subgroup during 1983–2002. Taken as a whole, these results suggest there was a weak relationship between fine particulate pollution and mortality during 1973–1982, but none during 1983–2002.”
(The reader should keep in mind that correlation is not causation. The use of umbrellas is highly correlated with rain, but umbrellas don’t cause rain, and the rooster’s crow doesn’t cause the sun to come up. Defenders of EPA argue that, because of the difficulty of proving causation, it is reasonable to base government regulations on strong correlations that do not constitute proof of ‘A’ causing ‘B.’ In this case, however, the purported correlations do not hold up logically.)
Congress seeks raw data before acting
Given the mounting evidence that EPA’s restrictions are contradicted by sound science, the position of Rep. Smith and the House Science Committee seems reasonable. The members of Congress are giving EPA an opportunity to demonstrate that the Six Cities and American Cancer Society studies do indeed show a correlation between fine particulate matter concentrations and human mortality. EPA, however, refuses to disclose the raw data.
From the partial data presented in the publication of the two studies, scientists have discovered the flaws discussed earlier in this article. Scientists suspect the raw data will reveal more glaring flaws if EPA releases that information. If the people at EPA truly believe the raw data support their regulations, they should welcome a review by non-EPA scientists.
If the data were to support EPA’s regulations, that would create political support for such regulations if they were truly justified. But the need for transparency goes beyond that, to the heart of the Scientific Method, which dictates that scientific theories be subjected to thorough examination. Before a scientist or government researcher asserts that scientific evidence supports a particular claim, that scientist or researcher should rigorously test that claim. When putting forth a theory, a scientist or researcher should encourage others to challenge that theory. But rigorous testing of scientific assertion requires access to the related scientific data.
By refusing to release the raw data to Congress and third-party scientists, EPA proposes that we toss the Scientific Method aside and instead trust that EPA will never make any scientific mistakes or propose scientifically unjustified restrictions. This requires the American people to take the federal agency at its word that it would not engage in politically motivated conduct or seek to enhance its own funding, staffing, and power by claiming higher degrees of environmental threats than actually exist.
EPA regulations punishing the economy
EPA’s scientifically suspect emissions restrictions are taking a tremendous toll on the American economy, and the costs threaten to rise even higher.
As economist Margo Thorning pointed out in a 2011 article in the Washington, D.C. newspaper The Hill, EPA’s own economic modeling concludes that EPA’s regulations under the Clean Air Act Amendments have “significant negative impacts on U.S. GDP growth over the 2010-2020 period. GDP declines by $79 billion in 2010 and by $110 billion in 2020 relative to the baseline forecast.”
Those costs are only going higher.
Evaluating six new emissions regulations proposed by EPA in 2010 and 2011, a National Association of Manufacturers study reported that “the cumulative impact of the EPA’s proposed regulations could cost, by conservative estimates, roughly $100 billion annually and more than 2 million jobs. In a worst-case scenario, the regulations could mean the loss of $630 billion, 4.2 percent of GDP and more than 9 million jobs.”
Congress battles for data
Congress is growing increasingly frustrated with EPA’s refusal to provide the scientific data the agency claims support its restrictions.
The House Science Committee reported on its website: “As one example, by its own estimates the EPA’s proposed limits on ozone will cost taxpayers $90 billion per year, making it the most costly regulation the federal government has ever issued. Some of the data in question is up to 30-years-old.” Rep. Smith, the committee’s chairman, said, “The American people deserve all of the facts and have a right to know whether the EPA is using good science.”
EPA counters that providing the data would jeopardize personal medical histories of people participating in the two studies. House Science Committee members point out, however, that the studies finished gathering data more than two decades ago. Moreover, the names of individual subjects and sensitive information can be redacted prior to releasing the raw data.
“The Committee recognized the privacy issues potentially implicated by the documents, and accordingly granted you the option to reply with de-identified data,” Smith wrote to EPA Administrator Gina McCarthy on September 3.
Smith’s letter indicates a potentially more troubling issue for EPA: the agency may not have even examined the data prior to using that information to justify imposing its expensive restrictions on the American economy.
