EPA’s War on Science

Posted on May 6, 2020 by Robert B. McKinstry, Jr.

Since its creation under President Nixon five decades ago, EPA has, for the most part, been an independent agency utilizing the best science available, even where the science led it to policy results contrary to the predilections of the party in power – that is until the Trump Administration.  Two recent actions by the Trump EPA, one final and the other proposed, exemplify the sad and stark departure by the Agency from this prior practice.

An example of the agency’s prior practice is the Advance Notice of Proposed Rulemaking issued following the Bush Administration’s defeat in Massachusetts v. EPA.  In that case, the agency staff drafted a lengthy and well-reasoned analysis indicating how greenhouse gases might be regulated under the Clean Air Act governed by the law and science.  The Bush Administration published that analysis, prefacing it with a number of letters by appointed officials presenting alternative views consistent with the Administration position rejected by the Supreme Court - - but the Agency’s well-reasoned analysis constituting the bulk of the notice was nevertheless published.

As noted, two recent actions by the Trump EPA represent a departure from this science-driven regulatory approach; they also share the distinction of being roundly condemned by EPA’s Science Advisory Board and the relevant scientific community.  Both have the transparent objective of preventing the adoption of regulations whose health benefits clearly outweigh their costs.  The two actions are: (1) the final rule reversing the necessary and appropriate finding underlying the Mercury and Air Toxics Rule (“MATS”), National Emission Standards for Hazardous Air Pollutants: Coal- and Oil-Fired Electric Utility Steam Generating Units—Reconsideration of Supplemental Finding and Residual Risk and Technology Review, Docket No. EPA-HQ-OAR-2018-0794 (“Revised Necessary and Appropriate Finding”); and (2) the proposal cynically entitled Strengthening Transparency in Regulatory Science, 83 Fed. Reg. 18768 (Apr. 30, 2018); Supplemental Notice, 85 Fed. Reg. 15396 (Apr. 17, 2020), Supplemental Notice, 85 Fed. Reg.21340 (Apr. 17, 2020) (extending comment period to May 18, 2020), proposing 40 C.F.R. pt. 30 (“Transparency Proposal”). 

Over the unanimous objection of the regulated industry and environmental groups, the Revised Necessary and Appropriate Finding reversed the Obama EPA’s finding that it was necessary and appropriate to regulate hazardous air pollutant emissions from the utility industry under section 112 of the Clean Air Act.  The finding that was reversed was made by the Obama EMA on remand from the Supreme Court’s decision in Michigan v. EPA, holding that EPA needed to consider cost in making a necessary and appropriate finding.  On remand, the Obama Administration EPA considered cost in several ways and renewed its finding that it was necessary and appropriate to regulate hazardous air emissions from the utility industry. 

After the change of administrations, the Trump EPA decided to revisit the necessary and appropriate finding and, in its Revised Necessary and Appropriate Finding, found that it was not necessary and appropriate to regulated hazardous air pollutant emissions from the utility industry (this was the fourth flip over four administrations on this issue).  The Revised Necessary and Appropriate Finding ostensibly left the substantive requirements of the MATS rule in place.  Indeed, the utility industry had already complied with those substantive requirements by either closing plants or installing control equipment. However, under the Clean Air Act, a necessary and appropriate finding is a prerequisite to regulating hazardous air pollutants from the utility industry.  EPA’s reversal of the finding therefore has the potential to increase its chance of success in the on-going challenges to the MATS rule.  Reversal of the rule could undermine the ability of regulated utilities to recover sunk capital costs.

In issuing the Revised Finding, EPA decided not to consider “co-benefits.” The vast majority of the monetized benefits arising from regulating air toxics from the utility industry arise from the fact that most of the toxic acid gases and fine particulates are a mixture of listed hazardous air pollutants and conventional pollutants.  It is, therefore, impossible in epidemiological surveys in most cases to segregate the impacts of the components that are listed hazardous pollutants from the impacts of the components that are conventional pollutants.  Moreover, the same pollution control equipment that removes the hazardous air pollutants will also remove the conventional air pollutants.  Thus, for example, hazardous hydrochloric acid, hydrofluoric acid and hydrocyanic acid all form acid aerosols having the same impact on the lungs as the nitric, nitrous, sulfuric and sulfurous acid aerosols formed from “conventional” NOx and SOx air pollutants; and the same treatment technologies will remove hazardous and “conventional” acid gases.  For that reason, the direct health benefits of controlling these pollutants are labeled as co-benefits, and, according to economists, can also be considered negative costs. 

In the Revised Finding, EPA, contrary to the recommendations of the Science Advisory Board, would exclude these benefits/negative costs, as well as non-monetized benefits, from its consideration.  Instead, EPA would consider only costs to benefits relating to control of a hazardous air pollutant whose impacts can be segregated from other pollutants that are not listed as hazardous and can be monetized.   

Because EPA did not reverse the requirements of the MATS rule, the only apparent reason for proceeding with the revised finding appears to be an intent to advance a rule for the consideration of costs that will make it more difficult to regulate pollutants in the future.  Most pollutants have a variety of impacts, are emitted into the atmosphere as a mixture of pollutants, mix with other pollutants in the environment, and have impacts on receptors that cannot be segregated.  For example, most of us have observed the blue skies and clear air resulting from the reduction in automobile, truck, and air traffic as a result of COVID-19 restrictions.  Under the rationale underlying the Revised Necessary and Appropriate Finding, proposals to reduce greenhouse gas emissions from these sources could not consider the many health, environmental and welfare benefits arising from reductions in NOx and fine particulates.  While perhaps that is the underlying intent, the rationale could also be extended to water pollution, hazardous and solid waste, and other regulations in the future.

The Transparency Proposal might seem to be a proposal that would promote sound science and good government procedure; it provides:

…when EPA develops regulations, including regulations for which the public is likely to bear the cost of compliance, with regard to those scientific studies that are pivotal to the action being taken, EPA should ensure that the data underlying those are publicly available in a manner sufficient for independent validation.

83 Fed. Reg. 18768.  In fact, the proposed regulation is an all too transparent attempt to preclude reliance on results that are crucial for the promulgation of regulations to protect health and the environment, even where those results have met the rigorous requirements of scientific peer review.

Most notably the proposal applies specifically to “dose response data and models” supporting a regulation, requiring that they be “publicly available in a manner sufficient for independent validation. . . in a fashion that is consistent with law, protects privacy, confidentiality, confidential business information, and is sensitive to national and homeland security.  40 C.F.R. § 30.5, proposed 83 Fed. Reg; 18773.  In fact, this qualification makes most data and studies critical to support regulations out of reach for agency reliance.  The underlying data in human health studies is invariably private information that, by the words of the proposal, would be unavailable, for example, many dose response animal studies are business confidential.  Additionally, most models upon which EPA relies are proprietary and are available only for a significant price.  The proposal would also seemingly preclude reliance upon metadata and review articles appearing in peer reviewed publications, since the proposal would require that the underlying data be available. 

In recognition of these fatal flaws, the proposal has been criticized by EPA’s Science Advisory Board and major scientific organizations.  Moreover, there has not been a showing of any need for the proposed regulation.  The only reasonable conclusion is that this proposed regulation, like the Revised Necessary and Appropriate Finding, is an effort to promote the Trump Administration’s anti-regulatory agenda contrary to the dictates of sound science, in short, a war against science.  There is still an opportunity to comment.  The comment period has been extended to May 18, 2020.