Section 112 of the Clean Air Act and MATS
Section 112 of the Clean Air Act mandates that EPA set emissions control standards for a number of hazardous air pollutants listed in the CAA itself. Section 112(c) directs EPA to list source categories that emit significant levels of the listed hazardous air pollutants. Section 112(d) mandates that EPA set pollution standards for source categories on the 112(c) list. EPA also has the authority to remove source categories from the list under certain circumstances. EPA has included power plants on the 112(c) list. A separate provision unique to power plants, Section 112(n)(1)(A), provides that EPA shall issue hazardous air pollutant standards for power plants if it determines that it is “appropriate and necessary” to regulate power plant hazardous air pollutants. EPA issued MATS under section 112(d) after making the appropriate and necessary determination under section 112(n)(1)(A).
After control technology requirements are implemented for a source category, Section 112(f) requires that EPA perform a risk and technology review. If the agency determines that remaining emissions pose a residual cancer risk of more than 1-in-1-million and that additional reductions in hazardous air pollution can be achieved, EPA must issue a second, more stringent set of pollution control requirements.
New Jersey v. EPA, 517 F.3d 574 (D.C. Cir. 2008)
Section 112(c) directs EPA to list source categories that include facilities emitting a certain amount of hazardous air pollutants, and power plants clearly fit the mold. EPA placed power plants on the 112(c) list via its 2000 finding that it was appropriate and necessary to regulate power plant hazardous air pollutants. In 2005, the Bush EPA found that it was not appropriate and necessary to regulate hazardous air pollutants from power plants under section 112(d) and also removed power plants from the section 112(c) list of source categories subject to section 112 requirements. In 2008, the D.C. Circuit held that once power plants were included on the section 112(c) list EPA could remove them from that list only by making the delisting showings required by 112(c)(9), regardless of EPA’s finding at the time that it was not appropriate and necessary to regulate power plant hazardous air pollutants.
MATS Background
Signed by EPA Administrator Lisa Jackson in December 2011, the Mercury and Air Toxics Standards required coal- and oil-fired power plants to reduce their emissions of mercury and a range of heavy metals and acid gases. As part of the rule, EPA had reinstated the appropriate and necessary finding issued in 2000 and rescinded in 2005. When it was issued, EPA projected that MATS would cut power plant mercury emissions by 90 percent, acid gases by 88 percent, and SO2 emissions by 41 percent. The agency also projected that those pollution cuts would avoid thousands of premature deaths and prevent 100,000 heart and asthma attacks annually. While compliance with the rule would cost a little under $10 billion annually, the human health value alone of the air quality improvements was projected to be $37 billion to $90 billion each year. By the spring of 2015, the majority of coal- and oil-fired power plants had met their MATS obligations, and the rest followed by the spring of 2016.
The Supreme Court: “Appropriate and Necessary” Includes “Cost”
In June 2015, while the Supreme Court rejected numerous challenges to MATS, it ruled (Michigan v. EPA, 576 U.S. ___) that EPA had erred by not taking account of cost in making its 2011 appropriate and necessary determination. The Court stated that it took no position on how EPA was to account for cost. In remanding the case to the D.C. Circuit Court of Appeals for further proceedings, the Court included no instructions on whether that court should vacate the rule altogether. The D.C. Circuit declined to vacate MATS, leaving the pollution control standards and compliance schedule in effect, and sent the appropriate and necessary finding back to EPA for review consistent with the Supreme Court’s opinion on the need to account for cost.
The Supplemental Appropriate and Necessary Finding
In April 2016 EPA issued a Supplemental Finding that after taking account of the cost of compliance with the MATS rule, it found no basis for altering the determination that regulating hazardous air pollutants emitted by power plants was appropriate and necessary. The finding relied on two alternative analyses. The first assessed the $9.8B cost of compliance with MATS as reasonable, noting that the utility industry’s annual revenue ranged from $277B to $356B in the years since 2000; the cost of compliance thus represented 2.7% to 3.5% of the industry’s annual revenues. EPA concluded:
“…the cost of complying with MATS—compared to historical annual revenues, annual capital expenditures, and impacts on retail electricity prices—is well within the range of historical variability. The EPA further finds that the power sector is able to comply with the rule’s requirements while maintaining its ability to perform its primary and unique function—the generation, transmission, and distribution of reliable electricity at reasonable cost to consumers.”
EPA asserted that the Michigan ruling did not require a formal benefit-cost analysis, but nevertheless applied a benefit-cost analysis to provide an alternative basis for the appropriate and necessary finding. Since compliance with MATS had the effect of reducing not just mercury, acid gases, and other hazardous air pollutants but also pollutants like SO2 and ambient fine particles, EPA included the full health benefits of reducing the entire suite of pollutants in its benefit calculations. The result: regulating hazardous air pollutants emitted by power plants would yield net benefits of between $27B and $80B and thus was “appropriate and necessary.”