“Your staff asserts that ‘much of the data’ subpoenaed by the Committee ‘are held solely by the outside research institutions that conducted these large-scale epidemiological studies,’” Smith wrote. He noted:
First, that is a remarkable assertion by an agency that purports to rely on such data—data that it apparently has not even obtained, much less reviewed—to impose extraordinary costs on the American people. Second and in any event, the subpoena is not limited to documents in your possession. It re quires production of all documents in your possession, custody, or control. “Control” is the legal right, authority, or ability to obtain documents upon demand. Thus, responsive documents that EPA has ‘the legal right, authority or ability to obtain from outside research institutions are within the scope of the subpoena. Under OMB Circular A-110, EPA has the right to “obtain, reproduce, publish, or otherwise use the data” from these studies.
Smith’s letter indicates EPA never attempted to validate the assertion that the Six Cities study and the American Cancer Society study show a scientifically sound connection between fine particulate matter levels and human mortality. Given that scientists have identified major flaws in the studies even from the little information the authors made available, EPA’s utilization of the two studies as backbone support for its economy-strangling emissions restrictions defies common sense.
EPA claims dubious economic benefits
EPA’s refusal to disclose the raw data enables the agency to claim its restrictions pass a cost-benefit analysis. So long as the raw data remain unavailable to Congress and third-party scientists, Americans must take EPA at its word that its emissions restrictions substantially reduce human mortality. Likewise, EPA tells us we should take the agency at its word regarding how much economic benefit results from its restrictions.
In 2012, Gina McCarthy attended and gave a presentation at the Energy, Utility and Environment Conference, which is an annual event for government bureaucrats, environmental activists and (mostly renewable) energy industry representatives held in Phoenix. At the time, McCarthy served as Assistant Administrator overseeing EPA’s air pollution policies. In her presentation, McCarthy asserted that EPA regulations, regardless of their costs, provide tremendous net economic benefits to the American economy.
McCarthy claimed, for example, EPA’s proposed Cross States Air Pollution Rule would provide between $120 billion and $280 billion in benefits each year. She also claimed EPA’s proposed Mercury and Air Toxics Standards (MATS) would provide between $37 billion and $90 billion in benefits each year. McCarthy said that, for every dollar energy producers spend to comply with MATS, society will gain between three and nine dollars in health benefits.
McCarthy defiantly challenged the audience, “If you are worried about whether this will benefit the economy, read those figures.”
To claim such economic benefits, however, McCarthy had to employ highly creative statistical and economic logic.
As noted, McCarthy and EPA must rely on the secret data and dubious interpretations of the Six Cities and American Cancer Society studies to come up with substantial asserted health benefits from the regulations.
Just as importantly, EPA then had to apply an even more dubious formula for assigning economic value to each life it claims to have extended through its regulations.
When weighing the costs and benefits of various regulations, each federal agency provides its own estimate of how much societal value is provided by each life extended. Much of EPA’s asserted regulatory benefits accrue through the agency’s calculation that each life it extends through regulations provides $9.1 million in societal value. EPA’s $9.1 million figure is higher than that asserted by any other federal agency.
For comparison, the Food and Drug Administration (FDA) asserts a life extended is worth $7.9 million, and the Department of Transportation (DOT) assumes a life extended is worth $6 million. During the latter years of the George W. Bush administration, EPA itself asserted each life extended was worth (merely) $6.8 million.
So how does EPA arrive at the asserted $9.1 million in benefit per life extended, and why does EPA claim a greater value for each life it extends than does any other federal agency? During a press conference at the EUEC event, McCarthy said she “can’t answer” those questions.
Instead of providing a formula or specific considerations EPA took into account when arriving at its $9.1 million valuation, McCarthy said EPA “works in close consultation with the White House” to arrive at its figure.
According to McCarthy, EPA works in close consultation with Barack Obama and his political advisors to claim an exceptionally high dollar value for each life it claims to extend through its regulations. This, of course, assures that virtually any EPA regulation can be claimed to provide more benefit than cost.
A follow-up question asked whether McCarthy’s assertion that she “can’t answer” the questions with any specificity meant EPA does not have a specific formula to determine the value of a life extended.
“I didn’t say that,” McCarthy responded. “I said I can’t answer the question.”
As Margo Thorning pointed out in The Hill, EPA employs very little specificity and a great deal of conjecture when formulating its asserted economic value of a life extended through regulation.