Murray Energy v. EPA (USCA Case #16-1127)
The Utility Air Regulatory Group, represented by the law firm Hunton and Williams where now-Assistant Administrator for Air and Radiation Bill Wehrum was a partner at the time, and Murray Energy, a long-time lobbying client of Acting Administrator Andrew Wheeler, challenged the supplemental finding in the D.C. Circuit (captioned Murray Energy v. EPA). EPA filed its brief defending the supplemental finding on January 19, 2017. After the challenger’s reply brief was been filed, the Trump EPA asked the court to hold the case in abeyance pending its review of MATS the supplemental finding.
Previewing the Potential Issues in the Pending Proposal
The proposal comes more than three years after the power sector came into compliance with MATS, thus already having incurred a substantial share of the overall costs of complying with MATS and, in many cases, recovering those costs from their customers via regulatory ratecases. During this time the public has been reaping the air quality and health benefits of the sweeping pollution reductions achieved through MATS compliance.
Option 1: Rescinding the Appropriate and Necessary Finding
Under this approach, the MATS standards would remain in place. The option might also include a determination that under the risk and technology review provisions of section 112(f), no further pollution control requirements are warranted. Together, these two outcomes might be most satisfactory to the utility industry, which has already incurred the bulk of its costs in complying with MATS and which would prefer not to be subject to a second round of requirements under section 112(f).
Rescinding the appropriate and necessary determination complicates both leaving MATS in place and asserting the authority to make a prompt determination that no further controls are warranted.
Advance Notice on Costs and Benefits. EPA may already have revealed its motivation for targeting the appropriate and necessary finding even though that action would leave the agency with a complicated path to the two outcomes proposed in this option. In early June 2018, EPA issued an Advance Notice of Proposed Rulemaking, “Increasing Consistency and Transparency in Considering Costs and Benefits in the Rulemaking Process”. The Advance Notice of Proposed Rulemaking specifically cited the MATS supplemental finding as potentially objectionable, at least to some, because it was an example of past actions where “the Agency has justified the stringency of a standard based on the estimated benefits from reductions in pollutants not directly regulated by the action (i.e., ‘ancillary benefits’ or ‘co-benefits’).” Again, in the supplemental finding’s benefit-cost analysis, EPA used the high-value public health and monetary benefits that would flow from the reduction in fine particles resulting from compliance with the MATS even though the targeted pollutant was mercury.
Challengers’ Briefs in Murray Energy. The challengers’ briefs in Murray Energy v. EPA also likely preview the current EPA’s approach in a proposal to revoke the supplemental appropriate and necessary finding. Among the arguments made in the briefs:
- Weighing costs against benefits is the only basis on which EPA can make an appropriate and necessary finding.
- In evaluating the costs of compliance, EPA must consider “all costs and disadvantages, including the impacts on coal companies, communities, and workers, as well as localized impacts”.
- Since section 112 addresses only a specified list of hazardous air pollutants, the benefits flowing from the reductions of other pollutants must be excluded from the benefit-cost analysis when determining whether regulation is appropriate and necessary.
Section 112 requires that EPA set emissions control standards for hazardous air pollutants based on the emissions reductions achieved by the highest-performing sources. The industry briefs also suggested that since section 112(n) required EPA to study “alternative” methods of emissions control, EPA should have included “alternative” – presumably less stringent – emissions compliance options the MATS rule.
Comment – The Appropriate and Necessary Finding: While it is unclear that the proposal will include all of these arguments, it is likely that it will focus on the argument that only the benefits the flow from reducing mercury and other power plant pollutants listed in section 112 can be included in calculating the benefits for purposes of making the appropriate and necessary determination. The proposal may point to the language of section 112(n)(1)(A) to support its restrictive approach to calculating benefits:
“The Administrator shall perform a study of the hazards to public health reasonably anticipated to occur as a result of emissions by electric utility steam generating units of pollutants listed under subsection (b) after imposition of the requirements of this Act…”
The Administrator shall regulate electric utility steam generating units under this section, if the Administrator finds such regulation is appropriate and necessary after considering the results of the study required by this subparagraph.
EPA may argue that it is offering the only, or at least the best, interpretation of this language in concluding that the scope of its benefits analysis must or should be narrow. The agency may even venture to argue that it is simply following the Court’s Michigan opinion. The Michigan majority, however, made it clear that it was answering only the narrow question of whether EPA was required to account for cost in making the appropriate and necessary finding. The opinion explicitly stated that the Court offered no view on how EPA took account of cost in making the funding.
EPA’s Legal Task. Meanwhile, the proposal will have to overcome legal arguments and analysis that were previewed in the Obama EPA’s Murray Energy brief defending the supplemental appropriate and necessary finding and supporting the agency’s accounting in its benefits-cost analysis for the full range of pollution and health benefits that would be reduced as a result of the regulation.
Were the proposal to follow the approach implied in the June Advance Notice and laid out in the industry briefs in the Murray Energy case, the agency would have to do two things. First, it would have to argue that even though the Michigan Court described “appropriate and necessary” as “capacious”, the agency did not have the discretion as a matter of law to take account of all of the pollution reduced by MATS in calculating benefits and was constrained to count only the “targeted” pollutant benefits.
Second, EPA would have to argue in the alternative that even if the agency had discretion, it was reasonable to ignore the reductions in SO2 emissions and ambient fine particles – and the health benefits of those reductions – resulting from MATS even when considering the full range of costs. In fact, the initial and supplemental appropriate and necessary determinations are based on a formidable record establishing the dramatic public health and economic benefits of reducing SO2 and ambient fine particles.
OMB Guidance. It might also have to explain away well-established guidelines from the Office of Information and Regulatory Affairs (OIRA) to federal agencies issued by the George W. Bush administration in 2003, which state: “Your analysis should look beyond the direct benefits and direct costs of your rule-making and consider any important ancillary benefits and countervailing risks” (OMB Circular A-4, Section E, subsection headed “Ancillary Benefits and Countervailing Risks,” September 17, 2003).
New Science. Since MATS was issued in 2012, the scientific case presenting, for example, the danger of fine particle concentrations even at low levels has grown only more compelling. The case for the regulation of mercury, one of the “targeted” MATS pollutants, has also grown stronger, with recent science showing that in 2011 EPA underestimated the benefits of reducing mercury.
The proposal thus will have to contend with the threshold question: does it account for the even stronger case for regulation made by post-2011 science, or does it try to justify ignoring information that was not in the record at the time of the issuance of MATS? If the proposal accounts for the broader range of costs identified in the Murray Energy industry brief, it will have to justify doing so without taking account of the full benefits of the regulation.
Actual Compliance Costs. With the last of the MATS compliance deadlines having passed almost two and a half years ago, the utility industry is in full compliance with the MATS requirements. That means that utilities have already invested capital to install pollution control equipment and in some states PUCs have incorporated a portion of those investments in the rate base. (In other states, such cases may still be pending). Industry reported this summer that it had spent a cumulative multi-year total of $18B to comply with the rule, which appears to be less than EPA projected when MATS was issued. Will the agency take into account these lower costs in a benefit-cost-analysis?
Comment – Leaving MATS in Place and Issuing a Risk and Technology Review Determination: Thanks to section 112(n)(1)(A), cited above, rescinding the appropriate and necessary finding removes the foundation for MATS and, indirectly, for any subsequent risk and technology review determination, even a negative determination. EPA will be compelled, therefore, to include a legal theory for its authority to offer this option. The proposal could argue that even in the absence of an appropriate and necessary finding, EPA’s earlier determination to include power plants on the list of source categories covered under section 112 remained in force. The proposal likely would point to the D.C. Circuit decision in New Jersey v. EPA holding that rescinding the appropriate and necessary determination did not by itself result in removing power plants from the list of covered source categories. EPA likely would then go on to argue that so long as power plants remained on the list of covered source categories, the agency had the authority to maintain MATS. In fact, the proposal could invoke the section 112(d)(1) mandate that EPA set standards for hazardous air pollutant emissions for any and all listed source categories and argue that this mandate prevails even in the absence of an appropriate and necessary finding. Then, with MATS remaining in place, the proposal could include a risk and technology review determination.
Option 2: Rescinding the Appropriate and Necessary Finding and the MATS Requirements
In contrast to option 1, EPA would argue that that section 112(n)(1)(A) is the controlling provision even if it is in deep tension with the mandate of section 112(d)(1). Again, 112(n)(1)(A) provides:
The Administrator shall regulate electric utility steam generating units under this section, if the Administrator finds such regulation is appropriate and necessary after considering the results of the study required by this subparagraph.
If the proposal includes this option, it may be for two motivations. First, EPA may simply find it hard to interpret this language so as to justify maintaining the standards once the agency has withdrawn the appropriate and necessary finding, and even notwithstanding the section 112(d)(1) mandate.
Second, section 112(f) directs the EPA to set additional pollution control standards – in this case for power plants subject to MATS – if it determines after a risk and technology review, that controls are necessary to ensure environmental protection. Section 112(f) also directs the EPA to set additional, more stringent standards if the more stringent controls are needed to achieve reductions in pollutants classified as known, probable or possible human carcinogens so as to reduce lifetime excess cancer risks to exposed individual to less than one in one million. Since sources are subject to the additional standards required under section 112(f) only if standards for toxic air pollutants have already been in place for at least 8 years, rescinding MATS could remove the grounds for any additional regulation of toxic air emissions from power plants. Including this option might be a “tell” that, in contrast to option 1, EPA would find it challenging to determine that no further reductions in hazardous air pollutant emissions are warranted. In that case, simply removing the grounds for further reductions would be the safer course in the EPA’s view.