Thorning reported that EPA’s asserted economic value of a life extended through regulation is “based on survey data that asks individuals what they would be ‘willing to pay’ (called ‘stated’ WTP) for a small increase in life expectancy and the wage differential between occupations of different riskiness, such as a commercial fishermen compared to an office worker (‘revealed’ WTP). The academic surveys of WTP used by EPA have no link to overall economic activity and do not address how (or if) WTP affects the components of GDP (consumption, investment, government spending, and net exports). ‘Willingness to Pay’ responses by survey participants or the wage differential between occupations with different levels of risk do not create any new jobs, cause any investment, or increase levels of spending in the U.S. economy.”
Emergency medical physician John Dale Dunn, a policy advisor for the American Council on Science and Health, agrees. Dunn points out that regardless of whether EPA inflates the subjective value it assigns a human life, it is misleading for the agency to claim its regulations are providing economic benefits.
“The costs that energy producers pay to comply with EPA regulations are certainly economic, and these economic costs certainly get passed down to consumers in the form of higher prices,” said Dunn. “EPA’s asserted benefits, however, are not economic. The asserted benefits are largely the subjective value that EPA places on a person allegedly living a little bit longer due to EPA regulations.
“When EPA bureaucrats say their regulations will ‘help the economy,’ they are being misleading at best,” Dunn explained. “Even if, for the sake of argument, EPA is not inflating the value it places on a life extended, its asserted benefits are noneconomic in nature. The economy pays dearly for EPA regulations, even if EPA says the noneconomic benefits are worth the price.
“You can say, ‘I think the subjective value people place on having their life extended is $9.1 million,’ but that certainly doesn’t mean the economy grows by $9.1 million each time a life is extended,” Dunn observed. “EPA is comparing economic apples to noneconomic oranges, and the agency is being misleading at best when it claims its regulations benefit the economy.”
At the Energy, Utility and Environment Conference press conference, McCarthy was asked how much time EPA emissions restrictions add, on average, to each life EPA allegedly extends.
McCarthy said she could not answer that question, either.
Dunn said the reason McCarthy could not, or would not, answer the question is that EPA can claim only modest health benefits even through its secret data and suspect formulas. The asserted mortality benefits are so modest that they equate to merely hours, or a few days at best, per person.
“There is a good reason why [then-]Assistant Administrator McCarthy said she could not answer the question: the answer is embarrassing for EPA,” said Dunn. “EPA is not preventing an army of twentysomethings from dying grossly premature deaths. Instead, even if we accept EPA’s highly suspect data-mining, EPA regulations add mere hours or days per life extended.”
“To say that adding a few hours or days to an elderly person’s life is worth taxpayers spending $9.1 million is quite a bold statement,” said Dunn. “And to say that adding a few hours or days to an elderly person’s life provides $9.1 million in economic value is simply preposterous.”
“It is also curious that EPA claims a higher monetary value per life saved than the Food and Drug Administration and the Department of Transportation,” observed Dunn. “DOT regulations are designed to prevent sudden tragedies that strike people of all ages. DOT is saving many more years of life per accident avoided than the few days of life extended by EPA regulations, yet EPA has the audacity to claim a higher benefit value than DOT.”
Dunn pointed out that the average person does not make $9.1 million in an entire lifetime of employment, which adds even more doubt to EPA’s asserted $9.1 million in economic benefits per life extended.
“EPA is cooking the books when it claims the benefits of its new and proposed regulation outweigh their extremely high costs,” Dunn summarized.
An agency deserving trust?
Against this backdrop, EPA and Congress have reached an impasse. Congress demands access to the data that EPA claims justify expensive restrictions on air emissions and on the American economy. EPA refuses to provide the data.
EPA is now in a difficult position. It must make the case that it deserves the American people’s trust—that the agency can be trusted to evaluate data and impose environmental restrictions without any accountability to Congress or substantive examination by third-party scientists who can validate or cast doubt on EPA’s scientific claims.
Congress, on the other hand, will likely face an EPA stonewall unless and until the American people demand transparency. EPA and other federal agencies under the Obama administration have perfected the art of refusing to cooperate with congressional watchdogs, of employing delaying tactics until Congress tires of pressing the issue and the news cycle turns to the next controversy.
In the meantime, EPA creates and enforces costly emissions restrictions, proposes still more costly future restrictions, and refuses to provide any hard scientific or economic data to support the restrictions. Congress and the public deserve better.
James M. Taylor is managing editor of Environment & Climate News, a columnist for Forbes, and a senior fellow at the Heartland Institute.