Option 3: Removing Power Plants from the Categories Listed as Subject to Section 112/De-Listing
The proposal may include an additional option, proposing a trifecta of rescinding the appropriate and necessary finding, withdrawing the MATS pollution control requirements, and removing power plants from the list of sources subject to section 112 – or proposing to conclude explicitly that the first two steps are the functional equivalent of de-listing power plants.
Option 3 may represent an attempt by EPA to achieve complete consistency in its approach. If EPA is proposing to conclude that it is not appropriate and necessary to regulate toxic air emissions from power plants then, the reasoning may go, power plants should be removed from the list of source categories covered by section 112. Completing the trifecta could advance the Trump administration’s relentless promotion of coal by removing once and for all power plants from coverage under section 112 or, at least, to creating another procedural obstacle to a successor administration’s efforts to reinstate limits on hazardous air pollution from power plants. Like option 2, this option would remove the grounds for the application of section 112(f), which mandates the imposition of additional, more stringent controls in certain circumstances. Finally, the proposed trifecta may be intended to preclude action under section 112(j), which provides that in the event that EPA has not promulgated a standard for a source category, those sources are required to apply for a permit that sets emissions limitations for hazardous air pollutants determined on a case-by-case basis to be equivalent to the standard that would exist were a standard to have been otherwise established.
New Jersey Impact. The delisting part of the trifecta is likely to run into difficulty, however. Section 112(c) directs EPA to regulate source categories that include facilities emitting a certain amount of hazardous air pollutants and EPA placed power plants on the 112(c) list via a notice of finding in 2000. In 2005, the Bush EPA found that it was not appropriate and necessary to regulate hazardous air pollutants from power plants and also removed power plants from the list of source categories subject to section 112 requirements. In 2008, the D.C. Circuit held that once power plants were included on the section 112(c) list the EPA could remove them from that list only by making the delisting showings required by 112(c)(9), regardless of EPA’s finding at the time that it was not appropriate and necessary to regulation power plant hazardous air pollutants.
Section 112(c)(9), meanwhile, allows EPA to delist a category of sources only if the agency finds that no source in the category emits carcinogens or, if the category includes such sources, they do so at levels that do not:
“… exceed a level which is adequate to protect public health with an ample margin of safety and no adverse environmental effect will result from emissions from any source (or from a group of sources in the case of area sources).”
The inventory of power plant emissions of hazardous air pollutants likely makes it extremely difficult to make this showing. As the New Jersey court ruled, a determination that regulating hazardous air pollutants from power plants is not appropriate and necessary simply does not serve as an analysis that meets these criteria. The Supreme Court’s Michigan decision offers no help either. After the D.C. Circuit upheld the MATS rule, including its appropriate and necessary determination, the Court took up only the issue of whether cost must be included in making an appropriate and necessary determination. Neither the Court nor, on remand, the D.C. Circuit vacated MATS, and neither disturbed the listing of power plants as 112(c) sources.
How will the proposal justify the trifecta, then? EPA may be tempted to pursue two arguments. First, the proposal may argue that it is nothing more than a continuation of EPA’s response to the Supreme Court’s Michigan opinion. Michigan required EPA to include cost in its appropriate and necessary determination. Since the agency is proposing to determine that it is not appropriate and necessary to regulate hazardous air pollutants emitted by power plants, the proposal must also address the legal consequences of that determination. One of the consequences, the proposal may go on to argue, is the introduction of tension between the requirements of section 112(d) – mandating the regulation of emissions from source categories listed under 112(c) – and the negative appropriate and necessary determination, which eliminates the foundation for power plant regulation. The proposal would then argue that it is in EPA’s discretion to resolve the tension by removing power plants from the 112(c) list. To the extent that sequence flows logically from EPA’s response to Michigan, the proposal would assert, New Jersey is no longer binding.
This line of argument may be easy to refute simply because nothing in the Michigan opinion affected the holding in New Jersey. Michigan addressed only one question: whether the EPA’s appropriate and necessary analysis required a consideration of cost. The Court declined to take up any other issue, including the EPA’s decision to list power plants pursuant to section 112(c). In effect, in doing so, the Court upheld the listing of power plants pursuant to section 112(c) foreclosing, rather than inviting, an EPA “response” to the listing question. Even if for the sake of argument this assertion is granted and even if the tension between section 112(d) and 112(n)(1)(A) is granted, the proposal would still be in defiance of the New Jersey ruling that EPA must meet the requirements of 112(c)(9) before it can remove a source category from the list.